Professional Documents
Culture Documents
D96091GC10 SG PDF
D96091GC10 SG PDF
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
Zero Data Loss Recovery
Appliance Overview
Student Guide
D96091GC10
Edition 1.0 | November 2016 | D98586
Alessandro Colonna This document contains proprietary information and is protected by copyright and
other intellectual property laws. You may copy and print this document solely for your
own use in an Oracle training course. The document may not be modified or altered
Technical Contributors in any way. Except where your use constitutes "fair use" under copyright law, you
and Reviewers may not use, share, download, upload, copy, print, display, perform, reproduce,
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
publish, license, post, transmit, or distribute this document in whole or in part without
Jeff Ferreira the express authorization of Oracle.
Kelly Smith
The information contained in this document is subject to change without notice. If you
Tim Chien find any problems in the document, please report them in writing to: Oracle University,
500 Oracle Parkway, Redwood Shores, California 94065 USA. This document is not
Marco Calmasini warranted to be error-free.
h a s
Publishers
Trademark Notice
m ) Oracle iand/or
d e ฺ
Pavithran Adka Oracle and Java are registered
ฺ c oowners. Gu its affiliates. Other names
trademarks of
Syed Ali
n k ng ent
may be trademarks of their respective
Raghunath M
n b a
S t ud
Jobi Varghese
u nio this
l l o @ use
a k be e to
l o (a cens
b el li
ak eem
Contents
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
1 Introduction
About This Course 1-2
Course Overview 1-3
Additional Resources 1-4
How Prepared Are You? 1-5
Introductions 1-6
a ble
fe r
2 Zero Data Loss Recovery Appliance Solution
ans
Objectives 2-2
n - t r
o
Agenda 2-3
s an
What Are Engineered Systems? 2-4
) ha eฺ
ฺ c om Guid
Engineered Systems: Oracle Zero Data Loss Recovery Appliance X6-2 2-5
n k ng ent
Oracle Zero Data Loss Recovery Appliance: Overview 2-6
ud
Engineered System: Simple to Install, Manage, and Maintain 2-9
n b a t
Agenda 2-10
nio this S
u
Oracle Recovery Appliance Cloud-Scale Solution 2-11
@ use
Agenda 2-12 l l o
a k be e to
Oracle Recovery Appliance Hardware Specifications 2-13
l o (a cens
Oracle Recovery Appliance X6-2 Hardware Details 2-14
bel li
Oracle Recovery Appliance X6-2: Compute Server 2-17
em
ake
Oracle Recovery Appliance X6-2: Storage Server 2-18
Oracle Recovery Appliance X6-2: InfiniBand Switches 2-19
Oracle Recovery Appliance X6-2: Management Switch 2-20
Agenda 2-21
Oracle Recovery Appliance Base Software 2-22
Oracle Integrated Lights Out Manager (Oracle ILOM) 2-24
Agenda 2-25
Implementation Workflow 2-26
Summary 2-28
iii
Inadequate Data Recovery = Big Business Impacts 3-6
Key Database Protection Goals 3-7
Agenda 3-8
Backup Appliances Are Not Designed for Database 3-9
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
l l o @ use
Delta Store: Efficient Indexing and Storage Management 3-34
k be e to
Delta Store: Virtual Full Backup Optimization 3-35
a
l o (a cens
Delta Store: Virtual Full Backup Purging 3-36
bel li
Delta Store: How It Works 3-38
em Delta Store: Receiving Backup Pieces 3-40
ake Delta Store: Container Files and Delta Pools 3-41
Delta Store: Delta Pool Plans 3-43
Internal Tasks 3-44
Recovery Appliance Metadata Database: CATALOG 3-46
RMAN Recovery Catalog and Virtual Private Catalogs 3-48
Data Encryption 3-50
Protected Databases 3-51
Recovery Appliance Storage 3-52
Recovery Appliance Storage Space Management 3-54
Recovery Appliance: Unique Benefits for Business and IT 3-56
End-to-End Data Protection Visibility 3-57
Fast Recovery with Oracle Recovery Appliance 3-58
Recovery Appliance: Unique Benefits for Business and IT 3-60
Policy-Based Cloud Scale Protection 3-61
Dynamic Policy-Based Space Allocation 3-63
iv
Modern Cloud-Scale Database Protection 3-64
Recovery Appliance Replication 3-65
Media Manager Integration 3-67
Enterprise Manager Cloud Control 3-68
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Reporting 3-70
Summary 3-72
l l o @ use
OEDA Configuration: Replication Network Screen 4-23
k be e to
OEDA Configuration: Compute Node OS Screen 4-24
a
l o (a cens
OEDA Configuration: Review and Edit Screen 4-25
bel li
OEDA Configuration: Cluster Info Screen 4-26
em OEDA Configuration: Cluster Details Screen 4-27
ake OEDA Configuration: Ingest Cluster Networking Screen 4-29
OEDA Configuration: Replication SCAN Screen 4-30
OEDA Configuration: Cell Alerting Screen 4-31
OEDA Configuration: Platinum Support Screen 4-32
OEDA Configuration: Auto Service Request Screen 4-33
OEDA Configuration: Configuration Manager Screen 4-35
OEDA Configuration: Tape Library Screen 4-36
OEDA Configuration: Comments Screen 4-38
OEDA Configuration: Generate Files Screen 4-39
OEDA Configuration: Finish Screen 4-40
OEDA Configuration: Installation Template 4-42
OEDA Installation: Steps Overview 4-43
OEDA Installation: Get the Current Software - Step 1 4-44
OEDA Installation: Pre-Install Check - Step 2 4-46
OEDA Installation: Installation Script - Step 3 4-48
v
OEDA Installation: RA Installation Script - Step 4 4-50
OEDA Installation: EM Discovery - Step 5 4-52
Network Test Throughput 4-53
Agenda 4-54
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
vi
Enrolling Protected Databases 6-21
Configuring Protected Databases 6-26
Agenda 6-29
Backing Up Protected Databases 6-30
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
vii
Recovery Appliance Replication Status 8-18
Summary 8-20
Objectives 9-2
Agenda 9-3
Tape Archival 9-4
Recovery Appliance Tape Configuration 9-6
Recovery Appliance Alternative Tape Configurations 9-8
Agenda 9-9
Integrated Media Manager Components 9-10
Integrated Media Manager Concepts 9-12
a ble
fe r
Integrated Media Manager Interfaces 9-13
an s
Media Manager Implementation in the Recovery Appliance 9-14
n - t r
o
an
Oracle Clusterware Media Manager Services in the Recovery Appliance 9-16
s
Oracle Clusterware Media Manager Directories in the Recovery Appliance 9-17
ha eฺ
)
Copy-to-Tape Backup Operations in the Recovery Appliance 9-18
ฺ c om Guid
Integrated Media Manager Concepts in the Recovery Appliance 9-19
k ng ent
Integrated Media Manager in the Recovery Appliance 9-21
n
Integrated Media Manager 9-22 n b a
S t ud
Summary 9-28 u nio this
l l o @ use
10 Maintenance
a k bande Patching
e to
(a 10-2
Objectives
l o c e ns
b el
Agenda li
10-3
em Recovery Appliance Patch Overview 10-4
ake Recovery Appliance Patch Planification 10-6
Recovery Appliance Patching 10-8
Agenda 10-18
Recovery Appliance Oracle Enterprise Manager Plug-In 10-19
Agenda 10-21
Recovery Appliance Powering Off, Powering On 10-22
Agenda 10-25
Replacing a Damaged Physical Disk 10-26
Safe Disk Removal 10-28
Replacing a Damaged Flash Card 10-29
Moving All Disks from One Cell to Another 10-30
Using the Cell Software Rescue Procedure 10-32
Summary 10-35
viii
11 Diagnostics and Troubleshooting
Objectives 11-2
Agenda 11-3
Recovery Appliance Main Areas for Issues 11-4
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
ix
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
ake
me
bel
l o
a k
li
l l o
(a cens
u
be e to
n
@ use
b a n
nio this
k
S t
ฺ c
ud
ng ent
)
s
om Guid
ha eฺ
an
o n - t r an
sfe r a b
le
1
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Introduction r a ble
s fe
Oracle Zero Data Loss Recovery Appliance - t r an
o n
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
About This Course
( a ak nse
e l lo l i ce
m b
e
ake
a k
l o (a cens
bel li
em
ake
recovery-appliance/index.html
• My Oracle Support
– http://support.oracle.com
• Oracle Technology Network Data Sheet
– http://www.oracle.com/technetwork/database/availability/recove
le
ry-appliance-ds-2297776.pdf rab fe
• Oracle Documentation Library n s
– http://docs.oracle.com/cd/E55822_01/index.htm n - tra
n o
h a sa
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved. 4
a a
( cen
l l o li
m be
e
ake
a a
( cen
l l o li
m be
e
ake
following information:
• Name
• Company affiliation
• Title, function, and job responsibility
• Experience related to topics presented in this course
r a ble
• Reasons for enrolling for this course n s fe
• Expectations from this course n - tra
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved. 6
a k
l o (a cens
bel li
em
ake
a k
l o (a cens
bel li
em
ake
Objectives
a k
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
Systems provide:
• Optimal performance and high availability
• Less deployment and operating risk
• End-to-end single-vendor support
• End-to-end unified monitoring of all
r a ble
components sfe n
n - tra
ENGINEERED TO an
o
s
WORK TOGETHERm) ha eฺ
ฺ c o Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Building
b eand li a custom platform requires top talent. Traditional custom systems that
l operating
k e emused to run Oracle Database backup solutions do not deliver the performance or reliability
are
a that an Engineered System can deliver. The components that are used in custom database
backup systems are often not balanced and are often misconfigured, thus creating
bottlenecks and reducing overall system performance. Complex inter-component failure
modes are not tested and components are not engineered together as a unit in custom
database backup systems.
The hardware components, database software and libraries, operating system and device
drivers, firmware, network configuration, and all other components in the Engineered System
are optimized to work together. Oracle Engineered Systems are built to provide the highest
performance and are the most widely available platform for running Oracle software
components.
Engineered Systems include a full range of specialized services to simplify deployment,
maintenance, and support. Engineered Systems are best suited for companies that want the
highest system performance, rapid deployment, reliability, and security, with the lowest total
cost of ownership.
with Oracle Database and eliminates data loss exposure for all databases
without impacting production environments.
• Integrated real-time Oracle redo transport eliminates data loss exposure.
• Incremental forever backups eliminate backup windows and database server
overhead.
• Continuous database-aware validation ensures data integrity and database
recovery to any point in time.
• Cloud-scale capacity with online expansion a ble
protects thousands of databases, enabling fe r
database protection as a service. an s
n - t r
• Autonomous tape archival enhances
tape resource utilization and investment a no
protection. h a s
• Unified control and automation simplifies m ) ideฺ
end-to-end management of the entire g ฺ co t Gu
n kn den
data protection lifecycle – from data creation
a
to disk to tape.
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Oracle’s l l o li Recovery Appliance (Recovery Appliance) is a ground-breaking data
eZero Data Loss
b
k e em solution that tightly integrates with the Oracle Database in order to address these
protection
a requirements head-on. It eliminates data loss exposure and dramatically reduces data
protection overhead on production servers. In addition, the Recovery Appliance scales to
protect thousands of databases, ensures end-to-end data validation, and implements full
lifecycle protection including disk backup, tape backup, and remote replication.
Delta Push
• DBs access and send only changes Tape Archival
• Minimal impact on production • Offloads Tape Backup
• Real-time redo transport instantly • Tapes utilized all day
protects ongoing transactions
a ble
fe r
an s
n - t r
Protects all DBs in your Data Center
Delta Store
a no
• Stores validated, compressed DB changes on disk
• Petabytes of data
• Oracle 10.2-12c, any platform
h a s
• Fast restores to any point-in-time using deltas
( a ak nse
Oracle’s e l
Zero
lo Data Loss
l i ceRecovery Appliance tightly integrates with the new Recovery
m b
k e e
Appliance specific capabilities in the Oracle Database and the Recovery Manager (RMAN)
a backup tool to provide data protection capabilities and performance that are not possible with
any other data protection solution.
The principal design goal for the Recovery Appliance is to eliminate the loss of critical
database data that is still possible by using existing data protection solutions.
The second design goal for the Recovery Appliance is to reduce backup-related processing
on production database systems to the absolute minimum, transmitting only the changed
data. With unnecessary backup processing eliminated, production systems can now focus on
their primary goal of serving business-critical workloads.
The third design goal for the Recovery Appliance is to provide a cloud-scale database
protection service for tens to thousands of databases in a data center.
this paradigm by supporting the reception of real-time redo blocks from all the hundreds
to thousands of databases across an enterprise, thereby dramatically reducing data loss
exposure from hours/days to sub-seconds.
• Zero-Impact Backups: With the Recovery Appliance, customers simply run the RMAN
in an incremental-forever strategy – that’s it. It means no more recurring full backups, no
more tape backups, and no more RMAN backup deletion and validation commands on
the production server. All expensive backup workloads are offloaded to the appliance,
where dedicated hardware and storage handle these tasks in the most efficient manner. a ble
fe r
• Restore to Any Point-in-Time: Each incremental block received is validated,
ans
- t r
compressed, and then indexed as a part of a new virtual full backup, which appears as a
n
no
normal full backup in the RMAN catalog. Essentially, this means validated full backups
a
h a s
are now available at any point in time, without ever incurring the cost of running full
backup operations.
m ) ideฺ
ฺ
• Cloud-Scale Protection: It has the ability togmassively co t G u at a cloud level by
scale
supporting from tens to thousands of databases
a n kn dacross
e n a data center, essentially
forming a private Data Protectionn
o b within
Cloud
S tuthe enterprise. There are several
technology components within
u i s
i Appliance that makes this possible.
n the Recovery
h
e t
• End-to-End Data Protection
e l lo@ o uVisibility:
s The Recovery Appliance provides a complete,
end-to-end view
a a kbinto thes e t protection lifecycle by using Oracle Enterprise Manager
data
l o ( –cfrom
Cloud Control n the time the backup is created by RMAN on the database to the
edisk,
b e
time l
it is l
stored i on on tape, and/or replicated to another Recovery Appliance in a
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
Optimization
Expertise
Build Your Own
Network
Administration
SAVINGS
Storage
Administration
Time
Cost
System
a b le
Administration fe r
an s
Enroll Protected
n - t r
Databases
a no
Disaster Recovery
h a s
Implementation
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Building l l o li system for an Oracle Database backup system requires a diverse
eyour own custom
b
k e emset and expertise that cover many hardware and software components, often which are
skill
a from different vendors. The initial deployment of a custom build-your-own system might
encompass tasks such as sizing, ordering, researching best practices, assembling, installing,
patching, and configuring the software. If you take into consideration all the activities that are
needed to put together a new system, it can take weeks to perform.
Because Oracle Recovery Appliance is an engineered system that has been purpose-built, it
greatly reduces the time and cost that are associated with the integration of the hardware and
software components that have been optimized for running the Oracle Database Backup
solution. All the hardware and software components are provided by Oracle, thereby enabling
customers to benefit from streamlined single-vendor support. With Oracle Recovery Appliance
X6-2, a single DBA can deploy a high-availability clustered database backup system in about
one hour.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
n i o i
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
e l
OraclebRecovery li
Appliance easily protects all databases in the data center by using a
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
r a
U19 b
Interconnect 2 InfiniBand QSFP+ 2 InfiniBand QSFP+ 2 InfiniBand QSFP+
U18
Storage Server (17)
s fe U18
an
U17 Compute Server U17
U16
U15
n - t r Compute Server
Storage Server (7)
U16
U15
no
U14 U14
FC SAN adapter 2 x 16 Gb QLogic 2 x 16 Gb QLogic 2 x 16 Gb QLogic U13 U13
sa
Storage Server (6)
(optional) FC SFP+ PCIe FC SFP+ PCIe FC SFP+ PCIe U12 U12
h a
) Servers ฺ
U11
U10
Storage Server (5)
U11
U10
Storage Servers 14 Storage Servers 18 Storage Servers 18 Storage
m i d e U09
Storage Server (4)
U09
co t Gu
U08 U08
(max conf) (168 Data Drives) (216 Data Drives) (216 Data Drives)
g ฺ U07
Storage Server (3)
U07
nb18) s Stu
U04 U04
Expansion 18) 18) U03 U03
n i o i
U02
Storage Server (1)
U02
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
Zero Datab l li Appliance is a backup solution that has been specifically designed
e Loss Recovery
m Oracle Database files. The Recovery Appliance can also recover any database to
etoeprotect
ak any point in time as well as automate tape archival of data to protect against site disasters.
Virtual full backups are possible to remote backup appliances. The Recovery Appliance is a
single, standalone, rack solution that contains Compute Servers and Storage Servers, all
housed within a customized Sun Rack 1242 enclosure. Compute Servers are based on the
Oracle Server X6-2 (8-drive) server and Storage Servers are based on the Oracle Server X6-
2L (12-drive) server. Engineered Systems, like this Recovery Appliance, are designed such
that specifically chosen hardware and software components are pre-integrated in the rack to
operate as a cohesive whole, reducing the cost and complexity of IT infrastructures while
increasing productivity and performance compared to standalone systems.
Servers Included (3) X6-2L Storage Servers (18) X6-2L Storage Servers
(upgradeable to 18) (total)
a k
l o (a cens
The hardware
b li
el components of Zero Data Loss Recovery Appliance X6 consists of (2) 1U
em Server X6-2s (Compute Servers) and multiple Oracle Server X6-2Ls (Storage
eOracle
ak Servers). The Recovery Appliance includes InfiniBand and Ethernet switches, and is available
configured as a full rack or as a base rack that can be upgraded with additional Storage
Servers when required.
operations. Database processing is pushed to Storage Servers, allowing all the disks to
operate in parallel, reducing CPU consumption on Compute Servers.
• Switches: InfiniBand and Ethernet switches are used for the various internal and external
networks as well as for the management of the Recovery Appliance. These switches
include:
- Leaf: (2) 36-port InfiniBand switches used for high-speed internal communication
between internal hardware components (Compute Servers, Storage Servers, fabric
a ble
interconnects, and so on)
fe r
ans
- Spine: (1) A 36-port InfiniBand switch used for high-speed external communication
between multiple Recovery Appliances (optional) n - t r
o
an
- Management: (1) A 48-port 1 GbE switch used to provide easy management of all
s
ha eฺ
internal hardware components (Compute Servers, Storage Servers, and so on)
)
• Networks
ฺ c om Guid
n k ng ent
- Private: (2) 36-port InfiniBand switches (commonly referred to as Leaf Switches)
n b a
S t ud
used for high-speed internal communication between internal hardware components
nio this
(Compute Servers, Storage Servers, system disk, and fabric interconnects)
u
l l o @ use
- Ingest: A required 10 GbE network that connects the protected Oracle Database
a k be e to
servers to the Recovery Appliance for backup within the same data center (also
l o (a cens
referred to as the backup network). This is a high-speed, private Ethernet network
using (2) 10 Gb connections to each of the (2) Compute Servers in the rack. These
bel li
connections may be configured as active/passive or as active/active. Compute
e m Servers support channel bonding, providing higher bandwidth and availability.
ake - Replication: An optional 10 GbE network used to connect local Recovery
Appliance(s) with remote Recovery Appliance(s)
• Fibre Channel (FC) SAN: An optional 16 Gb Fibre Channel network used to connect the
Recovery Appliance to the datacenter storage area network (SAN) for backups to tape
• Rack: A single, standalone rack based on Sun Rack 1242 that houses and provides AC
power to all internal components (Compute Servers, Storage Servers, and switches)
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
a k
l o (a cens
Compute
b l
eServers li
included with the Zero Data Loss Recovery Appliance X6 are Oracle Server
k e emservers, are only available in limited configurations, and optimized for the Zero Data
X6-2
a Loss Recovery Appliance X6. They handle all the processing required by the database
application software, except for any storage- or disk-related processing (which is handled by
Storage Servers). They are also used to provide host network connectivity to the Zero Data
Loss Recovery Appliance X6.
The dual-port 16 Gb QLogic FC SFP+ PCIe adapters are not installed at the factory and
shipped for installation with the Recovery Appliance. No other cards (FC or otherwise) are
allowed in the Recovery Appliance and the firmware version of the QLogic adapters cannot
be changed.
a k
l o (a cens
Storage
b l
eServers li with the Zero Data Loss Recovery Appliance X6 are Oracle Server
included
k e em servers, are only available in limited configurations, and optimized for the Zero Data
X6-2L
a Loss Recovery Appliance X6.
Storage Servers are the building blocks of the Recovery Appliance. They free up CPU
resources in Compute Servers and are highly optimized for use with database software to
dramatically accelerate database processing and to speed I/O operations. Database
processing is pushed to Storage Servers, allowing all disks to operate in parallel, thereby
reducing CPU consumption on Compute Servers.
The Recovery Appliance uses Storage Servers as the main storage for backups. This is the
destination for protected database backups. Redo staging areas for redo streams reside on
Recovery Appliance Storage Servers disks too. To expand past all the resources provided in
a Zero Data Loss Recovery Appliance X6 (Base Rack), customers can order additional
Storage Servers when required.
a a
( cen
InfiniBand l l o i used for high-speed internal communication among components and
e switches lare
b
k e em
for inter-rack communication.
a There are (2) Leaf Switches included with each Zero Data Loss Recovery Appliance. The
Leaf Switches are Sun Datacenter InfiniBand Switch 36 and are used to interconnect the
compute servers and storage servers to provide extremely scalable, reliable and high-
performance connectivity among all internal components within the Zero Data Loss Recovery
Appliance.
Leaf Switches also connect to Spine Switches when multiple Recovery Appliances are
connected together. Client connectivity is not recommended via InfiniBand connections. For
best performance, it is highly recommended that you use 10 GbE for all client connectivity.
The (2) ports on each InfiniBand HCA within each Compute and Storage Server connect to
different Leaf Switches to provide redundant communications between Compute Servers and
Storage Servers.
The Spine Switch is not included with the Zero Data Loss Recovery Appliance X6 by default.
A spine switch is required for interconnecting the Zero Data Loss Recovery Appliance X6 with
other Recovery Appliances. The Spine Switch used for interconnecting Recovery Appliances
together is the Sun Datacenter InfiniBand Switch 36, which is used to connect up to (18)
Recovery Appliances together.
a a
( cen
There bis e l l o li
one (1) Management Switch included with each Zero Data Loss Recovery Appliance
k e emThe management switch is the Cisco Catalyst 4948E-F switch and is used to connect to
X6.
a all components within the Zero Data Loss Recovery Appliance X6 on a 1 GbE management
network.
The management switch is used for network management tools. It is connected to all
components within the Recovery Appliance. This switch can be replaced by a customer-
supplied switch.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
• Hardware drivers
• RAC Database 12c with options
• Grid Infrastructure 12c: Oracle Clusterware and ASM
• Pre-Bundled Media Manager for tape offsite
• Oracle Enterprise Manager 12c Cloud Control Agent
• Zero Data Loss Recovery Appliance Software
• QLogic FC SFP+ PCIe adapter drivers and tools (optional)
• HealthCheck (exachk)
a ble
Oracle Recovery Appliance: Storage Server fe r
an s
• Oracle Linux 6
n - t r
•
•
Hardware drivers
Oracle Exadata Storage Server Software a no
s
Oracle Recovery Appliance: Additional Components ) ha ฺ
m i d e
• PDU firmware and HTML interface
Cisco Catalyst 4948 Ethernet Switch IOSkng
ฺco t Gu
•
n d e n
• Sun Datacenter Infiniband Switch 36 a
nb s Stu
software
n i o i
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
The Zero
b l i
eData LosslRecovery Appliance X6 base image includes Oracle Enterprise Linux as
them
kee
operating system for both Compute Servers and Storage Servers. It also includes the
a required firmware and drivers. Customers cannot modify the individual components of the
image package.
Compute Servers Integrated/Pre-Installed Software include:
• Oracle Linux 6
• Hardware drivers, management pack, and firmware drivers for various components
Restricted use of Oracle Database Enterprise Edition on the Recovery Appliance and the
following options:
• Oracle Real Application Clusters
• Oracle Advanced Compression Option
• Oracle Advanced Security Option
• Oracle Partitioning
a a
( cen
e l l o
OraclebIntegrated li Out Manager (ILOM) provides advanced service processor (SP)
Lights
k e em and software that you can use to manage and monitor your Oracle hardware.
hardware
a Oracle ILOM is pre-installed on all of Oracle Recovery Appliance Compute Servers, Storage
Servers, and Sun Datacenter Infiniband Switches. It is a vital management tool in the data
center.
Oracle ILOM enables you to experience a single, consistent, and standards-based service
processor across all Oracle Recovery Appliance servers and switches. This means you will
have:
Single, consistent system management interfaces for operators
• Support for rich and standard protocols
• Oracle Enterprise Manager 12c Cloud Control management tools and interfaces
• Integrated system management functions at no extra cost
The Oracle ILOM SP runs its own embedded operating system and has a dedicated Ethernet
port, which together provide out-of-band management capability. Oracle ILOM automatically
initializes as soon as power is applied to the server. It provides a full-featured, browser-based
web interface and has an equivalent command-line interface (CLI). An industry-standard
SNMP interface and an IPMI interface are also available.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
• Separation of duties
• Prerequisites for recovery appliance administration
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
A typical l l o li
eRecovery Appliance environment includes personnel with the following roles:
m b
a kee• Cloud Control administrator: The application administrator with this role administers
Oracle Enterprise Manager Cloud Control (Cloud Control).
• Recovery Appliance administrator: This administrator manages Recovery Appliance.
• Protected Database administrator: This administrator is responsible for configuring
backups to the Recovery Appliance by using the virtual private catalog account
assigned by the Recovery Appliance administrator.
You must work with the Oracle field engineers to install and set up an Oracle Recovery
Appliance.
Use the following tools to complete administrative tasks for the Recovery Appliance:
• Cloud Control: Cloud Control is a system management tool with a graphical UI that
enables you to manage and monitor the Recovery Appliance and its protected
databases. This is the preferred UI for Recovery Appliance tasks.
• SQL*Plus: SQL*Plus is a command-line tool that enables you to run DBMS_RA
program units, and query recovery catalog views. You use SQL statements and Oracle-
supplied PL/SQL packages to complete these tasks in SQL*Plus.
a k
l o (a cens
bel li
em
ake
a k
l o (a cens
bel li
em
ake
Objectives
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
In this b l l o
e you learn
lesson, li about the role that the Recovery Appliance has in your database
m infrastructure. In addition, you will get in-depth understanding of the Recovery
kee
protection
a Appliance solution covering hardware and software components functionality and features.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
fe r
an s
48%
Total cost of
n - t r
backup solutions
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den (Percent of respondents, n=1,200 )
i o nb s Stu
Source: IDG Research Services, May 2012; ESG Group - Trends in Data Protection Modernization in 2012
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
The fundamental li
problems with today’s database protection solutions stem from:
b
ak ee•m Backup and recovery methods that are based on decades-old nightly backups
paradigm, where up to a day’s worth of data can be lost on every restore
• High backup overhead on production servers and networks for processing all database
data during backups, whether it has changed or not
• Ever-increasing backup windows due to non-stop data growth
• Backup appliances that cannot scale to protect the hundreds to thousands of databases
in the data center
• Poor visibility and control of the full data protection lifecycle, from disk to tape to
replication
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
eprovides alilist of customer’s data loss consequences and the serious impacts in
This slide l
b
em decisions and results.
ebusiness
ak
a a
( cen
Some b e l l
examples
o li the number of outage days due to inadequate data recovery
showing
k e em
implementations
a
Business Goals
• Never lose critical business data
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
IT Goals
• Ensure database-level recoverability
• Centralized service to protect all databases a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
Current Backup Solutions Fail g o GuThese Goals
toฺcAchieve
n k n ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Currentbbackup li fail to achieve business and Information Technology department
el solutions
k e em
goals.
a
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
Window
Lose all data since
last backup Large performance
impact on production
Poor Database
Recoverability
Many Systems
to Manage a ble
fe r
Many files are copied, Scale by deploying an s
but the protection state more backup
n - t r
of the database is
unknown a no appliances
h a s
m ) ideฺ
g ฺ co t Gu
Treat Databases as Just a n kn todPeriodically
Files e n
Copy
n b S t u
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
The protection
b li described in "Traditional Database Backup Techniques" are not
el strategies
em to solve the challenges created by this transformation. Enterprises find themselves
edesigned
ak without a consistent backup and recovery strategy.
The following shortcomings are common to most or all of the traditional backup techniques:
• Data loss exposure: A database is only recoverable to its last valid backup, which may
have occurred hours or days ago. In addition, storage snapshots and third-party
appliances cannot validate Oracle data blocks, and so cannot detect Oracle block-level
corruptions.
• Daily backup windows: As database sizes increase, the lengths of the backup
windows also increase, creating additional load on production systems. Critical
databases cannot afford to be deprived of resources used for daily backups and related
maintenance activities.
• Poor Database Recoverability: Because most third-party backup snapshot and
Recovery Appliances lack Oracle integrated data block and database backup validation,
restore and recovery operations tend to fail. Such failures result in extended down time
and potentially larger data loss.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
• The file system data format does not change in the backup process.
– Ideal for storage deduplication technologies
– Deduplication ratio claims of 10-50x
• RMAN passes a prepackaged backup set.
– The backup stream/block format is Oracle proprietary.
– RMAN backup stream is largely opaque to external backup
a b le
applications and storage. fe r
n s
n - tra
— Full backups (production impact) are needed for deduplication.
a k
l o (a cens
Current
b el
disk-based li protection solutions impose large loads on production systems. This
data
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
Minimal Impact
Eliminate Data Loss
Backups
Real-time redo
Production databases
transport provides
only send changes; all
instant protection of
backup and tape
ongoing transactions
processing offloaded
a ble
Database-Level Cloud-Scale Protection
fe r
Recoverability
an s
End-to-end reliability,
Easily protect all
n - t r
databases in the data
visibility, and control of
a no center by using a
databases, and not of
h a s massively scalable
disjointed files
m ) ideฺ service
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Oracle’s l l o li Recovery Appliance tightly integrates advanced data-protection
eZero Data Loss
b
k e em
technologies with the Oracle Database to address these challenges head-on. The Recovery
a Appliance redefines the database protection landscape with an innovative, never-before-seen
approach that:
• Eliminates Data Loss: As backups go, you are only protected up to your last backup –
in the event of a full restore and recovery operation, any changed blocks and archived
logs not backed up result in some amount of data loss, spanning sometimes hours and
sometimes even days. The Recovery Appliance completely changes this paradigm by
supporting the reception of real-time redo blocks from all the hundreds to thousands of
databases across the enterprise, dramatically reducing data loss exposure from
hours/days to sub-seconds.
• Has Minimal Impact Backups: With the Recovery Appliance, customers simply run the
Recovery Manager (RMAN) in an incremental forever strategy – that’s it, no more
recurring full backups, no more tape backups, and no more RMAN backup deletion and
validation commands on the production server. All expensive backup workloads are
offloaded to the appliance, where dedicated hardware and storage handle these tasks in
the most efficient manner.
Protection Cloud within the enterprise. There are several technology components within
the Recovery Appliance that makes this possible.
In addition to these unique benefits, the Recovery Appliance also provides:
• End-to-End Data Protection Visibility: The Recovery Appliance provides a complete,
end-to-end view into the data protection lifecycle by using Oracle Enterprise Manager
Cloud Control – from the time the backup is created by RMAN on the database to the time
it is stored on disk, on tape, and/or replicated to another Recovery Appliance in a remote
a ble
data center.
fe r
• ans
Secure Replication: Backups on a local Recovery Appliance can be easily and quickly
n - t r
replicated via secure transport to a remote Recovery Appliance for protection against
no
disasters such as site outages or regional disasters. The replication topology can be
a
tailored to match the data center’s requirement. h a s
• Enables Autonomous Tape Archival: Optionalo m
tape i d eฺsends backups directly
)integration
g ฺ
from the appliance to tape, offloading resource-intensive
u
c t Gfull-backup-to-tape operations
k n n
from production database servers. This
b axn7,also de efficiency and reduces costs by
increases
t u
enabling tape systems to be used
i o n24
s S of only during off-peak backup windows.
instead
• Scales Efficiently: Built on
n
ua massively i
th scalable architecture, the Zero Data Loss
Recovery Appliance o @
llenables users
e
us to expand capacity seamlessly to petabytes of
b e
k downstime. t o
storage, withano
( a n e
e
o Enterprise
lic Data: The ability to replicate change data in real-time to another
b ellData
• Secures
Zero Loss Recovery Appliance, along with end-to-end data validation, protects
m
a kee business critical data from site disasters or software/hardware malfunctions.
Tape Library
Oracle Recovery Appliance Design Goals
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Delta Push
Tape Archival
• DBs access and send only changes • Offloads Tape Backup
• Minimal impact on production • Tapes utilized all day
• Real-time redo transport instantly
protects ongoing transactions
a ble
fe r
an s
n - t r
Protects all DBs in your Data Center
Delta Store
a no
• Stores validated, compressed DB changes on
• Petabytes of data
• Oracle 10.2-12c, any platform
disk
h a s
• Fast restores to any point-in-time using deltas
• No expensive DB backup agents
m ) ideฺ
• Built on Exadata scaling and resilience
• Scales to 1000s of DBs
co t Gu
• Enterprise Manager end-to-end control
g ฺ
a n kn den Replicate to Remote
n b S t u Recovery Appliance
n i o i s
@ u e th
b e llo to us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
( a ak nse
Oracle’s e l
Zero
lo Data Loss
l i ceRecovery Appliance tightly integrates with new Recovery Appliance-
m b
k e e capabilities in the Oracle Database and the RMAN backup tool to provide data
specific
a protection capabilities and performance that are not possible with any other data protection
solution.
The principal design goal for the Recovery Appliance is to eliminate the loss of critical
database data that is still possible by using existing data protection solutions.
The second design goal for the Recovery Appliance is to reduce backup-related processing
on production database systems to the absolute minimum, and transmitting only the changed
data. With unnecessary backup processing eliminated, production systems can now focus on
their primary goal of serving business-critical workloads.
The third design goal for the Recovery Appliance is to provide a cloud-scale database
protection service for tens to thousands of databases in a data center.
a a kb se t
l o ( cen
l
e Appliance
The Recovery
b li receives incremental backups and real-time redo from the protected
em
edatabases. The appliance contains the Recovery Appliance metadata database. This
ak database includes the following components:
• The RMAN recovery catalog, which is subdivided into multiple virtual recovery catalogs
• One or more storage locations. Recovery Appliance storage contains the delta store,
which includes multiple delta pools.
Media Manager software is pre-bundled with the Recovery Appliance to archive backups to a
tape library.
Multiple protected databases send backups and real-time redo to the Recovery Appliance.
Protected databases can run on different releases of Oracle Database.
Oracle Enterprise Manager Cloud Control running on a separate server in the environment
helps administrators manage all Recovery Appliances, protected databases, and tape devices
in the Recovery Appliance environment.
Minimal Impact
Eliminate Data Loss
Backups
Real-time redo
Production databases
transport provides
only send changes; all
instant protection of
backup and tape
ongoing transactions
processing offloaded
a ble
Database Level Cloud-Scale Protection
fe r
Recoverability
an s
End-to-end reliability,
Easily protect all
n - t r
databases in the data
visibility, and control of
a no center using massively
databases, and not of
h a s scalable service
disjointed files
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Data Loss:
Eliminates li As backups go, you are only protected up to your last backup – in the
b
k e em of a full restore and recovery operation, any changed blocks and archived logs not
event
a backed up result in some amount of data loss, spanning sometimes hours and sometimes
even days. The Recovery Appliance completely changes this paradigm by supporting the
reception of real-time redo blocks from all the hundreds to thousands of databases across the
enterprise, dramatically reducing data loss exposure from hours/days to sub-seconds.
Traditional
a ble
fe r
Full Backups
an s
+
n - t r
Protected
Incrementals
a no
Databases h a s
m ) ideฺ
ฺ c o Gu Recovery
n k ng+ ent Appliance
n b S t ud
aIncrementals
Redo
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
el backupliapproaches, if the online redo log is lost, then media recovery loses all
In traditional
b
m after the most recent available archived redo log file or incremental backup. An RPO
kee
changes
a of a day or more that might result from a traditional approach may be unacceptable.
The Recovery Appliance solves the RPO problem through a continuous transfer of redo
changes to the appliance from a protected database. This operation is known as real-time
redo transport.
a ble
fe r
Hub and
an s
Spoke
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu Tape Library
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
To protect l l o li or site outage, one Recovery Appliance can replicate backups to a
e against server
b
em Recovery Appliance.
edifferent
ak When the upstream Recovery Appliance sends the incremental backup to the downstream
Recovery Appliance, it creates a virtual full backup as normal. The downstream Recovery
Appliance creates backup records in its recovery catalog. When the upstream Recovery
Appliance requests the records, the downstream Recovery Appliance propagates the records
back.
If the local Recovery Appliance cannot satisfy virtual full backup requests, then it
automatically forwards them to the downstream Recovery Appliance, which sends virtual full
backups to the protected database. DBAs use RMAN as they usually do, without needing to
understand where or how the backup sets are stored.
( a ak nse
A robuste l lo strategy
backup l i ce protects data against intentional attacks, unintentional user errors
m b
kee
(such as file deletions), and software or hardware malfunctions. Tape libraries provide
a effective protection against these possibilities.
The fundamental difference between the two approaches is that the Recovery Appliance
backs up to tape, and not to protected databases. The Recovery Appliance comes with
preinstalled Media Manager software and supports optional Fibre Channel cards. Thus,
installation of a Media Manager is not necessary on the protected database hosts.
When the Recovery Appliance executes a copy-to-tape job for a virtual full backup, it
constructs the physical backup sets, and copies them to tape, and then writes the metadata to
the recovery catalog. If desired, the Recovery Appliance can also copy successive
incremental backups and archived redo log file backups to tape. Though the backup on the
Recovery Appliance is virtual, the backup on tape is a non-virtual, full physical backup. The
Recovery Appliance automatically handles requests to restore backups from tape, with no
need for administrator intervention.
media managers.
• Tape drives and tape libraries function more efficiently because the Recovery Appliance is
a single large centralized system with complete control over them. In other tape solutions,
hundreds or thousands of databases can contend for tape resources in an uncoordinated
manner.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
fe r
an s
Remote Replica
n - t r
a no
h a s
• Data validated on
• Data validated on receive
m ) ireceive,
d e ฺ restore, and
• Data periodically revalidated
g ฺ co t Gu periodically
• Data validated on restore
a n kn den on-disk
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
A basicbprinciple li
of backup and recovery is to ensure that backups can be restored
m
kee
successfully. To ensure that there are no physical corruptions within the backed-up data
a blocks, backups require regular validation.
The Recovery Appliance provides end-to-end block validation, which occurs in the following
stages of the workflow:
• Recovery Appliance validation: The Recovery Appliance automatically validates the
backup stream during the backup ingest phase, before writing the backups to disk. No
manual RESTORE VALIDATE step is required. Also, a background task running on the
Recovery Appliance periodically validates the integrity of the virtual full backups in the
delta pools
• Oracle ASM: Oracle ASM stores the backup and redo data for the Recovery Appliance.
Oracle ASM mirrored copies provide redundancy.
disk when a HARD check fails. Furthermore, Oracle Exadata Storage Server Software
automatically inspects and repairs hard disks periodically when hard disks are idle. If the
software detects bad sectors on a hard disk, the software automatically sends a request to
Oracle ASM to repair the bad sectors by reading the data from another mirror copy. By
default, the hard disk scrub runs every two weeks.
• Tape library: The Recovery Appliance validates blocks when it copies them to tape, and
also when it restores them from tape.
• Downstream Recovery Appliance in a replication configuration: If you configure a ble
fe r
s
replication, the downstream Recovery Appliance validates data during the backup ingest
an
and restore phases.
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
Minimal Impact
Eliminate Data Loss
Backups
Real-time redo
Production databases
transport provides
only send changes; all
instant protection of
backup and tape
ongoing transactions
processing offloaded
a ble
Database Level Cloud-Scale Protection
fe r
Recoverability
an s
End-to-end reliability,
Easily protect all
n - t r
databases in the data
visibility, and control of
a no center using massively
databases, and not of
h a s scalable service
disjointed files
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Minimal l l o li
eImpact Backups: With the Recovery Appliance, customers simply run the RMAN in
b
k e em
an incremental forever strategy – that’s it, no more recurring full backups, no more tape
a backups, and no more RMAN backup deletion and validation commands on the production
server. All expensive backup workloads are offloaded to the appliance, where dedicated
hardware and storage handle these tasks in the most efficient manner.
• After a one-time full backup, incrementals are used to create virtual full
database backups on a daily basis.
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
Protected
Day 0
Full
Day 1
Incr
Day N
Incr
n - t r
Databases n o Store
Delta
h a sa
m ) ideฺ
g ฺ co t Gu
a n kn den
No More Full Backups: b
o n Incremental
S tu Forever Architecture
u n i h i s
e t
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
l
Practically,
b li
e all backup-related processing is offloaded to the Recovery Appliance. This
a a
( cen
l l o
e main goals
One ofbthe li of the Recovery Appliance is to reduce backup-related processing on
em database systems to the absolute minimum by transmitting only the changed data.
eproduction
ak With unnecessary backup processing eliminated, production systems can now focus on their
primary goal of serving business critical workloads.
The Recovery Appliance implements an incremental forever backup architecture to minimize
impact on production systems.
This architecture is based on two innovative technologies: Delta Push and Delta Store.
Full/Incremental
Backup Sets
Remote File
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Validate
Server Process
Redo Blocks
Archived Log
Backup Sets
Protected Redo Blocks Redo
Databases Staging
Area
Archived Log Backups
Delta Store
a ble
Validate
fe
Replica r
Backup Sets
Data Blocks
Index Blocks
(Create Virtual Full)
an s
Appliance
Backup Staging
Compress + Write to Disk
n - t r
RMAN
Incrementals via
Recovery
(FLASH)
a no
Appliance
Backup HTTP Servlet
h a s Incremental Backup Sets
Module
m ) ideฺ
g ฺ co t Gu Archived Log Backup Sets
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
With Deltal l o li
e Push, protected databases only send incremental backups containing unique
b
em to the Recovery Appliance. There is no need for recurring full backups. Delta Push is
echanges
ak also known as “incremental forever” because, after a one-time full backup, only incremental
backups are run on production systems. Effectively, Delta Push is a highly optimized form of
source-side deduplication. Changed blocks on production databases are very efficiently
identified by using RMAN block change tracking, which eliminates the need to read
unchanged data.
Special integration between protected databases and the Recovery Appliance eliminates
committed undo, unused, and dropped tablespace blocks from the backup stream,
significantly reducing overhead and space consumption.
Because Delta Push sends only changed data and not full backups, network traffic is greatly
reduced compared to other solutions. This enables low-cost 10 Gig Ethernet to be used for
backups. Expensive dedicated Fibre Channel or Fibre Channel over Ethernet backup
networks are not needed. Also, minimizing network traffic allows the Recovery Appliance to
be located further away from the protected databases and even, in some cases, across a
WAN in a remote data center.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• Redo data is transferred from memory buffers on the database server; very
low impact.
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
• If the redo connection is lost, the appliance "buttons up" the received redo
data and creates a “partial archived log backup”.
– Preserves recovery until the last change received by the appliance
(sub-second RPO)
• When the connection is reinstated, a gap detection process on the
appliance automatically fetches missing archived logs from protected
databases.
Protected Databases Storage
a ble
Location
fe r
an s
- t r
X m
h a
Partial
a
) LogidBackup
co t GuArchived
e ฺ
no
sArchived
n
g ฺ
a n kn den Log Backup
nb s Stu
Fetch Missing
Archived Logs
n i o i
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
Redo data
b l i
e contains lrecords of all changes made to a database and is therefore critical in
k e em
minimizing data loss if data failure occurs. By using the real-time redo transport feature of
a Recovery Appliance, you substantially reduce the window of potential data loss that exists
between successive archived redo log backups. A typical RPO is zero to sub-second when
you enable real-time redo transport.
With real-time redo transport enabled, a protected database generates redo changes in
memory, and then immediately transfers them to the Recovery Appliance, which validates
them and writes them to a staging area.
When the protected database performs an online redo log switch, the Recovery Appliance
converts and assembles the redo changes into compressed archived redo log file backups.
The Recovery Appliance catalog automatically tracks these archived redo log backups in its
recovery catalog. The RMAN can restore and apply these archived redo log backups as
usual.
Appliance RAC database. Redo log files are assembled in the Recovery Appliance in the Redo
Staging Area (also known as the Landing Pad) located in the DELTA Store ASM Disk Group.
After that, the assembled redo log files per protected database are moved as Archived Redo
Log files into the DELTA Store ASM Disk Group too.
If a protected database loses connection to the Recovery Appliance, an INCOMPLETE_ARCH
task is performed to make the partial log file available as an archivelog backup. A gap detection
code runs to transfer missing logs from the protected databases to the Recovery Appliance.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
Day 1 C1 D1 fe r
Incremental Backup
an s
n - t r
Day 2 a no
D2 s
Incremental Backup
) h a ฺ
ฺ c om Guide
Day 3
n k nB3g eC3nt
Incremental Backup
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
The delta
b l is theli totality of all protected database backup data in the Recovery Appliance
estore
em location. All datafile and archived redo log backups reside in the delta store.
estorage
ak By default, the RMAN makes full backups. A full backup of a datafile includes every allocated
block in the file being backed up. A full backup of a datafile can be an image copy, in which
case every data block is backed up. It can also be stored in a backup set, in which case
datafile blocks not in use may be skipped.
An incremental level 0 backup is a full backup that happens to be the parent of incremental
backups whose level is greater than 0. Incremental backups at level 0 can be either backup
sets or image copies, but incremental backups at level 1 can only be backup sets.
The Recovery Appliance only supports incremental backups at level 0 and 1 to realize the
benefit of the optimization features provided by Virtual Full Backups. The Recovery Appliance
still supports full or partial backups (no incremental), but such backups are treated as simple
backup files as they appear to traditional file or tape storage systems and are not treated as
Virtual Full Backups.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• A virtual full backup is created in the Recovery Appliance catalog for each
received RMAN incremental backup.
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
l l o li the incoming changed data blocks, and then compresses, indexes,
e Store validates
The Delta
b
emstores them. These changed blocks are the foundation of virtual full database backups,
eand
ak which are space-efficient pointer-based representations of physical full backups as of the
point-in-time of an incremental backup.
Oracle Databases must be running with the block change tracking option that causes Oracle
Database to track datafile blocks affected by each database update. The tracking information
is stored in a block change tracking file. When block change tracking is enabled, the RMAN
uses the record of changed blocks from the change tracking file to improve incremental
backup performance by only reading blocks known to have changed, instead of reading whole
datafiles.
Recovery
Window =
Day 0 A0 B0 C0 D0 E0 3 Days
Full Backup
Day 1 - Virtual Full Backup
Day 1 C1 D1 A0 B0 C1 D1 E0
Incremental Backup
Day 2 - Virtual Full Backup
a b le
Day 2 A2 D2 fe r
ans
Incremental Backup A2 B0 C1 D2 E0
n - tr
no
Day 3 - Virtual Full Backup
Day 3 B3 C3
h a s a B3 C3 D2 E0
Incremental Backup
m ) ideฺA2
a a kb se t
l o ( cen
Backup b e is theliautomated stage in which a Recovery Appliance scans a backup that
l
ingest
k e emsent by a protected database. The Recovery Appliance decomposes the backup into
was
a smaller sets of blocks, writes the blocks into the appropriate storage location, and indexes the
backups. Indexing includes inserting rows into the Recovery Appliance metadata database to
describe the physical location of every block.
This effective indexing stage helps to the Recovery Appliance to optimize the storage and
improve the recovery operations.
Space is released for blocks not needed for the recovery window goal. In the previous
example, on Day 4, assuming a three-day recovery window, Day 1 Virtual Full Backup is no
more needed and could be purged upon space pressure.
• The optimize task is a background process that runs on the most recent
virtual full backup to “optimize” its read I/O access for future restore
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
operations.
B0 E0 C1 A2 D2
Day 0 A0 B0 C0 D0 E0
Full Backup
B3 C3 A4 C4 D4
Day 1 C1 D1
Incremental Backup free free free free free
a b le
Day 2 fe r
Incremental Backup
A2 D2
an s
Optimization
n - t r
Day 3 A4
aC4no D4 E0
B3
Incremental Backup
B3 C3
h a s
m ) C3ideD2
A2 ฺ B0 C1
Day 4
Incremental Backup
A4 C4 D4
g ฺ co t Gu
a n kn den free free free free free
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Virtualb
full l l o
e blocks areli reordered into contiguous sets for fewer and larger read I/O requests,
k e em is more efficient than many smaller read I/O requests. This stage is for periodic
which
a optimization for future restore operations requests.
As old blocks are deleted and new incremental backups arrive for updated datafiles, the
blocks in a backup can become less contiguous. This state can degrade the performance of
restore operations. The Recovery Appliance automatically reorganizes the blocks to maintain
contiguity during ordinary maintenance and validation.
Optimization includes updating rows into the Recovery Appliance metadata database to
describe the new optimized physical location of every block.
Day 0 A0 B0 C0 D0 E0 A0 D0
B0 C0 E0
Full Backup
C1 D1 A2 D2 B3
Day 1 C1 D1
Incremental Backup
C3 A4 C4 D4 free
Day 2 A2 D2
Incremental Backup
Purging
a b le
fe r
Day 3
B0 E0 C1 A2
a n s D2
Incremental Backup
B3 C3
-t r
n onC4 D4
B3 C3 A4
Day 4 A4 C4 D4 h a sa
Incremental Backup
m ) ideฺ free free free free free
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Appliance
The Recovery li prefers to have space already available for new allocations. To do
b
em
eso:
ak • The Recovery Appliance executes as normal PURGE tasks in preexisting schedulers.
• This task uses histogram-based purging that defines a freespace_goal, based
upon recent usage patterns.
• This task cannot be used to sacrifice a recovery window goal for any database.
The Recovery Appliance uses a PURGE_IMMEDIATE path when no space is left for
allocations. To do so:
• This task executes as a special purpose scheduler.
• This task will sacrifice recovery window goals if needed.
• Requesters have to wait for purge to find space.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
– If you lose a Storage Cell, you still have a copy of the data in
the Recovery Appliance.
• Delta Store hosts database blocks in Container files (2TB in
Length).
• Deleting a Database does not release space in the ASM le
group; instead, blocks are released in the Container Files. erab
n s f
-t r a
n on
h a sa
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
The base l l o
econfigurationli includes three storage servers that are internally connected using
b
em
ehigh-speed InfiniBand. It provides a highly available configuration with 94 TB of usable
ak capacity for incoming backups. The base rack can be upgraded incrementally by adding
additional storage servers into the rack, up to a maximum of 18 storage servers in a full rack.
Each storage server adds 32TB of usable capacity. The total usable capacity of a full rack is
580TB, with an effective capacity of up to 5.8 Petabytes of Virtual Full Backups.
The Recovery Appliance can protect databases whose total size is approximately the same
as the available capacity of the appliance for a typical recovery window of 10 days. For
example, a single full-rack configuration with 18 storage servers that has a 580TB usable
capacity can protect approximately 580TB of source database for a 10-day recovery window,
storing 10 580TB virtual full backups plus all the redo data generated for that 10-day period.
Accurate sizing of the Recovery Appliance depends on several factors related to protected
databases, including the initial database size and growth rate, storage consumed by temp and
undo, free space, database change rate, redo generation rate, desired recovery window, and
compressibility of the database.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
Data Flow
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Protected
Database
128MB
available in No Purge
storage
location?
immediate
a ble
fe r
Yes
an
Perform PURGEs
n - t r
operation
a a
( cen
e
Data Flow l l o li
b
k e em Databases send backup pieces in chunks of 128MB in size to the Recovery
Protected
a Appliance HTTP Servlet (XMLDB). If the Recovery Appliance has enough free space to
allocate them, it continues receiving chunks until it receives the total number of chunks in the
backup piece. If there isn’t enough free space left in the Delta Store, the Recovery Appliance
immediately executes a purge operation to release and free up space to make room for more
chunks. After all the backup piece chunks are received, the Recovery Appliance executes the
INDEX BACKUP task that indexes a backup piece into one or more Delta Pools (separate
Delta Pool for each datafile).
a a
( cen
l o
e Files li
Containerl
b
k e emRecovery Appliance uses Flash Cache (WriteBack mode) to store backups that it
The
a receives inside Container files. It improves the write-intensive operations because writing to
Flash Cache is much faster than writing to hard disks. After that, Storage Cell Software writes
them back onto the hard disk. In addition to improving write operations, Flash Cache improves
read operations to enable the Recovery Appliance to create Delta Pools.
Allocation units in Container files are used for the following three purposes:
• Collections of blocks that will be stored in a Delta Pool (chunks)
• Segments of backup pieces put in a storage location (may also include partial logs)
• Metadata that link segments of a backup piece together
a a
( cen
l l o li describing what data needs to be read from disk and in what order it
A Planbisea list of pointers
mbe read. Many Recovery Appliance internal tasks create plans to help complete their
kee
must
a work.
– Validate tasks
– Purge tasks
– Miscellaneous tasks
• Priority: Try to run tasks with the lowest number (highest priority)
first.
• States
a ble
– EXECUTABLE fe r
a n s
– RUNNING
n -t r
o
– COMPLETED
s an
– RETRYING
) ha eฺ
– FAILED
ฺ c om Guid
– CANCEL | CANCELLING
n k ng ent
– CANCELLED nba tud
u nio this S
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Internal
b tasks li
el are executed as jobs in the Recovery Appliance database. The Recovery
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a a
( cen
l o
e Applianceli Metadata Database
Recoveryl
b
k e emkey component of the Recovery Appliance is the Recovery Appliance metadata
The
a database. This database manages metadata for all backups, and also contains the RMAN
recovery catalog. The Recovery Appliance metadata database is preconfigured, pre-tuned,
and managed by the Recovery Appliance.
Recovery Appliance Schema
The Recovery Appliance schema contains metadata used internally by the Recovery
Appliance to manage backups on behalf of its protected databases. RASYS is the Recovery
Appliance administrative user who owns the Recovery Appliance schema. The Recovery
Appliance schema contains the Recovery Appliance catalog and the Delta Store Metadata.
account that is employed for backup and recovery operations. Each Recovery Appliance user
account is also a virtual private catalog account.
Recovery Appliance Delta Store Metadata
This metadata stores configuration data such as user definitions, protection policy definitions,
and client database definitions. The metadata database also stores the contents of the Delta
Store.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a k
l o (a cens
el RecoveryliAppliance for data protection of your protected databases, metadata for
To usebthe
em protected databases must be stored in the Recovery Appliance catalog and not in the
ethese
ak RMAN recovery catalog. Existing backup metadata that is currently stored in the RMAN
recovery catalog can be imported into the Recovery Appliance catalog.
To perform protected database backup or recovery operations by using a Recovery
Appliance, you must connect to the protected database and to the Recovery Appliance
catalog.
The RMAN Recovery Catalog includes functionality to create virtual private RMAN catalogs
(VPC) for groups of databases and users. In Oracle Database 12c Release 1 (12.1.0.2), the
RMAN recovery catalog is created and managed by using Oracle Virtual Private Database
(VPD), providing better performance and scalability when a large number of virtual private
catalogs are created.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
data.
– Options (implemented outside the Recovery Appliance)
— Transparent Data Encryption (TDE, protected database
tablespaces)
— Oracle Net Security (network transfer)
— Redo Encryption (real-time redo transport)
a ble
Tape Encryption (drive-based hardware) fe r
—
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Appliance
The Recovery li does not provide server-side encryption, which means that the
b
em does not itself encrypt and decrypt data.
eappliance
ak Options:
• TDE: Oracle recommends that you enable TDE on tablespaces in the database, and
then take incremental backups as usual. TDE requires the Advanced Security Option
(ASO).
• Oracle Net Security: It encrypts the transfer of backups and redo to the Recovery
Appliance.
• Redo Encryption: It encrypts redo both at rest on the Recovery Appliance and during
the network transfer to the appliance.
• Tape Encryption: The Recovery Appliance supports tape drive-based hardware
encryption. In this case, the tape drive encrypts the data, and the software.
a k
l o (a cens
el
A clientbdatabase li backups are managed by a Recovery Appliance is called a protected
whose
m
kee
database.
a A protected database can be a Standard or Enterprise Edition database on any storage or
volume manager (such as ASM, filesystem, ACFS, ZFS, and so on) and can use any of the
following database options: Oracle Real Application Clusters, Oracle Data Guard, Oracle
Streams or/and Oracle GoldenGate.
Recovery Appliance Backup Module: This backup module is the shared library that
transfers backup data to and from the Oracle Recovery Appliance. It must be installed in the
host of the protected database.
a a
( cen
Storage
l o
eLocation li
l
b
m ASM Disk Group serves as the destination for protected database backups. A
eAeDELTA
a k Recovery Appliance storage location can be shared among multiple protected databases.
The Recovery Appliance administrator decides which clients will use the storage location.
Benefits:
• Efficient disk usage: The Recovery Appliance uses common storage to absorb spikes
from all protected databases, thereby reducing the total amount of over-allocated
storage. Oracle recommends that protected databases continue to maintain fast
recovery areas for storage of local online and archived redo log files, control file auto-
backups, and flashback logs.
• Database-optimized backup deduplication and compression
• The shared disk backup pool is distributed based on a database protection policy, which
defines the disk recovery window goal for each database protected by the policy.
h a s
The Recovery Appliance checks the polling directory for newly
m ) ideฺbackups. When backups
created
exist, the Recovery Appliance copies the backups from
g ฺ cothe tpolling
G u directory to its internal
n n en
Recovery Appliance storage location, and thenkprocesses them. After enough time has passed
a the
for the Recovery Appliance to copy the backups,
b t d
uprotected database deletes the backups
n S
from the polling directory.
u nio this
Redo Staging Area
l l o @ use
For Recovery Appliance
a k beinstallations
e to that enable real-time redo transport, this is the destination
for redo streams
l o e ns databases transmit to the Recovery Appliance. The staging area
(athat cprotected
resides on
b el RecoveryliAppliance disks (the DELTA ASM disk group). The Recovery Appliance
collects
m data into archived redo log files, which it then converts to compressed archived redo
e e
ak backups that it writes to a Recovery Appliance storage location.
log
– The recovery window goal specifies the days, counting backward from the current time,
within which a point-in-time recovery is possible.
– It effectively extends the number of days, if free space is available.
– Sufficient Space: Backups created before a recovery window goal may be available (but
not guaranteed by default).
– Insufficient Space: The Recovery Appliance may purge backups attempting to respect the
recovery window requirement first.
• Reserved space (GB)
– Disk space (in GB) is guaranteed to each protected database to meet its recovery window
a ble
goal.
fe r
– Sufficient Space: Reserved space is ignored and the recovery window goal settings apply.
an s
– Insufficient Space: The recovery window goal is ignored and reserved space settings
apply. The Recovery Appliance purges backups for the database whose backups exceed n - t r
the Reserved Space by the highest percentage.
a no
• Guaranteed copy (NO/YES) h a s
– Older backups need to be copied to tape or a replicam ) iAppliance.
Recovery
d e ฺ
– NO (Default): Old backups are purged when more
g ฺ cospacet isGneeded
u even if the backups are
not copied or replicated.
a n kn old dbackup
e n before it has been copied or
nb s Stu
– YES: The Recovery Appliance never purges
replicated.
n i o i
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
e duty oflithe Recovery Appliance administrator is to plan for the appropriate
l
An important
b
em of disk space for a specified retention window and database size. The Recovery
eamount
ak Appliance avoids storage islands and over-allocation typical to host or LUN-oriented
provisioning. By using the Recovery Appliance, space is dynamically reallocated between
databases to meet recovery window goals, such as recovering to any time in the last 35 days.
Recovery Window Goal
The recovery window goal specifies the interval (typically specified in days), counting
backward from the current time, within which point-in-time recovery must be possible. The
Recovery Appliance attempts to retain sufficient backups to meet the recovery window goal
defined for each database.
Reserved Space
The reserved space defines the amount of disk space guaranteed to each protected database
to meet its recovery window goal.
• YES: When a guaranteed copy is enabled, the Recovery Appliance never purges a
backup before it has been copied to tape or replicated. After the Recovery Appliance has
consumed the reserved space with backups that have not been copied to tape or
replicated, it does not accept new backups or redo. This setting only changes the
behavior of storage management when the tape system or the replicated Recovery
Appliance is unavailable for an extended period.
The Recovery Appliance uses recovery window goals and reserved space settings to allocate
storage dynamically to meet business requirements. If the Recovery Appliance has purged as
a ble
e r
much backup data as possible while still meeting the recovery window goal for each database,
f
ans
and if more space is needed, then the Recovery Appliance evaluates the reserved space
n - t r
setting of each database. Recovery Appliance purges backups for the database whose backups
o
exceed the reserved space by the highest percentage.
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
Minimal Impact
Eliminate Data Loss
Backups
Real-time redo
Production databases
transport provides
only send changes; all
instant protection of
backup and tape
ongoing transactions
processing offloaded
a ble
Database Level Cloud-Scale Protection
fe r
Recoverability
an s
End-to-end reliability,
Easily protect all
n - t r
databases in the data
visibility, and control of
a no center using massively
databases, and not of
h a s scalable service
disjointed files
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
Database-Level li
Recoverability: Each incremental block received is validated, compressed,
b
k e emthen indexed as a part of a new virtual full backup, which appears as a normal full backup
and
a the RMAN catalog. Essentially, this means validated full backups are now available at any
in
point in time, without ever incurring the cost of running full backup operations.
End-to-end Oracle
Block Validation
DW BI CRM ERP
Application-oriented
Protection Policy
a ble
3 fe r
an s
n - t r
a no
h a s
m
Integrated ) ideฺ
EM Cloud Control:
ฺ co t Gu
Management
g
a n kn den
Policy-Based Management: Application-Oriented
n b S t uData Protection
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
The Recovery
b li understands internal Oracle database block formats, which enables
el Appliance
m levels of data validation. All backup data and redo blocks are automatically validated as
kee
deep
a they are received by the Recovery Appliance, as they are copied to tape, and as they are
replicated. In addition, backup blocks are also periodically validated on disk. This ensures that
recovery operations will always restore valid data – another unique differentiator that is only
possible because of the Recovery Appliance’s deep database integration. If a corruption is
discovered during validation, the Recovery Appliance’s underlying storage software
automatically reads the good block from a mirrored copy and immediately repairs the
corrupted block.
Traditional
• Minimum amount of data transferred Restore on
- Only restore virtual full backup Friday
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Faster Recovery
recovery Merge
- Virtual full backups are ready to use
- Only apply redo logs Restore
Thursday
• Quicker site-failure recovery Recovery Restore
Appliance Wednesday
- Direct restore from replicas or tape Restore
Restore
Tuesday
a ble
Apply Restore fe r
Redo Log
an
Monday s
n - t r
Restore
a no Restore
Virtual Full Full
a s
Backup
h eฺ Backup
)
ฺ c om Guid
Delta Push and Virtual Full Backupsa n kngRapid
Enable d e n t
Recovery to Any Point In Time
n b t u
u nio this S
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
The Recovery
b li uses virtual full backups to provide rapid recovery to any point in
el Appliance
emregardless of the amount of data being recovered. The on-disk recovery strategy of
etime,
ak Recovery Appliance has the advantage that RMAN can recover virtual full backups to any
point in time without applying incremental backups.
When a database is protected by the Recovery Appliance, RMAN must only restore a single
level 0 backup for the day of the RPO, and then recover up to the last second by using redo
log files sent via the real-time redo transport feature. For example, if the recovery window is 7
days, and if the RPO is 5 days ago, then RMAN can restore a single virtual full (level 0)
backup that is current to 5 days ago, and then recover it by using redo, and not level 1
incremental backups.
n b a
S t ud
Transmission
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
el Appliance
The Recovery
b li is able to reassemble virtual full backups located in the Delta Store
emphysical full backups and validate them. After that, these physical full backups are sent
einto
ak thought the network by using the HTTP Servlet. If no available virtual full backups are in, the
local Recovery Appliance automatically sends a request to a Recovery Appliance replica to
re-assemble and validate them into a physical full backup on behalf of it and transfer it back to
be sent directly through the network by using the HTTP Servlet. In addition, for archival
backups, it is able to automatically request the integrated Media Manager or third-party media
servers the physical full backups archived in the tape libraries to be sent directly through the
network by using the HTTP Servlet too.
Minimal Impact
Eliminate Data Loss
Backups
Real-time redo
Production databases
transport provides
only send changes; all
instant protection of
backup and tape
ongoing transactions
processing offloaded
a ble
Database Level Cloud-Scale Protection
fe r
Recoverability
an s
End-to-end reliability,
Easily protect all
n - t r
databases in the data
visibility, and control of
a no center using massively
databases, and not of
h a s scalable service
disjointed files
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
Database-Level li
Recoverability: Each incremental block received is validated, compressed,
b
k e emthen indexed as a part of a new virtual full backup, which appears as a normal full backup
and
a the RMAN catalog. Essentially, this means validated full backups are now available at any
in
point in time, without ever incurring the cost of running full backup operations.
policies
a ble
Silver Policy – Internal Critical
fe r
Disk: 10 days
an s
Tape: 45 days
Replica
n - t r
a no
h a s
Bronze Policy - Test/Dev
m ) ideฺ Replica Recovery
Disk: 3 days
g ฺ co t Gu Appliance also
Tape: 30 days
n k n en Policy-Based
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
The Recovery
b li introduces the concept of a protection policy, which defines recovery
el Appliance
em that are enforced on a per-database basis on the appliance and on tape. Using
egoals
ak protection policies, databases can be easily grouped by recovery service tier. The Recovery
Appliance includes predefined “Platinum”, “Gold”, “Silver”, and “Bronze” policies, which can
be customized to support various business service-level agreements. For example, database
backups under the Gold policy target a 35-day recovery window on a local Recovery
Appliance and 90 days on tape, while backups managed under the Silver policy target a 10-
day recovery window on a local Recovery Appliance and 30 days on tape. Tiered protection
policies are also independently applied on the remote replicated Recovery Appliance.
The Recovery Appliance uses protection policies to simplify the management of protected
database backups. A protection policy is a collection of attributes that defines recovery goals
and storage space requirements for one or more protected databases. A Recovery Appliance
contains multiple protection policies that define varied recovery goals. Each protected
database is assigned exactly one protection policy that determines how the Recovery
Appliance stores and maintains backup data for that protected database.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
Recovery
fe r
Sales DB
Window Goals
an s
n - t r
a no
h a s
m ) ideฺ
Web DB
g ฺ co t Gu
a n kn den Sales Web
n b t u
u nio this S
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
el
Using protection
b li as the basis, the Recovery Appliance fully manages all backup
policies
k e em space according to each database’s recovery window goals. For example, the
storage
a “Finance Database”, a member of the Gold policy, can be recovered within the past 35 days,
while the “Products Database”, a member of the Silver policy, can be recovered within the
past 10 days. If free space is available in the Delta Store, backups older than the recovery
window goal can be retained, effectively extending the recovery window. Upon space
pressure, the Recovery Appliance purges backups and automatically re-provisions space
between databases to meet the recovery window goals for every protected database. The
appliance may also purge backups proactively, in advance of any space pressure, based on
historical space usage – again, in order to meet recovery windows goals for all databases.
Space is purged in a database intelligent fashion with an understanding of the dependencies
between datafiles, redo logs, and control files.
This recovery window-oriented space management approach eliminates the need to manage
space at an opaque storage-volume level as is typical with generic backup appliances. With
this innovative approach, data protection is aligned with each application’s business criticality,
and manual rebalancing of space is eliminated.
a a kb se t
l o ( cen
l
e Appliance
The Recovery
b li is based on the Oracle Exadata architecture and therefore inherits
k e em
its proven scalability, redundancy, and performance. As additional databases within the
a enterprise are protected by the Recovery Appliance, compute servers and storage servers
can be easily added to the appliance, providing a simple, no-downtime, scale-out data
protection cloud that seamlessly supports business growth.
• When a Protection Policy is added to the Replication Server on Upstream Appliance, all future
incremental and archived log backups for its protected databases are queued for replication.
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
HTTP
a ble
fe r
Full (Level 0) Backup
an s
n - t r
no
Storage
Location
s a
Incremental and
Archived Log Backups h a
) ideฺ
Virtual Full
Backups
m
co t Gu
Upstream Downstream g
n en ฺ Archived Log
Backups
Appliance nkAppliance
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
In Recovery
b li replication, one Recovery Appliance (the upstream Recovery
el Appliance
m
kee
Appliance) forwards backups to another Recovery Appliance (the downstream Recovery
a Appliance). After initial configuration, replication is fully automatic. Each Recovery Appliance
in a replication topology manages its own protection and polling policies.
Because a downstream Recovery Appliance processes backups independently from the
upstream Recovery Appliance, a downstream Recovery Appliance can have completely
different policies for every database whose backups it is storing.
Recovery Appliance. In this case, each Recovery Appliance plays both the upstream and
downstream roles in the replication topology. Every Recovery Appliance serves as the
primary backup location for one set of protected databases, and the secondary backup
location for the other set. In this way, every Recovery Appliance is actively utilized while
also providing disaster recovery services for the other Recovery Appliance.
• Hub and Spoke: Backups flow from one set of databases to a local Recovery Appliance,
and from a different set of databases to a different local Recovery Appliance. The local
Recovery Appliances then forward these backups to a single remote Recovery Appliance,
a ble
which archives the backups to tape.
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• Pre-bundled Media Manager is the default Recovery Appliance tape archival solution.
• Backups are stored on tape as complete backup sets, and not as virtual backups.
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
Full Backup
an s
CTDB Day 1 Virtual Full
Day 0 Day 1
n - t r
Full Incr
a no
h a s
Incremental &
Archived Log Backupsom
) ideฺ
g ฺ c t Gu
Recovery Appliance
a n kn den Tape Library
b t u
n i on is S
@ u e th
b e llo to us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
( a ak nse
e l
The Recovery
lo Appliance
l i ce includes a built-in media manager. The Recovery Appliance offloads
m b
ak ee
tape backup operations from protected databases to the Recovery Appliance. Thus, protected
database hosts do not need the RMAN-integrated media management software module.
Instead, a single copy of the pre-bundled media manager module is installed on the Recovery
Appliance. The Recovery Appliance automatically manages the copy of backups to tape for
all protected databases.
Protected databases send backups to the Recovery Appliance, which stores them on disk in
the specified storage locations. To reclaim disk space and to create transportable tape
backups, business requirements may necessitate archival to tape.
Backups are stored on tape as complete backup sets, and not as virtual backups, so tape
backups are usable by RMAN without mediation by the Recovery Appliance.
Recovery
Appliance
a ble
fe r
Tape
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
Protected
a n kn den Remote
Databases
i o nb s Stu
Enterprise Manager Appliance
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Cloud b l l o
e lists every
Control li Recovery Appliance in the enterprise. The Recovery Appliance
k e em page is a command center that centralizes management of the Recovery Appliance
home
a environment. From this page, you can manage Recovery Appliance storage and performance,
and view recent activity and issues that may need attention. The storage locations page
expands on the information provided in the storage locations section on the Recovery
Appliance Home page.
It is easy to provision protected databases for the Recovery Appliance.
• Recovery Appliance Administrator
- Chooses new database in the Enterprise Manager list
- Assigns protection policy
- Sets new database credentials
• Database Administrator
- Selects Recovery Appliance target in Enterprise Manager
- Enables real-time redo transport
• Recovery Appliance replicas and Media Manager software have Enterprise Manager
Cloud Control plugins to enable the management and monitoring of these components by
using the web or CLI interface, and utilities of Cloud Control.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
• Pre-created reports
– Capacity Planning summary and details
– Recovery Window analysis Tape
– Protected Databases details
– Top 10 ranks
• Manual and automatic scheduling
r a ble
• Custom reports for querying metadata catalog views for extended
n s fe
a
reporting n-tr
no
s a
h a
) ideฺ
m
co t Gu
g ฺ
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Through l l o li the Recovery Appliance provides pre-created BI Publisher reports
ethe BI Publisher,
b
k e emenable you to meet the following goals:
that
a • Ensure that the Recovery Appliance has sufficient storage space for its needs: By
using the capacity reports, you can plan for additional storage, reduce the number of
new protected databases added to the Recovery Appliance, or adjust protection policies
so that the aggregate recovery window space decreases.
• Ensure that the network is not overloaded: The capacity reports also indicate
whether the Recovery Appliance has maximized network capacity. In some cases, you
can reduce network load by redistributing network traffic more evenly throughout the
day. If network traffic is not distributed, and if network peaks are close to maximizing
network bandwidth, then you may need to adjust the backup window times of some
protected databases.
• Provide a good view of system performance and activity for service requests.
• Obtain a brief or highly detailed status report for any protected database, which can
sometimes be useful for troubleshooting databases that are not meeting recovery
window goals
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a k
l o (a cens
bel li
em
ake
Objectives
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
• Installation Process
• Extending and Expanding the Recovery Appliance
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a a
( cen
l l o li the Oracle Exadata Database Machine, the Recovery Appliance, or
e can configure
The OEDA
b
emOracle SuperCluster. The OEDA deployment tool is used to perform the initial Oracle
ethe
ak Recovery Appliance configuration. It loads network settings, creates user accounts, installs
Oracle Recovery Appliance software, and secures the system based on information in the
configuration files created by the OEDA configuration tool.
You can choose from multiple platforms to run the OEDA tool: Microsoft Windows, Linux, Mac
OS X, and Solaris (SPARC and x86-64).
See the following support note for additional information: Zero Data Loss Recovery Appliance
Supported Versions (Doc ID 1927416.1): Patch Release History for the ZDLRA section
(OneCommand)
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a a
( cen
Consider l l o li guidelines:
ethe following
b
k e e•m Do not run the OEDA deployment tool on Exadata Storage Servers.
a • When used for deployment (installation), the deployment assistant is run on any of the
following systems that can access the Exadata database servers:
- Oracle Exadata Database Machine database server
- Oracle SuperCluster database domain
- Oracle SuperCluster database zone
- Microsoft Windows machine
- Oracle Linux machine
- Oracle Solaris SPARC machine
- Apple OS X machine
• The configuration utility is the config.sh script. This tool creates some files and
directories that must not be edited or changed.
a b le
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• You must select the Zero Data Loss Recovery Appliance option here.
• Do not import legacy configurations that are not ZDLRA
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
configurations.
• Do use import any time you have a new version of configurator and
an existing ZDLRA configuration.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
The OEDA l l o
e configures li the engineered system based on the configuration information entered
b
ak ebyem
the customer. The OEDA is provided as a compressed file. To use the OEDA, extract the
file on the client server. After the file has been extracted, change to the directory that has the
extracted folder, and run one of the following commands:
• On Linux, Apple, and UNIX: config.sh
• On Microsoft Windows: config.cmd
• The name prefix is used to generate a host name for all rack
components: compute nodes, storage nodes, switches, and
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Customer l l o li
e Details information:
b
k e e•m Customer Name (Required): The customer name can contain alphanumeric characters
a and spaces
• Application: The application used on the Oracle Exadata Database Machine or another
protected Oracle Database system
• Network Domain Name (Required): The name of the domain in which Recovery
Appliance will operate
• Name Prefix (Required): The prefix used to generate host names for all rack
components. The prefix can contain up to 20 alphanumeric characters. White space is
invalid. For example, a value of ra01 results in a compute server name of ra01db01
and a storage server name of ra01cel01.
• Region: The geographic area where Recovery Appliance will be installed. Select a
region from the list.
• Time Zone: The time zone for the installation site. Select a time zone from the list. The
displayed time zones are within the selected region.
• DNS: Up to three IP addresses for the DNS servers
• NTP: Up to three IP addresses for the Network Time Protocol (NTP) servers
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e selection
The hardware li page identifies the number of racks that you want to deploy at the
b
em time, and the number of storage servers within each rack. The Recovery Appliance
esame
ak supports a flexible configuration. You can have between 3 and 14 storage servers in each
rack. Multiple racks must be cabled together.
Oracle Recovery Appliance X5-2 and X6-2 support Elastic Rack configurations. Elastic Rack
configurations give you flexibility at the time you need to configure the number of Storage
Nodes and racks. You will learn more about this later. Oracle Recovery Appliance X4-2 has
predefined configurations, starting with three Storage Nodes and expanding up to 14 Storage
Nodes per Rack.
For example, to deploy two full Oracle Recovery Appliance X4-2 racks (14 storage servers)
and one minimally configured Oracle Recovery Appliance X4-2 rack (3 storage servers) at the
same time:
1. Expand the X4 ZDLRA folder.
2. Add X6-2 ZDLRA 14 Cell HC 4TB twice.
3. Add X4-2 ZDLRA 3 Cell HC 4TB once.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Appliance
The Recovery li comes with two compute nodes and the number of storage nodes of
b
emchoice.
eyour
ak Click the check box if you want to include a spine switch as a part of the rack configuration
(for multipack deployments). Do this for each rack that will have a spine switch.
• Ingest Network
• Accepts backups locally, similar to the client network in
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Exadata
Network Description Ports Required IP Requirement
Ingest Network This required network 4 x 10 GbE connections Enable Jumbo Frames.
connects the protected (UTP or Fiber)
Oracle Database servers to The preferred scenario is
the Recovery Appliance for Active via LACP.
backup. A private Ethernet
network must be designed
a ble
to support the transfer of fe r
large volumes of data by
an s
using techniques such as
n - t r
jumbo packets.
a no
You can use this network if
h a s
you are connecting to third-
party media servers for m ) ideฺ
tape.
g ฺ co t Gu
There is VLAN port tagging nk
n en
n b
support on this network. a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
el network
This required
b li connects the protected Oracle Database servers to the Recovery
em for backup within the same data center. Also known as a backup network, this high-
eAppliance
ak speed, private Ethernet network must be designed to support the transfer of large volumes of
data by using techniques such as jumbo packets. The Recovery Appliance connects to this
network by using two 10 GB connections to each of the two compute servers in the rack. You
can configure the two connections as active/passive (redundant) or active/active. The
compute servers support channel bonding to provide higher bandwidth and availability.
The single client access name (SCAN) supports failover between the two compute servers in
the Recovery Appliance. In an installation with multiple Recovery Appliance racks configured
as a cluster, VIP addresses support failover among the racks. The protected database
systems can resolve the host names to dynamically assigned addresses.
Third-party tape hardware and software also use the ingest network.
Refer to the following note for additional information on VLAN port tagging support: Enabling
8021.Q VLAN Tagging in Zero Data Loss Recovery Appliance Over Ingest Networks (Doc ID
2047411.1)
• Replication Network
• Not used in a local configuration
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
l l o
e replication
The optional li network connects the local Recovery Appliance (the upstream
b
em with a remote Recovery Appliance (the downstream appliance). Oracle
eappliance)
ak recommends a broadband, encrypted network, instead of an insecure public network,
wherever possible.
The Recovery Appliance supports the following configurations between the upstream and
downstream appliances:
• One-way: Data flows from the upstream appliance to the downstream appliance.
• Bidirectional: Data flows in both directions between the upstream appliance and the
downstream appliance.
• Hub and spoke: Data flows from multiple upstream appliances to one downstream
appliance.
• InfiniBand Network
• Non-routable network automatically configured
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
InfiniBand Network This network connects Internal to the Recovery You will require a subnet
the compute servers Appliance in OEDA for the IB
and storage servers by network.
using the InfiniBand IB partitions are not
a ble
switches. Oracle supported today. If you have to connect
fe r
Database uses this Because of this, to a second system
an s
network for Oracle RAC
cluster interconnect
Exadata or (Exadata or
SuperCluster running SuperCluster), you have n - t r
traffic and for accessing no
multiple IB partitions to use the same subnet
a
data on the storage
a s
cannot be connected. as that system, but
h
servers.
m ) ideฺ different IP addresses.
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e connects
This network li the compute servers and storage servers using the InfiniBand
b
em Oracle Database uses this network for Oracle RAC cluster interconnect traffic and
eswitches.
ak for accessing data on the storage servers. This non-routable network is fully contained in the
Recovery Appliance.
This network does not connect to your data center networks. It is automatically configured
during installation.
components
IP Address
Network Description Ports Required
Requirement
Fiber Channel Arbitrated If you are using the pre- You require two ports
Loop (FCAL) Network bundled media per server (four in a
manager, then you can rack).
a ble
back up the Recovery
fe r
Appliance to the storage You must use a SAN.
an s
area network (SAN) in
n - t r
your data center for
backups to tape. a no
The ingest network is
used to communicate
a s
with Automated
h
) ideฺ
Cartridge System
m
g ฺ co t Gu
Library Software
a a
( cen
If you are l l o li
eusing the pre-bundled Media Manager, then you can back up the Recovery
b
m to the SAN in your data center for backups to tape. The network connections
kee
Appliance
a depend on whether you have an Oracle tape solution or use third-party hardware.
• Pre-bundled Media Manager: When you use the Oracle-recommended tape solution, a
fiber channel adapter is installed in each compute server to provide a connection to the
fiber channel SAN. Tape backups are isolated on this network and, thus, do not interfere
with the performance of the other networks.
• Third-Party Tape Systems: When you use a third-party tape system, the backups to
tape use the 10 Gb ingest network. This is the same network that the local protected
databases use to back up to the Recovery Appliance.
• Management Network
• Connects through the Ethernet switch to the compute servers,
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
e l l
The management
o li connects through the Ethernet switch to the compute servers, the
network
b
em servers, and the InfiniBand switches. It connects directly to the PDUs.
estorage
ak This (required) 1 gigabit, Ethernet (GbE) network connects to your existing management
network. You use it to do administrative work on all components of the Recovery Appliance.
The management network connects to the servers, Oracle ILOM, and InfiniBand switches
through the Cisco Ethernet switch in the rack.
There are three uplinks to the management network:
• One from the Ethernet switch
• One from each of the two PDUs. Network connectivity to the PDUs is required only if the
electric current is monitored remotely.
Each server has two network interfaces for management. One provides management access
to the operating system through the NET0 Ethernet interface, and the other provides access
to Oracle ILOM through the ILOM Ethernet interface. The Recovery Appliance is delivered
with the ILOM and NET0 interfaces connected to the Ethernet switch on the rack.
Cabling or configuration changes to these interfaces is not permitted. Do not use the
management network interfaces for client or application network traffic. Instead, you can use
NET1, NET2, NET3, or the two interfaces on the PCI slot.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Appliance
The Recovery li uses up to four networks. Each network must be on a separate
b
em from the others:
esubnet
ak • Subnet 1 (required): The administration network provides access to the servers and
switches in the Recovery Appliance, either directly or through Oracle ILOM. This
network typically uses a copper cable.
• Subnet 2 (required): The ingest network connects the Oracle Exadata Database
Machine to a local Recovery Appliance. This network can be bonded, and it uses either
a copper cable (1 Gb or 10 Gb) or an optical fiber cable (10 Gb). The Link Aggregation
Control Protocol (LACP) is optional. The data center must provide network and switch
support.
• Subnet 3 (required): The private InfiniBand network connects components within the
rack and, optionally, multiple racks together. This network uses InfiniBand cables.
• Subnet 4 (optional): The replication network connects a Recovery Appliance to an
optional, remote Recovery Appliance. This network can be bonded, and it uses either a
copper cable (1 Gb or 10 Gb) or an optical fiber cable (10 Gb). A replication network is
also called a backup, DR, or independent client network.
- 1/10 Gb Copper Base-T for copper cables (typically for laptops and personal
computers)
- 10 GbE Optical for fiber optic cables (typically for data centers)
• Enabled LACP: The option for the ingest network if LACP is enabled in the data center
• Available Network: The option to configure the optional replication network
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
e serverli and storage server has two network interfaces for the administration
l
Each compute
b
k e em
network:
a • An Ethernet interface provides management access to the operating system.
• An ILOM Ethernet interface provides access to Oracle Integrated Lights Out Manager
(Oracle ILOM).
Other options in this screen include:
• Starting IP Address for Pool: The first IP address in a pool of consecutive addresses
assigned to the administration network port of the compute servers, storage servers,
and InfiniBand switches. Addresses in this pool are also assigned to the ILOM port of
the database and storage servers.
• Pool Size: This value is calculated from the hardware selection. The pool should consist
of consecutive IP addresses. If consecutive IP addresses are not available, then you
can modify individual IP addresses on the review page.
• (*) Sample host names: It displays sample names on the administration network, using
the prefix entered on the Customer Details page. To modify the name format, click
Modify.
(*) You can change the default administration host names by using the Administration
Network Format Masks dialog box. Names are typically changed to meet corporate naming
standards or for compatibility with existing systems.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
The ingestl l o li
e network provides the connection between Oracle Database systems and Recovery
b
em for bulk data transfers. It is the backup network.
eAppliance
ak This screen shows the following options:
• Starting IP Address for Pool: The first IP address in a pool of consecutive addresses
assigned to the ingest network port of the compute servers.
• Pool Size: This value is calculated from the hardware selection. The pool should consist
of consecutive IP addresses. If consecutive IP addresses are not available, then you
can modify individual IP addresses later in the configuration process.
• Ending IP Address for Pool: This value is calculated from the starting IP address and
the pool size.
• Is the Default Gateway for Compute Servers: Select this option if the gateway IP
address is the default gateway for the compute servers. Oracle recommends this
configuration.
• Defines the Host Name for the Compute Servers: Select this option if the ingest
network names define the host names for the compute servers. (*)
• Sample Ingest Names: Displays sample names on the ingest network, using the prefix
entered on the Customer Details page. To modify the format, click Modify.
(*) You can change the default ingest network host names on the Ingest Network Format
Masks dialog box.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e InfiniBand
The private li network connects the compute servers and storage servers through the
b
em gateway switches in the rack. The InfiniBand network can also connect multiple
eInfiniBand
ak Recovery Appliance racks.
This screen options:
• Starting IP Address for Pool: The first IP address in a pool of consecutive addresses
assigned to the InfiniBand network port of the compute servers, storage servers, and
InfiniBand switches.
• Pool Size: This value is calculated from the hardware selection. The pool should consist
of consecutive IP addresses. If consecutive IP addresses are not available, then you
can modify individual IP addresses on the review page.
• Ending IP Address for Pool: This value is calculated from the starting IP address and
the pool size.
• Enable Active Bonding on Compute Node Network: Select this option to support
failover between the two compute servers on the InfiniBand network.
• Sample InfiniBand Names: Displays sample compute server and storage server
names on the InfiniBand network, using the prefix entered on the Customer Details
page. To modify the name format, click Modify.
You can change the format of the InfiniBand network host names by using the Private (IB)
Network Format Masks dialog box.
replication.
• When using replication, a separate network is highly recommended.
Without it, you would use up ingest network bandwidth.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e replication
The optional li network connects the Recovery Appliance to another (downstream)
b
em Appliance. It provides dedicated bandwidth for this task. Complete this page if you
eRecovery
ak have a replication network.
If a replication network is unavailable, then you can use the ingest network to connect to a
downstream appliance. However, this configuration creates competition for bandwidth on the
ingest network.
This screen shows the following options:
• Starting IP Address for Pool: The first IP address in a pool of consecutive addresses
assigned to the replication network port of the storage servers. The pool size identifies
the required number of IP addresses. If consecutive IP addresses are not available, then
you can modify individual IP addresses on the review page.
• Pool Size: This value is calculated from the hardware selection.
• Ending IP Address for Pool: This value is calculated from the starting IP address and
the pool size.
• Sample Replication Name: Displays a sample compute server name on the replication
network, using the prefix entered on the Customer Details page. To modify the format,
click Modify.
You can change the replication host name format by using the Replication Network Masks
dialog box.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Appliance
The Recovery li runs the Linux operating system. There are no choices to be made
b
ak eonem
this page.
regenerate data.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Use this l l o li the network names and IP addresses assigned to each device. If you
epage to review
b
em a pool of IP addresses that are not all available for use, then replace the duplicate
eidentified
ak addresses on this page. You can edit the host names and IP addresses just by clicking the
field. All host names and IP addresses must be unique.
• Regenerate Data: Click this button to see the current list of generated names, after
modifying the names or format masks.
• Lookup IP: It checks that the IP address matches the device name, and corrects it if
they do not match.
configuration.
• Ensure that the cluster name is unique here.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
You can
b i
l one lcluster
ecreate on the Recovery Appliance.
em
eOptions:
ak • Cluster Name (required): A unique name for the cluster. The name must start with an
alphanumeric character. It can contain additional letters, numbers, periods (.), and
hyphens (-). The default name is cluster-clu1.
• Available: It lists all compute servers and storage servers included in this installation.
• Selected: It lists all compute servers and storage servers that are assigned to the
cluster. Click All to assign all servers to the cluster.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li have values that you provided on earlier configuration pages. You
e on this page
Many fields
b
k e emreview these values and revise them if necessary. The greyed out fields are read only,
can
a and are shown only to provide additional information about the installation.
• Cluster: The values at the top of the page are derived from the rack configuration
settings on the Customer Details screen.
• Users and Groups: This section is read only. It displays the users and groups defined
for Oracle Database on the Recovery Appliance. You do not access this database
directly, because it exclusively supports the backup and recovery of protected
databases.
• Software Locations: This section identifies the location of the Oracle Database
software on the Recovery Appliance. Most fields are read only.
• Disk Group Details: This section defines the disk groups for Oracle Database on the
Recovery Appliance. Most fields are read only. Oracle recommends HIGH redundancy
for DELTA ASM disk groups only for mission-critical applications.
• Initial Database: It includes the Cluster Database name and character set.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Use this l l o li the network names and IP addresses on the SCAN, ingest network,
epage to review
b
emnetwork, and replication network. If you identified a pool of IP addresses that are not all
eVIP
ak available for use, then replace the duplicate addresses on this page. You can edit the host
names and IP addresses just by clicking the field. All host names and IP addresses must be
unique.
• Regenerate Data: Click this button to update the list of generated names, if you
modified any names or format masks.
• Lookup IP: Checks that the IP address matches the device name, and corrects it if they
do not match.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li if you previously configured a replication network. The Recovery
e appears only
This page
b
em defines a SCAN for the rack, and a VIP for the compute servers. All names and IP
eAppliance
ak addresses must be unique.
• Regenerate Data: Click this button to update the list of generated names, if you
modified any names or format masks.
• Lookup IP: Checks that the IP address matches the device name, and corrects it if they
do not match.
The Recovery Appliance does not require cell alerting but you
must configure one of the following alerting options:
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
• Cell alerting
• OCM
• ASR
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li when a problem is detected in a storage server. The alert can be
e are triggered
Cell alerts
b
em using the Simple Mail Transfer Protocol (SMTP), Simple Network Management
edelivered
ak Protocol (SNMP), or both.
The Recovery Appliance does not require cell alerting, but you must configure one of these
alerting options:
• Auto Service Request (ASR)
• Cell alerting
• Oracle Configuration Manager (OCM)
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
OraclebPlatinum li provide enhanced support at no additional cost for qualified
Services
k e em
engineered systems. This is a connected service and requires the installation of Oracle
a Advanced Support Gateway software within the system.
The Oracle Advanced Support Gateway is configured with the user information that is
required by the Platinum Service monitoring agent.
• Customer Details: These include customer name, customer support identifier, and My
Oracle Support email address that corresponds to the CSI.
• Platinum Gateway Details: The Recovery Appliance can use an existing Oracle
Advanced Support Gateway in the network. Otherwise, you can create a gateway on a
server or a supported virtual machine, which is connected to the same management
network or ingest network as the Recovery Appliance.
• Platinum Agent Configuration Details: Oracle Platinum Services uses an agent to
monitor the health of the Recovery Appliance. Use the default configuration settings for
the agent, unless you have a specific reason to change them. For example, the default
port number might be assigned already to another device.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Request
Auto Service li (ASR) automatically opens service requests when specific hardware
b
em occur.
efaults
ak The Recovery Appliance does not require an ASR, but you must configure one of the
following alerting options:
• ASR
• Cell alerting
• Oracle Configuration Manager
Options in this screen:
• Enable Auto Service Request: To use ASR with the Recovery Appliance and then
enter the configuration details
• ASR Manager Host Name: The host name of the server where the ASR Manager is
installed
Note: Oracle recommends a stand-alone server that has connectivity to the Recovery
Appliance.
• ASR Technical Contact Name: The name of the technical contact person for the
Recovery Appliance
• Proxy Port: The port number for the HTTP proxy server. The default is 80.
• HTTP Proxy requires authentication: Select if the HTTP proxy server requires
authentication
• HTTP Proxy User: The username for the proxy server. If the connection requires a
password, then you must provide it during installation.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li Manager collects and uploads configuration information to an
e Configuration
The Oracle
b
em repository for access when needed by Oracle Support.
eOracle
ak The Recovery Appliance does not require the Oracle Configuration Manager, but you must
configure one of the following alerting options:
• ASR
• Cell alerting
• Oracle Configuration Manager
• Most libraries are Default SCSI. This is an industry standard using the FC-
SCSI interface for robotic control.
• You can add one or more tape libraries here, but usually just one is added.
• All tape libraries will be hosted inside the same domain.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
You can l l o li tape library for the Recovery Appliance only if you use the pre-bundled
econfigure the
b
em Manager for tape backups. You can manually configure other media managers as
eMedia
ak clients only.
Options:
• Use Tape: Select this option if you plan to use the pre-bundled Media Manager for tape
backups. Then you can enter the configuration details.
• Number of Tape Libraries: It denotes the number of tape libraries available to the
Recovery Appliance. All tape libraries must be hosted in the same domain. A separate
subpage is created for each library, so that you can enter the configuration details.
• Enable Admin VIP: Select this option to create a VIP on the ingest network and enable
the Media Manager as a Cluster Ready Services (CRS) service. Then you can enter the
VIP details.
• VIP Name: This is the alphanumeric name of the VIP.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
• Installer Comments
• There might be default questions here to help the person
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Enter any
l o
eadditionalliinformation about the deployment in the text field, particularly anything of
l
b
k e emto the person performing the installation. The information appears at the bottom of
help
a InstallationTemplate.html, which is one of the generated files.
Save.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Selectb eldirectory lini which you will save the configuration files.
the
k e em
a
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
e page listsli the names of the generated files. Click Finish to close the utility.
l
A summary
m b
a kee
checkip.sh
-rw-r--r-- 1 user staff 2989 Jul 26 11:21 Test_Customer-demo-
platinum.csv
-rw-r--r-- 1 user staff 6038 Jul 26 11:21 Test_Customer-demo-
preconf_rack_0.csv
-rw-r--r-- 1 user staff 218159 Jul 26 11:21 Test_Customer-
demo.xml
a ble
-rw-r--r-- 1 user staff 218116 Jul 26 11:20 Test_Customer- fe r
ans
demo.xml.26-Jul-16_112101
n - t r
-rw-r--r-- 1 user staff 27853 Jul 26 11:21 Test_Customer-
a no
demo.zip
h a s
-rw-r--r-- 1 user staff 8761 Jul 26 11:21
m d ฺ
) databasemachine.xml
e
ฺ co t Gu i
host:ExadataConfigurations user $ g
kn den
a n
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
a a
( cen
e l l o li
b
k e em
a
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Template
The Installation li report includes the following sections with detailed information:
b
ak ee•m Management Network Configuration
• Ingest Network Configuration
• Private InfiniBand Network Configuration
• Replication Network Configuration
• Clusters Information
• Rack Machines Overview
• Platinum Configuration
• Tape Library Connectivity
• Alerts
• Appendix A: Network Names and IPs
• Appendix B: Required Files and Patches
a a
( cen
After an
l o
eOracle ACSliengineer completes these steps, the installation is complete and the
l
b
em is ready to hand off to the customer.
esystem
ak
a k
l o (a cens
As instructed
b i
el by thelOEDA Installation Template, you must download the following files from
emOracle Support (MOS) into the OEDA's WorkDir directory in the first Compute Node:
eMy
ak • Database 12c software
- linuxamd64_12102_database_1of2.zip
- linuxamd64_12102_database_2of2.zip
• Grid Infrastructure 12c software
- linuxamd64_12102_grid_1of2.zip
- linuxamd64_12102_grid_2of2.zip
• Database Patch for Engineered Systems and DB In-Memory 12.1.0.2.13 (OCT2015).
• p22304421_1210213DBEngSysandDBIM_Linux-x86-64.zip
OPatch patch of version 12.2.0.1.5 for Oracle software releases (installer) 12.1.0.x and
12.2.0.x (JUL 2016)
• p6880880_121010_Linux-x86-64.zip
Exachk utility. Refer to the following support note to get the latest version: Oracle Exadata
Database Machine exachk or HealthCheck (Doc ID 1070954.1) and Zero Data Loss Recovery
Appliance Supported Versions (Doc ID 1927416.1) a ble
fe r
• exachk.zip
a n s
r
-t in addition
ZDLRA RA Automation RPM (includes ra_preinstall.pl and ra_install.pl perlnscripts
o
to other required files)
s an
• ra_automation-12.1.1.1.7-22712784.x86_64.rpm
) ha eฺ
QLogic QConvergeConsole GUI/CLI Management Tools
ฺ c om G u id Library integration is
(if the Tape
n k ng ent
enabled) Go to http://driverdownloads.qlogic.com/QLogicDriverDownloads_UI for download.
n b a
• QConvergeConsoleCLI-1.1.03-22.x86_64.rpm
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
b el li
k e em
a
upgrades.
• The script is included in the Recovery Appliance patch .zip
file.
• It should be run from there prior to new installs and upgrades.
[root@zdlraadm01 WorkDir]# /usr/bin/perl ra_preinstall.pl \
a ble
--config_xml=/u01/ra_install/linux-64/ExadataConfigurations/ZDLRA1AutoInstall-zdlra01.xml \
fe r
--oeda_dir=/u01/ra_install/linux-64/ --oracle_uid=4000 --dba_gid=4000 --oinstall_gid=4001
Updating OEDA for ra_automation-12.1.1.1.7-23494283.x86_64
an s
ra_automation-12.1.1.1.7-23494283.x86_64.rpm and /u01/ra_install/linux-
n - t r
no
64//WorkDir/ra_automation-12.1.1.1.7-23494283.x86_64.rpm match. Skipping copy.
[root@zdlraadm01 WorkDir]#
s a
h a
) ideฺ
m
co t Gu
g ฺ
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o li performs initial checks and prepares the Recovery Appliance for new
b
ra_preinstall.pl
k e em
installations and upgrades.
a The command shown in this slide shows an example of executing the ra_preinstall.pl
script.
Script to patch the config xml with the new patch bundles and
update es.properties with the current rpm name.
ARGUMENTS
The script takes the following arguments
--config_xml (REQUIRED)
--config_xml=CONFIG.XML_PATH The path where the config.sh xml
a ble
file is located.
fe r
--oeda_dir (REQUIRED)
a n s
--oeda_dir=OEDA_DIRECTORY_PATH The path of the oOEDA n r
-t directory.
--oracle_uid (OPTIONAL) s an
--oracle_uid=NEW_USER_ID The new oracle ) hausereฺid. Value must be
>= 1001 ฺ c om Guid
--oinstall_gid (OPTIONAL) n k ng ent
n b a Thetudnew oinstall group id. Value
nio to
--oinstall_gid=NEW_GROUP_ID
i s S
must be >= 1001 and not uequal
@ e t h 1011 or 1012.
e
--dba_gid (OPTIONAL)
b llo to us
a ak nse
--dba_gid=NEW_GROUP_ID
( The new dba group id. Value must be >=
1002 andllonot equal e
lic to 1011 or 1012.
b e
e m--help (OPTIONAL)
a ke Provides the scripts help message.
EXAMPLES
./ra_preinstall.pl --config_xml=/scratch/ZDLRA7_Auto_Install-
scao09.xml --oeda_dir=/scratch/linux-x64 --oracle_uid=5555 --
oinstall_gid=5555 --dba_gid=5555 --help
n i o i
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
e l
The install.sh
b li uses the files generated by OEDA to configure the networks, Oracle
script
em and other software components. The script is run by the root user.
eDatabase,
ak The command shown in the slide shows an example for listing all the steps for a specific
installation.
ARGUMENTS:
-l List all the steps that exist
-cf config file name [Full path please]
-s <step #> Run the step # at a time
-r <num-num> Run the steps one after the other as long as no
errors
are encountered
r a ble
-u <num-num> | <step#> Undo a range of steps or a particularfestep
a n s
-h Usage
n -t r
o
Version: 15.153.07:00
s an
) ha eฺ
[root@slcm03adm03 linux-x64]#
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
b el li
k e em
a
n
Step 6 - ZDLRA DB Backup Setup
a
nb s Stu
Step 7 - Enable ZDLRA Services
n i o i
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
e l
The ra_install.pl
b li script performs additional configuration steps for the Recovery
em infrastructure. You must run this script while logged in to a Recovery Appliance
eAppliance
ak compute server, and not from another server on the same network.
The command shown in the slide shows an example for listing all options and steps.
• Discovering Targets:
– Cluster: Oracle Cluster and High Availability Services
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li Manager to administer the Recovery Appliance, you must deploy the
e the Enterprise
To enable
b
em
emanagement agents to each compute server, then discover the targets for the Recovery
ak Appliance. The targets include the cluster, database, listeners, ASM, and the appliance itself.
Refer to the following support note for additional information: Prerequisites for Using the
Oracle Zero Data Loss Recovery Appliance Plug-in (12.1.0.1) (Doc ID 1929507.1)
You will learn more later about configuring the Recovery Appliance with Oracle Enterprise
Manager Cloud Control.
• It requires SSH.
• The execution node may be one of the nodes being tested.
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
e helps youlito identify the theoretical network throughput between a Protected
This scriptl
b
em system and a Zero Data Loss Recovery Appliance, or between two Zero Data Loss
eDatabase
ak Recovery Appliances in a replication configuration.
Refer to the following support note for additional information: Zero Data Loss Recovery
Appliance Network Test Throughput script (Doc ID 2022086.1)
Reference:
# ./network_throughput_test.sh -h
Usage: ./network_throughput_test.sh -s [sending_nodes_file] -r
[receiving_nodes_file] -i [interface]
• Installation Process
• Extending and Expanding the Recovery Appliance
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
• Base Rack
– 2x Compute Servers with high-speed connectivity
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
h a sAppliance
g ฺ co t Gu
a n kn den Fully
nb s Stu
Redundant
n i o i
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
l
e Appliance
The Recovery
b li racks have a flexible configuration that allows a minimum of three
em servers. You can extend a rack up to a total of 18 storage servers for Recovery
estorage
ak Appliance X6.
The base configuration includes two compute servers and three storage servers internally
connected by using high-speed InfiniBand. It provides a highly available configuration with 94
TB of usable capacity for incoming backups. The base rack can be upgraded incrementally by
adding additional storage servers into the rack, up to a maximum of 18 storage servers in a
full rack. Each storage server adds 32 TB of usable capacity. The total usable capacity of a
full rack is 580 TB, with an effective capacity of up to 5.8 PB of virtual full backups.
Accurate sizing of the Recovery Appliance depends on several factors related to protected
databases including the initial database size and growth rate, storage consumed by temp and
undo, free space, database change rate, redo generation rate, desired recovery window, and
compressibility of the database.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• Scale to 18 Racks
– Up to 216TB/hour Delta Ingest and Restore
– More than10 PB of usable capacity
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
You can l l o li
eadd racks together to increase system capacity and performance. The Recovery
b
em has a flexible configuration. A rack has two compute servers and a minimum of
eAppliance
ak three storage servers. A Recovery Appliance X6 or X5 full rack has 18 storage servers and a
Recovery Appliance X4 full rack has 14 storage servers. The maximum configuration is 18 full
racks.
If additional capacity is required, a second base rack can be connected via high-speed
InfiniBand to the first rack. The second rack includes its own pair of compute servers which
add connectivity and processing power to the configuration. As with the first rack, storage
capacity can be easily expanded by incrementally adding storage servers. Up to 18 fully
configured racks can be connected together into a single appliance, providing more than 10
PB of usable capacity, i.e. more than 100 Petabytes of Virtual Full Backups.
The power and flexibility of the Recovery Appliance scale-out architecture is revealed when
there is a need to support additional databases, or when business data grows. Storage,
compute, and network capacity is incrementally added in a balanced fashion that maintains
high performance. This architecture is far superior to traditional backup appliances which are
usually limited to two controllers and therefore cannot scale storage, networking, and compute
in a balanced, bottleneck-free fashion.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a k
l o (a cens
bel li
em
ake
Objectives
a k
l o (a cens
bel li
em
ake
• Configuration: Overview
• Recovery Appliance Administrative Interfaces
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
l l o li
m be
e
ake
General tasks:
• Planning
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a k
l o (a cens
bel li
em
ake
• Configuration: Overview
• Recovery Appliance Administrative Interfaces
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
l l o li
m be
e
ake
( a ak nse
e l
Administration l ce
lo Interfaces
i
m b
e• Cloud Control: This is a system management tool with a graphical user interface that
ake enables you to manage and monitor the Recovery Appliance and its protected
databases. This is the preferred UI for Recovery Appliance tasks.
• Secure Shell (SSH) allows remote access, but local access might be necessary during
initial power-on or troubleshooting.
• ILOM provides a remote management console using multiple protocols, such as
HTTP/HTTPS, IPMI, SNMP, and so on.
• With local access, you have access to the node operating system as well as the ILOM.
• You use SQL statements and Oracle-supplied PL/SQL packages to complete these
tasks in SQL*Plus or SQL Developer.
• The dcli utility facilitates centralized management across an server realm by
automating the execution of OS and utilities commands on a set of servers. The utility
then returns the output to the centralized management location where the dcli utility
was run.
• Configuration: Overview
• Recovery Appliance Administrative Interfaces
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
l l o li
m be
e
ake
a ble
fe r
ZDLRA EM Agent
an s
and OMS plug-ins
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
b
Recovery
l
eAppliance li Discovery Prerequisites
k e em Server
Compute
a If agents are already installed on the Recovery Appliance compute servers and uploading to
the Oracle Management Server, proceed to the Recovery Appliance Discovery Prerequisites
section below. Otherwise, management agents must be pushed to each compute server by
using the following procedure:
1. From the Setup menu, select Add Target, select Add Targets Manually, choose Add
Host Targets, and then click Add Host.
Enterprise Manager displays the Add Host Targets: Host and Platform page.
2. On the Host and Platform page, for the Platform choice, select Same for All Hosts. Use
the Add button to add all Recovery Appliance compute servers to the list (specify the
fully qualified host name of each server). Select the proper platform for the first server in
the list. Click Next.
Enterprise Manager displays the Add Host Targets: Installation Details page.
Guided Discovery
a ble
Workflow
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Recoveryb l
eAppliance li Manual Discovery
k e em
Steps
a After discovering the cluster and cluster database, the Recovery Appliance target itself can be
discovered. As a part of discovering the top-level Recovery Appliance target that will include
all Enterprise Manager Recovery Appliance software and hardware management
functionality, a separate Recovery Appliance hardware target will also be discovered.
Follow these steps to manually discover a Recovery Appliance target:
1. From the Setup menu, choose Add Target, then select Add Targets Manually.
Enterprise Manager displays the Add Targets Manually page.
2. On the Add Targets Manually page, select the Add Targets Using Guided Process (Also
Adds Related Targets) option.
3. From the Target Types drop-down list, choose Recovery Appliance, then click Add
Using Guided Process. Enterprise Manager displays the Recovery Appliance Hardware
Discovery page.
the Target Creation Summary page, showing a summary of all hardware targets created,
including the Database Machine target representing the Recovery Appliance hardware
and all member targets for all hardware components.
6. On the Target Creation Summary page, click Continue With Recovery Appliance
Discovery. Enterprise Manager displays the Recovery Appliance Discovery: Properties
page.
7. On the Recovery Appliance Discovery: Properties page, select the Target Name for the
a ble
Recovery Appliance. The Recovery Appliance Hardware target name is already filled-in
fe r
ans
with the target name discovered above.(If the Recovery Appliance hardware discovery
- t r
was completed without completing the full Recovery Appliance discovery process, that
n
no
may leave one or more Recovery Appliance hardware targets in a state where they are
a
h a s
not associated with a Recovery Appliance target. In this case, when Recovery Appliance
) ideฺ
discovery is re-initiated from the Add Targets Manually page, the discovery process will
m
g ฺ co t Gu
not automatically go in to the Recovery Appliance hardware discovery wizard. Instead, the
a n kn den
process will go directly to the Properties page in order to allow selection of an
i o nb s Stu
unassociated Recovery Appliance hardware target. In this case, the hardware target
name will not be pre-filled. Click the Select Target icon to launch the Search and Select
u n t h i
Targets popup, which shows a list of Recovery Appliance hardware targets that have not
l lo@ o us e
e
yet been associated with a Recovery Appliance target. Select the correct Recovery
kb se t
a
Appliance hardware target.)
a
8. l
In the l o
Recovery
l i en
( cAppliance Administrator Credentials section, specify the database user
m be
credentials for the Recovery Appliance recovery catalog owner. This user is also the
a kee Recovery Appliance administrator. You can use Named or New credentials. Click More
Details to see detailed information about the credentials.
9. In the Recovery Appliance Monitoring Credentials section, specify the database user
credentials that will be used to monitor the Recovery Appliance. If necessary, that user
will be granted the role required to access monitoring information in the Recovery
Appliance database. Optionally, you can select the option to Use base catalog user
credentials as monitoring credentials which disables the username and password fields.
10. Click Next to move to the Media Manager Domain discovery page.
11. If Media Manager is installed on the Recovery Appliance, specify the installation home
directory and monitoring credentials for the Media Manager domain. Otherwise, check the
option to skip Media Manager discovery.
12. Click Next to move to the Review page.
13. On the Review page, you can review the target discovery selections. The page lists the
Target Name, Target Type, Recovery Appliance Hardware, Database System, and
Cluster Database.
14. Click Submit to complete the discovery process. The Recovery Appliance target is
created.
• Configuration: Overview
• Recovery Appliance Administrative Interfaces
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
l l o li
m be
e
ake
Separation of Duties
• Cloud Control Administrator duties:
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
i o nb s Stu
– Perform backup and recovery operations
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
A typical l l o li
eRecovery Appliance environment includes personnel with the following roles:
b
•m Cloud Control administrator: The application administrator with this role administers
a kee Oracle Enterprise Manager Cloud Control (Cloud Control).
- Discovering targets, including the Recovery Appliance
- Managing one or more protected databases
- Managing one or more Recovery Appliances
• Recovery Appliance administrator: This administrator manages the Recovery
Appliance. Typical duties include:
- Creating the protection policies
- Assigning protected databases to protection policies
- Managing space on Recovery Appliance
- Configuring tape and replication operations
- Creating the Recovery Appliance user accounts that own virtual private catalogs
- Monitoring Recovery Appliance, and generating reports
• Protected Database administrator: This administrator is responsible for configuring
backups to the Recovery Appliance using the virtual private catalog account assigned
by the Recovery Appliance administrator.
User accounts:
• Cloud Control super-user: SYSMAN
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
ZDLRA_ADMIN:
g ฺ co t Gu
sample Cloud
Control administrator
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
• be Controllisuper-user (SYSMAN): This application account exists by default. Its
l
Cloud
m purpose is to administer Cloud Control itself, and is not directly related to managing a
e
ake Recovery Appliance or protected databases.
• Cloud Control administrator: This is a Cloud Control user account that has been
granted the roles and privileges needed to discover Recovery Appliances and Targets
and to manage a specific protected database or a specific Recovery Appliance. Multiple
Cloud Control administrative accounts may exist, depending on your business
requirements.
– Username
– Password
– Password Profile
– Email
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li
e administrator
m b
Cloud Control (Recovery Appliance Target)
e following steps describe how to create an Enterprise Manager administrator (for example,
ake
The
ZDLRA_DB_ADMIN) that will have the necessary roles, target privileges, resource privileges,
and credentials to discover and manage one or more Recovery Appliances for use with Cloud
Control:
1. Log in to Enterprise Manager as an Enterprise Manager administrator (for example,
SYSMAN) that has the privileges to create other Enterprise Manager administrators.
2. From the Setup menu, choose Security then select Administrators. Click Create, which
displays the Create Administrator wizard.
3. On the Properties page, supply the name (for example, ZDLRA_EM_ADMIN),
password, password profile for security best practices and email for alert notification.
In this example we use the MGMT_ADMIN_USER_PROFILE password profile. Create
or choose a password profile that best adapts to your business requirements.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
b
Cloud Control li
el administrator (Recovery Appliance Target)
e m
ake
Steps
4. On the Roles page, ensure EM_USER is in the Selected Roles.
5. On the Target Privileges page, complete these steps:
a) In the Privileges applicable to all targets section, if the administrator needs the
ability to discover a Recovery Appliance, enable the Add Any Target privilege.
b) In the Target Privileges section, add the following privileges for the specified
targets, depending on whether the administrator needs the ability to discover
and/or manage the Recovery Appliance.
Discover:
- View privilege for the Agent targets that are monitoring the Recovery
Appliance compute nodes.
- Full privilege for the Recovery Appliance target. (This privilege will be
automatically configured for a user that discovers the Recovery Appliance
target.)
- View privilege for the targets corresponding to each client database that will be
protected by the Recovery Appliance.
6. On the Resource Privileges page, complete these steps:
a) For the Job System resource type, click the Manage Privilege Grants icon. In the
a ble
Resource Type Privileges table, check Create and continue.
fe r
ans
b) The administrator will either need access to existing named credentials or the ability
n - t r
to create named credentials. The credentials involved depend on whether the
a no
administrator needs the ability to discover and/or manage the Recovery Appliance:
h a s
- To give the administrator the ability to create named credentials, click the
m ) ideฺ
g ฺ co t Gu
Manage Privilege Grants icon on the Resource Privileges page under the
Named Credential resource type. In the Resource Type Privileges table, check
n kn den
Create new Named Credential and continue.
a
i o nb s Stu
- To grant the administrator access to existing credentials, from the Resource
u n t h i
Privileges page, for the Named Credential resource type, click the Manage
l lo@ o us e
Privilege Grants icon. In the Resource Privileges table, click Add, then search
e
kb se t
and select for the required named credentials and continue.
a a
(discover n manage a Recovery Appliance target, the administrator will need
o
c) llTo l i c eand/or
m be to have access or be able to create the following credentials:
a kee - Credentials for the Recovery Appliance Administrator (for example, RASYS).
The named credentials (as viewed by choosing Security from the Setup menu
and then selecting Named Credentials) must be of type Recovery Appliance
Administrator Credentials.
- Credentials for the host of the Recovery Appliance. The named credentials
must be of type Host Credentials.
- If different than the Recovery Appliance Administrator, credentials for the
Recovery Appliance Monitoring user (for example, DBSNMP). The named
credentials must be of type Recovery Appliance Administrator Credentials.
- Credentials for the Media Manager Domain administrator (for example,
ADMIN). The named credentials must be of type Media Manager Domain
Credentials.
8. Click Finish.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
– Username
– Password
– Password Profile
– Email
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li
e administrator
m b
Cloud Control (Protected Databases Target)
e following steps describe how to create an Enterprise Manager administrator (for example,
ake
The
ZDLRA_DB_ADMIN) that will have the necessary roles, target privileges, resource privileges,
and credentials to manage one or more protected databases for use with Recovery
Appliance:
1. Log into Enterprise Manager as an Enterprise Manager administrator (for example,
SYSMAN) that has the privileges to create other Enterprise Manager administrators.
2. From the Setup menu, choose Security then select Administrators. Click Create, which
displays the Create Administrator wizard.
3. On the Properties page, supply the name (for example, ZDLRA_DB_ADMIN),
password, password profile for security best practices and email for alert notification.
In this example we use the MGMT_ADMIN_USER_PROFILE password profile. Create
or choose a password profile that best adapts to your business requirements.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Cloudb Control li
el administrator (Protected Databases Target)
• Configuration: Overview
• Recovery Appliance Administrative Interfaces
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
l l o li
m be
e
ake
Tasks
1. Plan for needed policies by grouping protected databases
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Appliance
The Recovery li provides a Data Protection as a Service infrastructure allowing you
b
m an effective data protection private cloud environment including disk backup
ak etoecreate
(Recovery Appliance), copy-to-tape and/or replication.
Basic tasks involved in managing protection policies:
• During the planning phase, group the databases into tiers, and decide the recovery
requirements for each tier.
• During the configuration phase, create one protection policy for each database tier.
• During the ongoing maintenance phase, modify protection policies as needed.
Gold
35 days on disk, 90 days on Database backups, real-time redo
transport, replication, and tape
a ble
tape backups (if tape is available).
fe r
Database backups, real-time redo
an s
Silver
35 days on disk, 90 days on
tape available). n -
transport, and tape backups (if tape is
t r
a no
Bronze
h a s
35 days on disk, 90 days on Database backups and tape backups
(if tape is available). There is no real-
tape
m ) ideฺ time redo transport.
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Appliance
The Recovery li provides a Data Protection as a Service infrastructure, allowing you
b
m an effective data protection private cloud environment, including disk backup
ak etoecreate
(Recovery Appliance), copy-to-tape, and/or replication.
Protection policies define a class of service of recovery requirements, allowing you to
standardize data protection by database tier. For example, you may want to create a “Gold”
protection policy for mission-critical databases, “Silver” for business-critical and “Bronze” for
Test/Dev databases.
Protection policies are the foundation for Data Protection As a Service. Each database under
Recovery Appliance management must be associated with a protection policy and can then
be managed as a group for defining replication and/or copy to tape schedules.
A protection policy is a named collection of properties that you can assign to multiple
protected databases. Using a single policy for multiple databases reduces Recovery
Appliance administration time, and enables you to change the properties of multiple protected
databases with one operation. To accommodate databases with differing backup and
recovery requirements, create as many protection policies as required.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
ea protection
To create li policy using Oracle Enterprise Manager Cloud Control:
b
ak ee1.m Access the Protection Policies page.
2. Click Create. The Create Protection Policy page appears.
3. Enter values as follows:
- In the Name field, enter the name of the new protection policy. For example, enter
bronze_dev.
- In the Description field, enter a description for the new policy. For example enter:
Policy with disk recovery window of 3 days and no tape
backup.
- In the Recovery Window field of the Disk Recovery Window Goal section, specify
a recovery window goal that the Recovery Appliance should attempt to meet for
point-in-time recovery using disk backups, and then select the units. For example,
enter 3 and then select days.
- In the Threshold field of the Unprotected Data Window Threshold section, enter
the maximum tolerable interval for data loss. For example, enter 5 and then select
minutes.
backups unless you explicitly purge them or space pressures exist within a storage
location.
• Optionally, in the Backup Polling Location section, define a backup polling policy. For
example, to specify that backup polling is disabled, leave the fields blank.
- In the Location field, specify a directory accessible by the Recovery Appliance.
- In the Frequency field, specify a time interval, and then select the time units.
- To specify that the Recovery Appliance must delete the backups from the polling
a ble
location after copying them, select Delete Backups After Copy. fe r
ans
•
n - t r
In the Backup Copy Policy section, specify whether the Recovery Appliance must
a no
replicate backups or copy backups to tape before deleting them. For example, select
Always accept new backups even if it requires purging existing backups not yet copied
to tape or replicated. h a s
m ) ideฺ
4. Click OK.
g ฺ co t Gu
The Protection Policies page appears, with
a n knthe newly
d e n created protection policy listed.
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
a a
( cen
l l o li
m be
e
ake
SQL Developer
Creating Protection Policies
• You must log in to the metadata database as RASYS.
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
polling_policy_name IN VARCHAR2 DEFAULT NULL,
recovery_window_goal IN DSINTERVAL_UNCONSTRAINED,
fe r
max_retention_window IN DSINTERVAL_UNCONSTRAINED DEFAULT NULL,
an s
recovery_window_sbt IN DSINTERVAL_UNCONSTRAINED DEFAULT NULL,
unprotected_window IN DSINTERVAL_UNCONSTRAINED DEFAULT NULL,
n - t r
guaranteed_copy IN VARCHAR2 DEFAULT 'NO',
allow_backup_deletion IN VARCHAR2 DEFAULT 'YES');
a no
h a s
PROCEDURE create_polling_policy(
m ) ideฺ
polling_policy_name IN VARCHAR2,
polling_location IN VARCHAR2,
g ฺ co t Gu
delete_input IN BOOLEAN DEFAULT FALSE);
a kn den
polling_frequency IN DSINTERVAL_UNCONSTRAINED DEFAULT NULL,
n Query a
b t u Protection Policy
n i on is S
@ u e th
b e llo to us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
( a ak nse
l loprotection
l i e
cpolicy,
b e
To create a execute the DBMS_RA.CREATE_PROTECTION_POLICY
em
eprocedure:
ak 1. Start SQL*Plus or SQL Developer, and then log in to the metadata database as RASYS.
2. (Optional) Run the DBMS_RA.CREATE_POLLING_POLICY procedure.
Parameters:
• polling_policy_name: The user-assigned name of the polling policy.
• polling_location: The directory that the Recovery Appliance periodically examines
for new backups. Do not specify the same directory name in multiple polling policies.
• polling_frequency: The frequency with which the Recovery Appliance examines
the specified directory for new backups. System load may cause backup polling to occur
less frequently. Specify the window as any valid INTERVAL DAY TO SECOND
expression, such as INTERVAL '2' DAY (2 days), INTERVAL '4' HOUR (4 hours),
and so on.
n k ng ent
days), INTERVAL '4' HOUR (4 hours), and so on.
b a ud
• max_retention_window: The maximum length of time that the Recovery Appliance
n S t
nio this
must retain backups for databases that use this protection policy. Recovery Appliance
u
l l o @ use
only holds backups longer than the specified period when they are required to preserve
a k be e to
the recovery window goal for a database. If null, then the Recovery Appliance does not
purge backups unless caused by explicit user actions or space pressures within a storage
o
location.
l (a cens
bel li
• recovery_window_sbt: The recovery window for SBT backups of databases that use
e m
ake
this protection policy. For each protected database, the Recovery Appliance keeps
backups long enough on tape to guarantee that a recovery is possible to any time within
the specified interval, counting backward from the current time. If this parameter is not
null, then you must also create an SBT job for this protection policy, and then schedule it
using a scheduling facility such as Oracle Scheduler. See CREATE_SBT_JOB_TEMPLATE.
Specify the window as any valid INTERVAL DAY TO SECOND expression, such as
INTERVAL '2' DAY (2 days), INTERVAL '4' HOUR (4 hours), and so on.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li policy using Oracle Enterprise Manager Cloud Control:
e a protection
To update
b
ak ee1.m Access the Protection Policies page.
2. In the Protection Policies table, select the protection policy that you want to edit. For
example, select the BRONZE_DEV row.
3. Click Edit. The Edit Protection Policy page appears.
4. Change the desired values, and then click OK. For example, in the Recovery Window
field of the Disk Recovery Window Goal section, enter 6.
The Protection Policies page appears, with the newly updated protection policy listed.
an s
unprotected_window IN DSINTERVAL_UNCONSTRAINED DEFAULT dbms_ra_int.intervalnull('p4'),
guaranteed_copy IN VARCHAR2 DEFAULT NULL,
n - t r
no Update a
allow_backup_deletion IN VARCHAR2 DEFAULT NULL);
s a Protection Policy.
PROCEDURE update_polling_policy (
polling_policy_name IN VARCHAR2, h a
) ideฺ
polling_location IN VARCHAR2 DEFAULT NULL,
m
co t Gu
ฺ
polling_frequency IN DSINTERVAL_UNCONSTRAINED DEFAULT NULL,
g
kn den
delete_input IN BOOLEAN DEFAULT NULL);
a n
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li policy, execute the DBMS_RA.UPDATE_PROTECTION_POLICY
e a protection
To update
b
em
eprocedure. Parameters that are null retain their existing values. For example, if
ak guaranteed_copy is currently NO for a protection policy, and if you specify null for this
parameter in DBMS_RA.UPDATE_PROTECTION_POLICY, then the value remains NO.
1. Start SQL*Plus or SQL Developer, and then log in to the metadata database as RASYS.
2. Run the DBMS_RA.UPDATE_PROTECTION_POLICY procedure.
3. Optionally, query the recovery catalog to confirm the update of the policy.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
ea protection
To delete li policy using Oracle Enterprise Manager Cloud Control:
b
ak ee1.m Access the Protection Policies page.
2. In the Protection Policies table, select the protection policy that you want to delete. For
example, select the BRONZE_DEV row.
3. Click Delete. A confirmation window appears.
4. Click Yes.
The Protection Policies page appears, with the deleted protection policy no longer listed.
(Optional) Delete the Backup Polling Policy if it is also no longer needed.
PROCEDURE delete_polling_policy (
a ble
polling_policy_name IN VARCHAR2);
fe r
an s
n - t r
a no
h a s
m ) ideฺ
ฺ co t Gu
Query Protection
g
n not exist.
Policykdoes
n e n
a d
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
ea protection
To delete li policy, execute the DBMS_RA.DELETE_PROTECTION_POLICY
b
em
eprocedure:
ak 1. Start SQL*Plus or SQL Developer, and then log in to the metadata database as RASYS.
2. Confirm that the protection policy that you intend to delete is not currently associated
with any protected databases.
3. Delete the policy.
4. Optionally, confirm the deletion.
(Optional) Delete the Backup Polling Policy if it is also no longer needed.
a ble
fe r
an s
VPC user n - t r
creation
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
e connectlito the Recovery Appliance catalog when backing up to a Recovery
RMANbmustl
k e em
Appliance. In this step, you create a virtual private catalog user for a specific protected
a database or set of protected databases.
1. With SQL*Plus or SQL Developer, connect to the Recovery Appliance database as
SYSTEM.
2. Create a database user account with CREATE SESSION privileges to own the virtual
private catalog.
• Configuration: Overview
• Recovery Appliance Administrative Interfaces
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
l l o li
m be
e
ake
a a
( cen
The key
l o
ecomponentlofi the Recovery Appliance is the Recovery Appliance metadata
l
b
em This database manages metadata for all backups, and also contains the RMAN
edatabase.
ak recovery catalog. The Recovery Appliance metadata database is preconfigured, pre-tuned,
and managed by the Recovery Appliance.
• RFS = Remote File Server
• SCAN = Simple Client Access Network
• XMLDB = XML Database
– Benefits
— High Availability (node failure)
— Scalability (1000s Protected Databases)
— Load Balance (distribute load between RAC instances)
— IB network for cluster interconnection and ASM storage traffic
Topology
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
OraclebClusterware i Release 1 is the integrated foundation for Oracle Real Application
l12c
k e em (RAC) and the High Availability (HA) and resource management framework for all
Clusters
a applications on any major platform.
Additional information about Oracle Clusterware benefits:
http://www.oracle.com/technetwork/database/database-
technologies/clusterware/overview/index-090666.html
--------------------------------------------------------------------------------
ob_dbfs
ONLINE ONLINE host1 STABLE
ONLINE ONLINE host2 STABLE
ora.CATALOG.dg
ONLINE ONLINE host1 STABLE
le
ONLINE ONLINE host2 STABLE
ora.DELTA.dg
r a b
ONLINE ONLINE host1 STABLE
s fe
ONLINE ONLINE host2 STABLE
- t r an
o n
an
ora.LISTENER.lsnr
ONLINE ONLINE host1 STABLE
STABLEha
s
ONLINE ONLINE host2
m ) ideฺ
ora.LISTENER_REPL.lsnr
g ฺ coSTABLE
t G u
ONLINE ONLINE host1
host2 an
kn deSTABLE
n
nb s Stu
ONLINE ONLINE
ora.asm i o
n host1thi
ONLINE ONLINE u
l o @
l to u s e Started,STABLE
ora.net1.network k
b e
ONLINE ONLINE host2 Started,STABLE
( a a nse
e l lo ONLINE i
ONLINE
l ceONLINE host1 STABLE
kee
ora.net2.network
a ONLINE ONLINE host1 STABLE
ONLINE ONLINE host2 STABLE
ora.ons
ONLINE ONLINE host1 STABLE
ONLINE ONLINE host2 STABLE
osbmedia
ONLINE ONLINE host1 STABLE
ONLINE ONLINE host2 STABLE
rep_dbfs
ONLINE ONLINE host1 STABLE
ONLINE ONLINE host2 STABLE
ora.LISTENER_REPL_SCAN3_NET2.lsnr
1 ONLINE ONLINE host1 STABLE
ora.LISTENER_SCAN1.lsnr
1 ONLINE ONLINE host1 STABLE
ora.LISTENER_SCAN2.lsnr
1 ONLINE ONLINE host1 STABLE
ora.LISTENER_SCAN3.lsnr
1 ONLINE ONLINE host1 STABLE
a ble
fe r
ora.MGMTLSNR
ans
1 ONLINE ONLINE host1
t
111.111.111.111 … ,STABLE
n - r
ora.cvu
o
1 ONLINE ONLINE host1 STABLE
s an
ora.mgmtdb
) ha eฺ
1 ONLINE ONLINE host1
c o m uid
Open,STABLE
ora.oc4j ฺ
ng eSTABLE tG
host1 nk
n
1 ONLINE ONLINE
n b a
S t ud
ora.scan1.vip
1
io
ONLINE ONLINE un host1thi
s STABLE
ora.scan1_net2.vip @ e
llo to ushost2
b
k se e
1
ora.scan2.vip( a a
ONLINE ONLINE
n
STABLE
1e l l o l i c e
m b ONLINE ONLINE host1 STABLE
kee 1
ora.scan2_net2.vip
a ora.scan3.vip
ONLINE ONLINE host1 STABLE
slcm03osbadm-vip1
1 ONLINE ONLINE host1 STABLE
--------------------------------------------------------------------------------
[oracle@host1 ~]$
Oracle Real Application Clusters (RAC) is an option to the award-winning Oracle Database
Enterprise Edition. Oracle RAC is a cluster database with a shared cache architecture that
overcomes the limitations of traditional shared-nothing and shared-disk approaches to provide a ble
fe r
highly scalable and available database solutions for all your business applications.
ans
Additional information about Oracle RAC Database benefits: n - t r
o
an
http://www.oracle.com/technetwork/database/options/clustering/overview/index.html
s
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
iDB Protocol
over InfiniBand InfiniBand Storage Network
a b le
fe r
a n s
Oracle Linux Oracle Linux Oracle Linux -t
n r
Cell
n o
Control CLI CELLSRV MS CELLSRV MS a MS
CELLSRV
s
(cellcli/dcli)
IORM IORM ) a
h IORM ฺ RS
SSH
RS RS
m i d e
g ฺ co t Gu
a n kn den
Exadata Cell
o nbExadata S tu
Cell Exadata Cell
u n i h i s
e t
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
l
e Appliance
The Recovery
b li software architecture includes components on the database servers
emon the Exadata cells. The overall architecture is illustrated in the slide. The following
eand
ak components reside on each database server:
• ZDLRA database servers use Oracle Linux x86_64 as the operating system.
• ZDLRA database servers can run Oracle Database 12c Release 1 with a pre-installed
and pre-configured RASYS schema that contains the Recovery Appliance Metadata.
These servers also include version 12.1 of Oracle Grid Infrastructure to provide
clusterware infrastructure to the resources.
• Automatic Storage Management (ASM) is required and provides a DELTA PUSH,
DELTA STORE and CATALOG for Oracle Recovery Appliance Database.
g ฺ co t Gu
a n knCell den
Exadata Cell
n
Exadata
b S t u Exadata Cell
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
The software
b li that reside in each Exadata cell include the following:
el components
ak ee•m Oracle Linux x86_64 provides the Exadata Storage Server operating system.
• Cell Server (CELLSRV) is the primary Exadata Storage Server software component and
provides the majority of Exadata storage services. CELLSRV is a multithreaded server.
It serves simple block requests, such as database buffer cache reads, and Smart Scan
requests, such as table scans with projections and filters. CELLSRV also implements
IORM, which works in conjunction with DBRM to meter out I/O bandwidth to the various
databases and consumer groups that are issuing I/Os. Finally, it collects numerous
statistics relating to its operations. Oracle Database and ASM processes use
LIBCELL to communicate with CELLSRV, and LIBCELL converts I/O requests into
messages that are sent to CELLSRV by using the iDB protocol.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
h a s
DBMCLI>
m ) ideฺ
co t Gu
CellCLI> exit
quitting
g ฺ
[celladmin@host1 ~]$
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
e utility isli the command-line administration tool for configuring database servers,
l
The DBMCLI
b
emmanaging objects in the server environment. DBMCLI runs on each server to enable you
eand
ak to configure an individual database server. You use DBMCLI to start and stop the server, to
manage server configuration information, and to enable or disable servers. The command-line
utility is already installed when Oracle Recovery Appliance is shipped.
Management Server on database servers implements a web service for database server
management commands, and runs background monitoring threads. The management service
provides the following:
• Comprehensive hardware and software monitoring including monitoring of hard disks,
CPUs, and InfiniBand ports.
• Enhanced alerting capabilities.
• Important system metric collection and monitoring.
• A command-line interface called DBMCLI to configure, monitor, and manage the
database servers. DBMCLI is pre-installed on each database server. DBMCLI
configures Auto Service Request, capacity-on-demand, Infrastructure as a Service, and
database server email alerts.
DBMCLI is a part of the Management Server that runs on database servers.
DB Listeners (back-end)
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
SCAN Listeners (front-end) nkn e n
a d
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
When b ae l l o li
Protected Database submits a request, the SCAN listener listening on a SCAN IP
k e em and the SCAN port is contacted on a client's behalf. Because all services on the
address
a cluster are registered with the SCAN listener, the SCAN listener replies with the address of
the local listener on the least-loaded node (Each scan listener keeps updated cluster load
statistics) where the service is currently being offered. Finally, the client establishes
connection to the service through the listener on the node where service is offered. All of
these actions take place transparently to the client without any explicit configuration required
in the client.
For additional details: Grid Infrastructure Single Client Access Name (SCAN) Explained (Doc
ID 887522.1)
a a
( cen
Automatic l l o li
e Storage Management (ASM), a feature of the Oracle Database, provides Oracle
b
k e em with simplified storage management that is consistent across all server and
customers
a storage platforms. With ASM, storage is managed as a small number of storage pools called
ASM Disk Groups. Database related files are assigned to ASM Disk Groups and ASM
manages the layout and data organization ensuring optimal performance and protection from
storage hardware failure. ASM Disk Groups provide a convenient and easy means for
consolidating storage and simplifying the administrative tasks previously required of DBAs
and System Administrators.
For additional details: http://www.oracle.com/technetwork/database/database-
technologies/cloud-storage/index.html
• Views
– Benefits
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
— Simplifies access to
Recovery Appliance
Metadata: Recovery
Appliance Catalog and
RMAN Recovery Catalog
— Easy integration with
a a
( cen
e l l o li
m b
e
ake
Internal Tasks
• Benefits
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a k
l o (a cens
el
OraclebScheduler li Scheduler) enables database administrators and application
(the
k e em
developers to control when and where various tasks take place in the database environment.
a The Scheduler provides complex enterprise scheduling functionality, which you can use to:
• Schedule job execution based on time or events
• Schedule job processing in a way that models your business requirements
• Manage and monitor jobs
• Execute and manage jobs in a clustered environment
Name Description
INDEX_BACKUP Indexes a backup piece into one or more Delta Pools (separate Delta Pool for each datafile)
BACKUP_ARCH Copies a raw archivelog received by RFS into an archivelog backup piece stored in a storage location
INCOMPLETE_ARCH Copies an incomplete archivelog read by RFS into a backup of archivelog in the storage location
PLAN_DF Creates a plan for purging or restoring a datafile from a Delta Pool
a ble
RESTORE Assists with constructing a backup piece from one or more Delta Pools
fe r
POLL Performs a scan of a polling directory
an s
BACKUP_SBT Copies a backup piece to a tape or downstream replication ZDLRA
n - t r
o
RESTORE_SBT
an
Restores a backup piece from a tape or downstream replication ZDLRA
s
RECONCILE
ha eฺ
Synchronizes the upstream/downstream catalog data, so the upstream can have access to the
)
om Guid
downstream backups
ฺ c
CROSSCHECK_DB
n k ng ent
Verifies that the catalog reflects the set of backups stored on tape and downstream replicants
VALIDATE
n b a
S t ud
Validates the backups stored for a database
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
Name Descriptions
PURGE_SB Removes backups that are beyond RECOVERY_WINDOW_SBT from tape storage
T
PURGE Asynchronously purges old backups from Delta Store to free space for new backups
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
Name Descriptions
DEFERRED_DELETE Removes backup files per user request
GCOPY_COMPUTE Computes the amount of data not backed up to tape for one Protected DB (used for
Guaranteed Copy)
DELETE_DB Removes a database from the ZDLRA (used for delete_db API)
a k
l o (a cens
b el li
m
a kee
Task Priorities
• Schedulers try to run tasks with lowest number (highest priority) first.
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
Tasks ordered by
an s
highest priority
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
RA_TASK View
• Contains a number of columns that are useful. Favorites are task_id, task_type, state,
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
-ฺc
om TASK_VIEW
job before
Use the RA G
id to understand whether
indexing is complete.
u
n k ng Indexing
e n tis complete.
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
b
Examples el li
k e e•m What tasks have been waiting the longest?
a select task_id, task_type, interrupt_count, systimestamp-
creation_time wait_time from ra_task where creation_time =
(select min(creation_time) from ra_task where archived='N') and
archived='N';
• What types of tasks currently need to be processed?
select task_type, count(*) from ra_task where archived=’N' group
by task_type order by task_type;
• Which tasks are being blocked by other tasks?
select t.task_id, t.task_type, b.task_id blocker, b.task_type
blocker_type from ra_task t join ra_task b on (t.waiting_on =
b.task_id) where t.archived = ’N‘ order by t.task_id;
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
Name Descriptions
a ble
RETRYING Session has failed, but task has not been cleaned up yet
fe r
an s
FAILED Task did not succeed, but is not eligible to be retried
n - t r
o
CANCEL
CANCELING an
Task is in the process of being run down
s
CANCELED ) ha eฺ
Some agent killed the task
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
Name Descriptions
ORDERING_WAIT An INDEX_BACKUP task is waiting for previous backups to be processed into the Delta Pool
TASK_WAIT Task has been suspended until a more important task completes or until a lock held by the
blocking task is released. The blocking task is identified by RA_TASK.WAITING_ON
SERVLET_WAIT Task is being blocked by an HTTP Servlet that is locking an object needed by the task
a ble
SPACE_WAIT Task is waiting for space to be released in the storage location
fe r
an s
OPT_DF_WAIT Task is an OPT_DF task that is waiting for space to be released in the storage location
n - t r
o
an
TEST_WAIT Task has been queued by the test_task mechanism
RESOURCE_WAIT
s
Task is in penalty box because undo or temp tablespace is temporarily out of space
ha eฺ
STALL_WHEN_WAIT Used by stall_when mechanism )
LIBRARY_WAIT
ฺ c om Guid
A PLAN_DF task is running for a BACKUP_SBT task whose library is stalled
POOL_FULL_WAIT
n k ng ent
The task cannot proceed because the RA_POOL tablespace is full
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
• Configuration: Overview
• Recovery Appliance Administrative Interfaces
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
l l o li
m be
e
ake
Databases
– More efficient disk usage
– Database backup deduplication and
compression
• Storage Locations
– DELTA Store (default)
— Usable Capacity: Base Rack 94TB - Full Rack 580TB
(32TB/Storage Node)
a ble
Normal Redundancy fe r
—
an s
– Additional Stores (non default)
n - t r
— Expansion and space allotment
a no
–
—
Backup Pooling
as
Storage Nodes same/different rack and/or Multirack
) h eฺ
External NFS Storage ฺ c om Guid
ngbackups t to DELTA Store based on
—
( a ak nse
Recovery e l lo
Appliance l i ce locations occupy space in Oracle ASM disk groups. By default,
storage
m b
k e e
the delta pool is stored in a normal redundancy Oracle ASM disk group called DELTA, which
a means that the Recovery Appliance maintains two copies of all on-disk backups. Database
backups can survive the loss of any one disk or storage server. The Recovery Appliance
metadata database, which tracks the files and blocks, is stored in a high redundancy Oracle
ASM disk group called CATALOG.
A more complex implementation may require multiple storage locations, for reasons such as
expansion and space allotment. For example Recovery Appliance with two storage locations:
DELTA (default) and DELTA2 (nondefault).
A Backup Polling Location is a file system directory where a protected database places
backups without interacting directly with Recovery Appliance. The backup polling directory is
an NFS mount point, and is not in a Recovery Appliance storage server. The polling policy
defines the file system path to the storage and how often it will be searched for new backups.
a k
l o (a cens
b el
Create Additional li
Storage Locations - DELTA Store
m
e create a Recovery Appliance storage location, follow these steps:
ake
To
1. From the Targets menu, choose Recovery Appliance. Enterprise Manager displays the
Recovery Appliance page. You may first see the Recovery Appliance Login page, where
you can enter the required credentials.
2. Click the Name of the Recovery Appliance for which you want to create a storage
location. The Recovery Appliance home page displays.
3. From the Recovery Appliance menu, choose Storage Locations. Enterprise Manager
displays the Storage Locations page.
4. Click Create. In the Create Storage Location dialog box, enter the Name of the new
storage location. Next, use the Search icon to select a Disk Group from the ASM Disk
Group dialog box and then click OK. Enter the amount of space in the Space field and
choose the unit of measure from the drop-down list.
5. Click OK to add the new storage location.
You could edit an existing DELTA Store to increase the storage location space but never to
decrease.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li
ea DELTA Store
Deletingb
k e em
To delete a storage location from a Recovery Appliance, follow these steps:
a 1. From the Targets menu, choose Recovery Appliance. Enterprise Manager displays the
Recovery Appliance page.
2. Click the Name of the Recovery Appliance for which you want to delete a storage
location. The Recovery Appliance home page displays.
3. From the Recovery Appliance menu, choose Storage Locations. Enterprise Manager
displays the Storage Locations page.
4. Select the Storage Location from the table that you want to delete and then click Delete.
5. Click OK to delete the storage location.
a k
l o (a cens
el
ExpandbRecovery li
Appliance Store
m
e steps
ake
Overall
• Preparing to Extend Recovery Appliance
- Estimating InfiniBand Cable Path Lengths: Ensure that you have the required
number and lengths of InfiniBand cables for your installation.
- Obtaining the Current Configuration Information: Use the current configuration
of the Recovery Appliance rack to plan patching requirements, configure new IP
addresses, and so on (image history, IP addresses, cell servers conf, compute
servers HugePages mem conf, net switches conf, OS users/groups, cluster status,
and patch info)
Version : 12.1.2.2.0.150917
Image activation date : 2015-11-28 13:55:30 -0700
Imaging mode : patch
Imaging status : success
a ble
fe r
[root@cell01 ~]#
ans
n - t r
opatch info sample: a no
•
h a s
m ) ideฺ
[oracle@host1 OPatch]$
g ฺ co t Gu
[oracle@host1 OPatch]$ ./opatch n
a kn den -oh /u01/app/12.1.0.2/grid
lsinventory
o nb versionS tu 12.1.0.1.9
Oracle Interim Patch Installer
u i h i s
n Corporation.
@
Copyright (c) 2016, Oracle
e t All rights reserved.
l o
el to u s
k b
a nse
( a
e l
Oracle lo Home lice : /u01/app/12.1.0.2/grid
b
m Central
e Inventory: /u01/app/oraInventory
ake from : /u01/app/12.1.0.2/grid/oraInst.loc
OPatch version : 12.1.0.1.9
OUI version : 12.1.0.2.0
Log file location: /u01/app/12.1.0.2/grid/cfgtoollogs/opatch/opatch2016-
07-28_01-14-54AM_1.log
-------------------------------------------------------------------------
-------
Local Machine Information::
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Hostname: host1.us.oracle.com
ARU platform id: 226
ARU platform description:: Linux x86-64
a ble
Oracle Grid Infrastructure 12c
fe r
12.1.0.2.0
ans
There are 1 products installed in this Oracle Home.
n - t r
a no
h a s
m ) ideฺ
Interim patches (5) :
g ฺ co t Gu
a kn Mayd25
nWed e n
Patch 22304421 : applied bon
n S t u 17:16:25 MDT 2016
Unique Patch ID: 19700676
u nio this
Created on 22 o
l l @2015,
Dec
u s e03:47:46 hrs PST8PDT
Bugs fixed:
a k be e to
l o (a cens
22304421
l patch
beThis li overlays patches:
e m
ake
21694919
This patch needs patches:
21694919
as prerequisites
…
-------------------------------------------------------------------------
-------
OPatch succeeded.
[oracle@host1 OPatch]$
- Generating the Network Configuration Files: Use the Oracle Exadata Deployment a ble
fe
Assistant to list the IP addresses for the new servers. Specify only the number of new r
ans
n - t r
servers. Define all other items, such as the Oracle home location and owner, the same as
a no
the existing rack configuration. In particular, the bonding configuration of the new servers
must match the existing servers in the rack. The Oracle Exadata Deployment Assistant
h a s
InfiniBand enables you to select the type of bonding. You must install the Recovery
m ) ideฺ
Appliance software in the new storage servers, after installing them in the rack.
ฺ o Gu
cdelete
- Moving the Audit and Diagnostic Files: Move
k g
n ent files in the
to
$GRID_HOME/rdbms/audit directory
b a n and the
u d$GRID_HOME/log/diagnostics
n t
Srecommends doing this task a day or two
u n io Oracle
directory before extending a cluster.
h i s
before the planned extension to allow
@ e t sufficient time.
- Synchronizing Release l o
el andtoPatch s
u Levels: The servers probably have a later release or
k b
a thenrack
patch levelathan
( e se where they will be added. You can either update the rack to the
b eallloservers,
laterlrelease
that
oric
l reimage the new servers to match the current rack. In either case, ensure
and all Sun Datacenter InfiniBand Switch 36 switches, are at the same
m
a kee patch level. Also consider how to handle the Grid Infrastructure and database home
releases and bundle patch updates.
Refer to the document: Zero Data Loss Recovery Appliance Owner's Guide for additional
information.
First two
Data LUNs only
Storage Grid
Partition Disk
System Area
a ble
e r
nstof
OR Cell OR Visible
Disk
- tr aASM
no n
a Disk
sGrid
LUN h a
) ideฺ (hot part)
Other m
co t GuGrid Disk
LUNs
g
n enฺ (cold part)
n k
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Each Exadata
b i
el High lCapacity Storage Server contains 12 hard disk drives. Exadata cell
OR
FLASHLOG
a ble
fe r
Flash
a n s
Flash Cell Cache -t r
LUN Disk n on
a
s Flash
) h a ฺ Log
om Guid e
ฺ c
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
el X6-2 lHigh
Each Exadata
b i Capacity Storage Server contains 12.8 TB of high performance flash
k e em distributed across four PCI flash memory cards. Therefore, each flash card has a
memory
a capacity of 3.2 TB.
Essentially, each flash device is like a physical disk in the storage hierarchy. Each flash
device is visible to the Exadata cell software as a LUN and the initial cell configuration
process creates flash-based cell disks on all the flash devices.
By default, nearly all the available space on each flash-based cell disk is allocated to Exadata
Smart Flash Cache and only a small portion (128 MB on each flash-based cell disk, 512 MB
in total) is allocated to Exadata Smart Flash Log.
a ble
fe r
Exadata Cell CELL1 Exadata Cell CELL1 Exadata Cell
a n s
CELL1
n-t r
Failure Group
DELTA_2
a no
CELL1
Disk Group
CELL2 Failure Group
h a s CELL3 Failure Group
DELTA_3
m ฺ Failure Group
) ideCELL3
CELL1 Failure Group
Disk Group
CELL2 Failure Group
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
After grid l l o li
edisks are configured, ASM disk groups can be defined to consume Recovery
b
em storage. The slide illustrates an example where two ASM disk groups are defined.
eAppliance
ak The DELTA_1 disk group is defined across all of the red grid disks, and the DELTA_2 disk
group is defined across the blue grid disks. When data is loaded into each disk group, ASM
evenly distributes the data across all of the grid disks in each disk group.
Using NORMAL or HIGH ASM redundancy in conjunction with at least 3 failure groups for
each disk group is recommended on Recovery Appliance. This ensures that at least two
copies of data are maintained to protect from storage failure. EXTERNAL redundancy disk
groups are not supported because they provide no protection from storage failures and
prohibit the use of online storage maintenance procedures, such as rolling patches.
Note that the initial configuration process for Recovery Appliance creates 2 disk groups using
NORMAL(DELTA) or HIGH(CATALOG) ASM redundancy.
To protect against the failure of an entire Exadata cell, separate ASM failure groups are
automatically associated with each cell, which ensures that mirrored ASM extents are placed
on different Exadata cells. This is also illustrated in the slide. By default, when failure groups
are automatically created, their names correspond to the cell name. So, different disk groups
can have the same failure group names. For example, the DELTA_1 disk group has a failure
group named CELL1, and the DELTA_2 disk group also has a failure group named CELL1.
a no
[celladmin@zdlra1cel01 ~]$
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Inside b e l l
Recovery
o li
Appliance, each Exadata Storage Server runs independently from all the
k e em Exadata Storage Servers. In line with that autonomy, each cell is administered
other
a individually. Most administration functions are performed using the Exadata cell command-
line interface (CellCLI). CellCLI can be used only from within a cell to manage that cell.
However, you can run the same CellCLI command remotely on multiple cells with the dcli
utility, which is described later in this lesson.
CellCLI works in conjunction with the Exadata Storage Server Management Server. CellCLI
provides the command interface while MS performs the administrative functions, such as
creating and dropping grid disks.
Three operating system users are configured for each Exadata Storage
Server:
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a kb se t
l o ( cen
Three b e l
operating li users are configured for each Exadata Storage Server: root,
system
k e em
celladmin, and cellmonitor. The slide describes the function of each user account.
a After Exadata is initially configured, the operating system user accounts are set with initial
passwords. The default initial password for root is welcome1. The default initial password
for the cellmonitor and celladmin users is welcome. It is recommended that the initial
passwords for all the user accounts should be changed to more secure passwords after initial
configuration is completed.
fe r
$ exacli -c celladministrator@zdlra1cel01
an s
Password=************
n - t r
no
ExaCLI>
s a
$ exacli -c celladministrator@zdlra1cel01 --cookie-jar -e list cell
Password=************
zdlra1cel01 h a
) ideฺ
online
$ m
co t Gu
g ฺ
a
CD_00_zdlra1cel01 kn den
$ exacli -l celladministrator -c zdlra1cel01 -e list celldisk
n normal
...
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Exadata l l o li 12.1.2.2.0 introduces the ExaCLI utility. ExaCLI is essentially the
esoftware release
b
k e em as CellCLI; however, the main difference is that ExaCLI manages cells from a remote
same
a host, typically an Exadata database server, whereas CellCLI runs directly on a cell.
ExaCLI enables users to perform most management functions without the requirement to
establish an SSH connection to the cell. Consequently, access to the cell operating system
can be tightly controlled to address various operational and compliance requirements.
To use ExaCLI, non-default users must be created on the cell, and the users must be
assigned roles that grant appropriate privileges. Creation of users and roles can only be
performed directly on the cell using CellCLI.
All communication between ExaCLI and Management Server (MS) running on the cell is over
https. Security certificates allow the cells to confirm their identity to ExaCLI. MS is deployed
with a default self-signed security certificate, or you can upload a different security certificate
issued by a Certificate Authority (CA).
ExaCli supports the same command syntax as CellCLI. However, not all CellCLI commands
can be executed through ExaCLI. Unsupported commands include:
• ALTER CELL with the RESTART, STARTUP or SHUTDOWN options
• CREATE USER, ALTER USER, DROP USER
• CALIBRATE
a ble
$ dcli –g mycells date
fe r
exa1cel01: Sun Oct 30 20:48:09 CDT 2015
an s
exa1cel02: Sun Oct 30 20:48:09 CDT 2015
n - t r
no
$ dcli –c exa1cel01,exa1cel02 cellcli –e list cell
a
exa1cel01: exa1cel01
exa1cel02: exa1cel02
online
online
h a s
Examples
m ) ideฺ
co t Gu
$ dcli –g mycells –x cellclicommands.scl
g ฺ
a n kn den
$ dcli –g mydbservers –l root –x dbwork.sh
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li centralized management across Recovery Appliance by
e utility facilitates
The dcli
b
em
eautomating the execution of a command on a set of servers and returning the output to the
ak centralized management location where dcli was run. The types of commands supported by
dcli includes operating system commands, CellCLI commands, operating system scripts,
and CellCLI scripts. However, it does not support interactive sessions.
The dcli utility runs commands on multiple servers in parallel threads. By default, the dcli
utility is located at /opt/oracle/cell/cellsrv/bin/dcli on each Exadata Storage
server and at /usr/local/bin/dcli on each database server. You can also copy the
dcli utility to a server outside of Recovery Appliance and launch commands from that
server.
The dcli utility requires Python version 2.3 or later on the server running dcli. You can
determine the version of Python by running the python -V command. In addition, dcli
requires prior setup of SSH user-equivalence between all the servers. You can use the dcli
utility initially with the -k option to set up SSH user-equivalence between a group of servers.
Command output (to stdout and stderr) is collected and displayed after command execution is
finished on all the specified servers. The dcli options allow command output to be
abbreviated to filter output, such as removing messages showing normal status.
• Command types:
– ExaCLI commands
– ExaCLI scripts
• Commands are executed in separate parallel threads using ExaCLI:
– Requires access through a user with role-based privileges
– Uses the same cookie jar mechanism as ExaCLI
• Interactive sessions are not supported.
• Command output is collected and displayed in the terminal session executing the exadcli utility.
a ble
$ exadcli –c exa1cel01,exa1cel02 -l celladministrator
fe r
--cookie-jar –e list cell
an s
r
Password=************
exa1cel01: exa1cel01
exa1cel02: exa1cel02
online
online
n - t
$ cat mycells
a no
exacel01
exacel02
h a s
Examples $ cat cmd.txt
m ) ideฺ
co t Gu
list cell
g ฺ
kn den
$ exadcli -g mycells -l celladministrator -x cmd.txt
exa1cel01: exa1cel01 online
a n
exa1cel02: exa1cel02 online
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Exadata l l o li 12.1.2.2.0 introduces the exadcli utility. Like dcli, exadcli
esoftware release
b
k e em centralized management across Exadata by automating the execution of a
facilitates
a command on a set of cells and returning the output to the centralized management location
where exacli was run.
The main difference is that dcli uses SSH for communication, while exadcli uses https in
the same way as ExaCLI. By removing the requirement for SSH, exadcli enables tight
control over access to the cell operating system, which is useful in addressing various
operational and compliance requirements.
Another difference is that exadcli can only be used to issue ExaCLI commands to be run on
multiple cells. Unlike dcli, exadcli cannot be used to execute other commands, such as
shell commands.
To use the exadcli utility, you must set up non-default users and roles on all the cells in the
same way as required by ExaCLI. Also, exadcli uses the same cookie jar mechanisms as
ExaCLI.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
Need large
Need high RPM and
I/O channel
fast seek time
a ble
fe r
an s
OLTP Analytics n - t r
(Small random I/O) no
(Large sequential I/O)
a
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
nb s Stu
CALIBRATE
n i o i
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
e l
The CALIBRATE
b li
command runs raw performance tests on Exadata Storage Server disks and
emmodules. This enables the measurement of two important metrics – IOPS and MBPS:
eflash
ak • IOPS (I/O per second): This metric represents the number of small random I/O that can
be serviced in a second. The IOPS rate typically depends on how fast the disk media
can spin and how many disks are present in the storage system.
• MBPS (megabytes per second): This metric represents the rate at which data can be
transferred between the server and storage array. This mainly depends on the capacity
of the I/O channel that is used to transfer the data.
Aggregate random read throughput across all hard disk luns: 1604 MBPS
Aggregate random read throughput across all flash disk luns: 4242.9 MBPS
Aggregate random read IOs per second (IOPS) across all hard disk luns: 4927
Aggregate random read IOs per second (IOPS) across all flash disk luns:
148695 Controller read throughput: 1608.05 MBPS
Calibrating hard disks (read only) ...
Lun 0_0 on drive [20:0 ] random read throughput: 153.41 MBPS, and 412 IOPS
a ble
Lun 0_1 on drive [20:1 ] random read throughput: 155.38 MBPS, and 407 IOPS
fe r
ans
...
n - t r
no
Lun 0_8 on drive [20:8 ] random read throughput: 154.46 MBPS, and 424 IOPS
a
a s
Lun 0_9 on drive [20:9 ] random read throughput: 154.63 MBPS, and 426 IOPS
h
) ideฺ
Calibrating flash disks (read only, note that writes will be significantly
m
slower)
ฺ o Gu
cthroughput:
Lun 1_0 on drive [[10:0:0:0]] random read g
kn read n t 269.11 MBPS, and 19635
n
IOPS Lun 1_1 on drive [[10:0:1:0]] arandom
u d e
n b t throughput: 268.86 MBPS, and
19648 IOPS
u n io his S
...
@ e t
l o s
el to u random read throughput: 268.33 MBPS, and 19717
k b
Lun 5_2 on drive [[11:0:2:0]]
IOPS Lun 5_3 ona drive s e
( a enresults
19693 IOPSoCALIBRATE
[[11:0:3:0]] random read throughput: 268.14 MBPS, and
e m
eCALIBRATE
a k stress test is now running...
Calibration has finished.
a a
( cen
l l o
e you may
Occasionally, li need to start, stop, or restart the Exadata Storage Server software
b
k e embe achieved
services:
can
Restart Server (RS), Management Server (MS), and Cell Server (CELLSRV). This
by using the CellCLI ALTER CELL command. An example is shown on the
a
slide. Following is a summary of the relevant syntax:
ALTER CELL { SHUTDOWN | RESTART | STARTUP } SERVICES { RS | MS |
CELLSRV | ALL}
CellCLI>
s a
h a
) ideฺ
m
co t Gu
g ฺ
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li in an Exadata Storage Server can be used, cell disks must be created
e disk storage
Beforebthe
k e em
by using the CREATE CELLDISK command.
a Note that in most cases, the initial configuration process, performed by using the Oracle
Exadata Deployment Assistant, will configure Exadata Storage Server cell disks and grid
disks.
The example in the slide shows the use of the CREATE CELLDISK ALL HARDDISK
command to create 12 disk-based cell disks with default names. In most cases, the default
cell disk names are used.
The preceding example also shows the use of the LIST CELLDISK command. For a High
Capacity Exadata Storage Server, output from the LIST CELLDISK command shows the 12
disk-based cell disks along with 16 flash-based cell disks that are normally used in
conjunction with Exadata Smart Flash Cache. The command should return a status of normal
for all the cell disks.
a a
( cen
After cell l l o li
edisks are created, grid disks must be provisioned using the CREATE GRIDDISK
b
em Like cell disks, grid disks are typically created as part of the initial configuration of
ecommand.
ak Recovery Appliance. However, there are times when existing grid disks may be removed and
replaced by new ones, such as when the disk groups are resized. This case is considered in
further detail later in the lesson.
The slide shows examples of how to create grid disks on empty new cell disks. In the example
in the slide, the ALL PREFIX option is used to automatically create one grid disk on each
cell disk. When the ALL PREFIX option is used, the generated grid disk names are
composed of the grid disk prefix followed by an underscore (_) and then the cell disk name.
It is best practice to use the planned ASM disk group name as the prefix name for the
corresponding grid disks. In the example, prefix values delta_1 and delta_2 are the
names of planned ASM disk groups that will consume the grid disks. Grid disk names must be
unique across all cells within a single Recovery Appliance deployment. By following the
recommended naming conventions for naming the cell disks and grid disks, you automatically
get unique names.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
$ cat /etc/oracle/cell/network-config/cellinit.ora
a ble
ipaddress1=192.168.10.1/24
fe r
ipaddress2=192.168.10.2/24
an s
$ cat /etc/oracle/cell/network-config/cellip.ora
n - t r
no
cell="192.168.10.5; 192.168.10.6"
cell="192.168.10.7; 192.168.10.8"
cell="192.168.10.9; 192.168.10.10"
s a
h a
) ideฺ
m
co t Gu
g ฺ
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
eof the initialli configuration for Recovery Appliance, two important configuration files
As a part l
b
emcreated on each database server, which enable it to access Exadata Storage Servers:
eare
ak • The cellinit.ora file contains the database server IP address that connects to the
storage network. This file is host specific, and contains the IP addresses of the
InfiniBand storage network interfaces for that database server. The IP addresses are
specified in Classless Inter-Domain Routing (CIDR) format.
• The cellip.ora file contains the IP addresses for the InfiniBand storage network
interfaces of the Exadata Storage Servers that are accessible to the database server.
If there are ever any issues with connectivity between a database server and Exadata
Storage Servers, then verify these files to ensure that the correct settings are present. If
changes are ever required to these files, then:
1. Stop the database and the Oracle ASM instances on the database server host.
2. Make the required file changes.
3. Restart the database and the Oracle ASM instances on the database server host.
a a
( cen
l l o
e appropriate
At all times, li versions of the various Exadata software components must be
b
em
emaintained in order to ensure correct operation. This includes the version of Oracle Database
ak software used for the ASM and database instances. Refer to My Oracle Support note Zero
Data Loss Recovery Appliance Supported Versions (Doc ID 1927416.1) for an up-to-date list
of the supported versions for the Exadata software components.
To ensure that ASM discovers Exadata grid disks, set the ASM_DISKSTRING initialization
parameter. A search string with the following form is used to discover Exadata grid disks:
o/<cell IP address>/<grid disk name>
Wildcards may be used to expand the search string. For example, to explicitly discover all the
available Exadata grid disks set ASM_DISKSTRING='o/*/*'. To discover a subset of
available grid disks having names that begin with data, set
ASM_DISKSTRING='o/*/data*'.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a k
l o (a cens
bel li
em
ake
a k
l o (a cens
bel li
em
ake
Objectives
a a
( cen
l l o li
m be
e
ake
Databases
• Enable Real-Time Redo Transport in a Protected Database
• Enroll Protected Databases
• Take a Backup of a Protected Database
• Recover a Protected Database a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
o
– Backup and Recovery operations
s an
• Maintain separation of duties between:
) ha eฺ
PDB
– Different application administrators
– Application administrators and DBA ฺ c om Guid PDBA PDBB PDBC
k n g nCDB t CRAC
– Users within application
a n ud e
• Provides isolation
n b S t
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Non-CDB
b li
elArchitecture
eIn m 11g database, the only kind of database that is supported is a non-CDB. The old
eOracle
a k architecture is referred to as the non-CDB architecture. The term non-CDB will be used as a
shorthand for an occurrence of a pre-12.1 database that uses the pre-12.1 architecture. This
database requires its own instance and therefore its own background processes, memory
allocation for the SGA. It needs to store the Oracle metadata in its data dictionary. The
database administrator can still create Oracle 12c non-CDBs with the same pre-12.1
architecture. These databases are not multitenant container databases or non-CDBs.
CDB Architecture
Consolidating many non-CDB databases onto a single platform reduces instance overhead,
avoids redundant copies of data dictionaries, and consequently storage allocation. This
benefits from fast provisioning, time-saving upgrading, better security through the separation
of duties and application isolation. The new 12c database that consolidates databases
together is a multitenant container database (CDB), and a database consolidated within a
CDB, a pluggable database (PDB).
Database occurrence. You can also upgrade a single PDB by unplugging it and
plugging it into a CDB at a different Oracle Database version.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
( a ak nse
e
This slide l lo
describes l i
the
cehigh-level tasks in using the Recovery Appliance to store and
m b
ak ee
manage backups for multiple protected databases in the enterprise. These tasks can be
performed using Cloud Control or RMAN. Depending on the management interface used,
there may be minor variations in the steps to perform certain tasks.
Multiple interfaces
• Oracle Enterprise Manager Cloud Control
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
• RMAN Client
• SQL*Plus or SQL Developer
Oracle Databases
Cloud Control
Tape Library
a ble
RMAN SQL*Plus
fe r
an s
n - t r
a no
h a s
) ideฺ
Recovery Appliance
m
g ฺ co t Gu
SuperCluster Exadata a n kn den Recovery Appliance
n b S t u Replica
n i o i s
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
l
e Appliance
The Recovery
b li provides multiple interfaces to manage backup and recovery
em for protected databases.
eoperations
ak • Oracle Enterprise Manager Cloud Control (Cloud Control): Cloud Control provides a
GUI for administering, managing, and monitoring a Recovery Appliance environment. It
also enables you to configure, back up, and recover protected databases.
• RMAN client: The Recovery Appliance is integrated with RMAN and you can use the
RMAN client installed on your protected database to configure, back up, and recover
protected databases.
• SQL*Plus: SQL*Plus is a command-line tool that you can use to query the Recovery
Appliance catalog and run the DBMS_RA PL/SQL package.
12.1.0.2
Features Available for Oracle Database Releases 10.2.0.5 11.2.0.3 11.2.0.4
and later
h a sa
Cross-endian backup polling
m ) idexฺ x
o
ฺc t G u
n g
k den
a n
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
e databaseli can be Standard or Enterprise Edition on any storage or volume
l
A protected
b
m (such as filesystem, ASM, ACFS, ZFS) and use any of the following: Oracle Real
kee
manager
a Application Clusters, Oracle Data Guard, Oracle Streams and/or Oracle GoldenGate. Note
that polling locations are only supported on NFS-based locations, not ASM.
Note
1. Virtual full backups are not supported for RMAN encrypted backup sets.
2. Real-time redo is available for 11.2.0.3, but not as an option to be configured using
Enterprise Manager, so you need to do it manually. Enterprise Manager provides the
option to configure real-time redo in versions 11.2.0.4+.
12.1.0.2
Features Available for Oracle Database Releases 10.2.0.5 11.2.0.3 11.2.0.4
and later
a k
l o (a cens
Note bel li
e m
3. The Recovery Appliance itself cannot serve as a cascader that transfers received redo
ake to a standby database.
For additional information about features supported by version please review the following
support note: Zero Data Loss Recovery Appliance Features Available per Oracle Database
Release (Doc ID 1995866.1)
Also check the following support note for recommended database patches: Recommended
Protected Database Patches for Zero Data Loss Recovery Appliance (Doc ID 1982542.1)
Databases
• Enable Real-Time Redo Transport in a Protected Database
• Enroll Protected Databases
• Take a Backup of a Protected Database
• Recover a Protected Database a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a a
( cen
The Zero l l o i
eData LosslRecovery Appliance Backup Module (Recovery Appliance Backup
b
em is an Oracle-supplied SBT library that RMAN uses to transfer backup data over the
eModule)
ak network to the Recovery Appliance. An SBT library transfers data to and from a backup
device type, either a tape device or the Recovery Appliance. RMAN performs all backups to
the Recovery Appliance, and all restores of complete backup sets, by means of this module.
The Recovery Appliance Backup Module must be installed in the following locations:
• In the Oracle home of every protected database that sends backups to a Recovery
Appliance. For example, a single host might have an Oracle Database 11g Oracle
home, and an Oracle Database 12c Oracle home. Each Oracle home might support five
protected databases, for a total of ten databases running on the host. In this case, only
two Recovery Appliance Backup Modules must be installed: one in each Oracle home.
• For Recovery Appliance replication environments, on every upstream Recovery
Appliance that sends backups to downstream Recovery Appliances
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• Prerequisites
– ra_install.jar file
– Java 1.5 or later
• Input parameters
– ZDLRA Username and Password (RASYS or other)
– ZDLRA Host, Port and ServiceName (SCAN)
• Output parameters
– Wallet File Location: $ORACLE_HOME/dbs/ra_wallet (default) a ble
– Config File Location: $ORACLE_HOME/dbs/ra<ORACLE_SID>.ora (default) fe r
an s
– Lib File Location: $ORACLE_HOME/lib (default)
n - t r
a
$ java -jar ra_install.jar -dbUser rasys -dbPass ra -serviceName zdlra4 no -port 1521 -host zdlra-scan5 \
h a s
-walletDir $ORACLE_HOME/dbs/ra_wallet -libDir $ORACLE_HOME/lib/ -proxyHost www-proxy.demo.com
g
Recovery Appliance initialization file /u01/app/oracle/product/11.2.0.3/dbhome_1/dbs/raLUZ11.ora created.
a n kn den
Downloading Recovery Appliance Software Library from file ra_solaris_sparc64.zip.
Downloaded 28540438 bytes in 44 seconds. Transfer rate was 648646 bytes/second.
nb s Stu
Download complete.
n i o i
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
e installs theli credentials and software that are required to use the ZDLRA library.
This utility
b l
em output for arguments:
eSample
ak host:ra_installer user$ java -jar ra_install.jar
Recovery Appliance Install Tool, build 2015-11-03
No arguments supplied
Usage: java -jar ra_install.jar
-dbUser: Database Username
-dbPass: Database Password
-host: SCAN hostname of ZDLRA Server
-port: ZDLRA Listening port
-servicename: ZDLRA service name of database
-walletDir: Directory to store wallet
-configFile: File name of config file
-libDir: Directory to store library
-libPlatform: Platform of library to download Recovery
Appliance Install Tool.
…
Zero Data Loss Recovery Appliance Overview 6 - 13
The file downloaded from OTN is called “ra_installer.zip” and includes two files:
“ra_install.jar” and “ra_readme.txt”.
Review the ra_readme.txt file for additional information.
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
backup module.
• Can be created manually when missing or damaged
– Use the mkstore tool
– Parameters:
— Wallet location
— credential name
— VPC User [oracle@host1 ~]$ mkdir -p $HOME/oracle/wallet
[oracle@host1 ~]$ mkstore -wrl $HOME/oracle/wallet \
> -createALO -createCredential crmdb vpccrmuser
Oracle Secret Store Tool: Version 11.2.0.4.0 - Production
a ble
Copyright (c) 2004, 2013, Oracle and/or its affiliates. All rights reserved.
fe r
Your secret/Password is missing in the command line
an s
Enter your secret/Password:
Re-enter your secret/Password:
n - t r
a no
Create credential oracle.security.client.connect_string1
[oracle@host1 ~]$
h a s
) ideฺ
[oracle@host1 ~]$ mkstore -wrl $HOME/oracle/wallet -listCredential
m
co t Gu
Oracle Secret Store Tool: Version 11.2.0.4.0 - Production
ฺ
Copyright (c) 2004, 2013, Oracle and/or its affiliates. All rights reserved.
g
a n kn den
List credential (index: connect_string username)
1: crmdb vpccrmuser
nb s Stu
[oracle@host1 ~]$
n i o i
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
l li the credentials of the Recovery Appliance user that will be used by
e wallet stores
An Oracle
b
emprotected database to authenticate with the Recovery Appliance. These same credentials
ethe
ak are used for sending backups and redo, if configured. When you install the Recovery
Appliance backup module, an Oracle wallet is automatically created. You can also create the
wallet and add required entries manually.
Sample output for arguments:
[oracle@host ~]$ mkstore
Oracle Secret Store Tool: Version 11.2.0.4.0 - Production
Copyright (c) 2004, 2013, Oracle and/or its affiliates. All rights
reserved.
mkstore [-wrl wrl] [-create] [-createSSO] [-createLSSO] [-createALO]
[-delete] [-deleteSSO] [-list] [-createEntry alias secret] [-
viewEntry alias] [-modifyEntry alias secret] [-deleteEntry alias] [-
createCredential connect_string username password] [-listCredential]
[-modifyCredential connect_string username password] [-
deleteCredential connect_string] [-help] [-nologo]
[oracle@host ~]$
Databases
• Enable Real-Time Redo Transport in a Protected Database
• Enroll Protected Databases
• Take a Backup of a Protected Database
• Recover a Protected Database a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
n - tr
Real-Time no
Redo Transport
h a sa
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li - Oracle Enterprise Manager
e Redo setting
Real-Time
b
k e emeach Database in the Backup Settings:
For
a 1. Select the Recovery Appliance that you are protecting.
2. Ensure you select the Correct VPC user.
3. Select Enable Real-Time Redo Transport. This is an OEM procedure. You can view its
progress in OEM.
Ensure that you take a backup after completing the REDO Transport operation to ensure all
Logs are collected.
During the load, the Recovery Appliance tries to maintain a lag of less than a second for the
unprotected data window.
Remember that when you configure real-time redo transport, redo data from the protected
database is directly transported and stored on the Recovery Appliance. This reduces the
window of potential data loss that exists between successive archived log backups.
Archived Log
Destination points to
the Recovery Appliance
a ble
fe r
an s
Redo Transport User
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Real-Time
b li - CLI
el Redo Setting
k e emeach database, execute the following steps:
For
a 1. Ensure that the Recovery Appliance user that the protected database uses to send
backups to the Recovery Appliance is configured. This same user will be used for redo
transport.
- You have learned how to create this user in the previous lesson. In this lesson you
will learn how to assign the privileges (enrolling) needed to send backups and for
redo transport.
- Also ensure that an Oracle wallet is created on the protected database that
contains credentials for the Recovery Appliance (and redo transport) user. This
wallet is created automatically when you installed the Recovery Appliance Backup
Module in the Protected Database’s Oracle Home directory.
2. Ensure that the following conditions are met for the protected database:
- ARCHIVELOG mode is enabled
- DB_UNIQUE_NAME parameter is set
Databases
• Enable Real-Time Redo Transport in a Protected Database
• Enroll Protected Databases
• Take a Backup of a Protected Database
• Recover a Protected Database a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
a no
h a s
m ฺ
) ide•Reserved
ฺ c o Gu Space
u nio this S
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Enrolling
b l
eProtected li
Databases with the Recovery Appliance - Cloud Control
Appliance using Cloud Control, the Recovery Appliance allocates a default reserved
space of 2.5x the database size. You can accept or change this amount.
k) In the Recovery Appliance User section, enter the credentials for the appropriate
virtual private catalog account.
l) In the Credential Access Grantee section, in Enterprise Manager Users, select the
Enterprise Manager user accounts that need access to the Recovery Appliance user
credentials.
a ble
m) Click OK.
fe r
n) Click Close to return to the Protected Databases page. ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
a k
l o (a cens
Enrolling
b ethe li Databases with Recovery Appliance - CLI
l Protected
k e emeach Database execute the following steps:
For
a 1. Install the Recovery Appliance Backup Module. Follow the steps explained previously in
this lesson.
2. Only for Oracle Database 10g: Enrolling these protected databases requires alternate
manual configuration steps for the first part of the enrolment process.
a. Add a connect descriptor for the Recovery Appliance to the tnsnames.ora file.
Sample:
ZDLRA=
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = scaz15ingest-scan1)(PORT = 1521))
(CONNECT_DATA =
(SERVER = DEDICATED)
(SERVICE_NAME = zdlra9)
)
)
the wallet.
For example :
$ /u01/app/oracle/product/10.2.0/db_1/bin/mkstore -wrl
/u01/app/oracle/product/10.2.0/db_1/dbs/ra_wallet/ -
createCredential "zdlra9" "rauser10" "welcome1"
e. Ensure that the sqlnet.ora file contains the location of the Oracle wallet.
For example : a ble
fe r
$ cat /u01/app/oracle/product/10.2.0/db_1/network/admin/sqlnet.ora
ans
# sqlnet.ora Network Configuration File:
n - t r
a no
/u01/app/oracle/product/10.2.0/db_1/network/admin/sqlnet.ora
h a s
# Generated by Oracle configuration tools.
m ) ideฺ
ฺco t Gu
NAMES.DIRECTORY_PATH= (TNSNAMES, EZCONNECT)
SQLNET.WALLET_OVERRIDE = true ng
a n k den
WALLET_LOCATION =
i o nb s Stu
(SOURCE =
u n t h i
(METHODll=o@
FILE) s
u e
k b e t o
( a a nse
(METHOD_DATA =
b lice
o (DIRECTORY
ell/u01/app/oracle/product/10.2.0/db_1/dbs/ra_wallet)
=
ee m )
ak )
f. Copy the libra.so file from the ORACLE_HOME/lib directory of the Recovery
Appliance to the ORACLE_HOME/lib directory of the protected database.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
a k
l o (a cens
Configuring
b li Databases Backup Settings - Cloud Control
el the Protected
k e emeach Database execute the following steps:
For
a 1. Access the home page for the protected database.
2. From the Availability menu, select Backup & Recovery, and then select Backup
Settings. The Backup Settings page for the protected database is displayed.
3. In the Device tab, configure the following optional settings:
a) To write redo data asynchronously from the protected database to the Recovery
Appliance, in the Recovery Appliance Settings section, select Enable Real-Time
Redo Transport.
b) If you want to configure backup polling for the protected database, then specify the
polling location in the Disk Backup Location setting of the Disk Settings section.
4. In the Maximum Backup Piece (File) Size field of the Backup Set tab, specify the
maximum size of each backup set that RMAN creates.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Configuring
b li Database Settings - CLI
el the Protected
k e emeach Database execute the following steps:
For
a 1. Use RMAN to connect to the protected database as TARGET.
Sample local connection:
rman target /
2. Use the CONFIGURE command to configure the required backup settings.
Sample command:
CONFIGURE CHANNEL DEVICE TYPE 'SBT_TAPE'
PARMS
'SBT_LIBRARY=/u01/app/oracle/product/11.2.0.4.0/dbhome_1/lib/libra.s
o,
ENV=(RA_WALLET=location=file:/u01/app/oracle/product/11.2.0.4.0/dbho
me_1/dbs/zdlra
credential_alias=ra-scan:1521/zdlra5:dedicated)' FORMAT '%U_%d';
3. (Optional) Configure Real-Time Redo Transport. Follow the steps described in previous
slides in this lesson.
Databases
• Enable Real-Time Redo Transport in a Protected Database
• Enroll Protected Databases
• Take a Backup of a Protected Database
• Recover a Protected Database a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
Oracle Suggested
Backup
a a
( cen
After you l l o li protected database, you can create and schedule protected database
econfigure the
b
em The Recovery Appliance uses the incremental-forever backup strategy for protected
ebackups.
ak database backups. In this strategy, an initial level 0 incremental backup is followed by
successive level 1 incremental backups. Except when using real-time redo transport, include
archived redo log files in all backups to ensure that you can perform complete recovery for the
protected database.
If the protected database is running in NOARCHIVELOG mode, then you must perform
consistent backups which requires shutting down the protected database.
Suggested Backup.
5. Specify the method used to back up redo data.
- If real-time redo transport is not configured for the protected database, select Also
back up all archived logs on disk.
- If required, select Delete all archive logs from disk after they are successfully
backed up.
6. Click Next to display the Schedule Oracle-Suggested Recovery Appliance Backup: a ble
fe r
Schedule page.
ans
7. Schedule the backup job.
n - t r
o
an
8. Click Next to display the Schedule Oracle-Suggested Recovery Appliance Backup:
s
Review page.
) ha eฺ
9. Click Submit Job.
ฺ c om Guid
10. Click View Job to display the status of the
k g job.
nbackup n t
a n ud e
n b S t
u nio this
l l o @ use
a k be e to
l o (a cens
b el li
m
a kee
Recovery Appliance
Backup Module
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
After you l l o li protected database, you can create and schedule protected database
econfigure the
b
em The Recovery Appliance uses the incremental-forever backup strategy for protected
ebackups.
ak database backups. In this strategy, an initial level 0 incremental backup is followed by
successive level 1 incremental backups. Except when using real-time redo transport, include
archived redo log files in all backups to ensure that you can perform complete recovery for the
protected database.
If the protected database is running in NOARCHIVELOG mode, then you must perform
consistent backups which requires shutting down the protected database.
Backing Up the Protected Databases - CLI
To create a full backup of the whole protected database:
1. Connect RMAN to the protected database as TARGET and the Recovery Appliance
catalog as CATALOG.
2. Ensure that the configuration steps for the Enrolling and Configuring tasks are
completed.
{
ALLOCATE CHANNEL c1 DEVICE TYPE sbt_tape
PARMS='SBT_LIBRARY=/u01/oracle/product/11.2.0.4/dbhome_1/lib/libra.so,
ENV=(RA_WALLET=location=file:/u01/oracle/product/11.2.0.4/dbhome_1/dbs/ra
credential_alias=ra-scan:1521/zdlra5:dedicated)'
FORMAT'%U_%d';
ALLOCATE CHANNEL c2 DEVICE TYPE sbt_tape
a b le
PARMS='SBT_LIBRARY=/u01/oracle/product/11.2.0.4/dbhome_1/lib/libra.so, fe r
n s
n - tra
ENV=(RA_WALLET=location=file:/u01/oracle/product/11.2.0.4/dbhome_1/dbs/ra
credential_alias=ra-scan:1521/zdlra5:dedicated)'
a no
FORMAT'%U_%d';
h a s
ALLOCATE CHANNEL c3 DEVICE TYPE sbt_tape
m ) ideฺ
g ฺ co t Gu
PARMS='SBT_LIBRARY=/u01/oracle/product/11.2.0.4/dbhome_1/lib/libra.so,
a n kn den
ENV=(RA_WALLET=location=file:/u01/oracle/product/11.2.0.4/dbhome_1/dbs/ra
i o nb s Stu
credential_alias=ra-scan:1521/zdlra5:dedicated)'
u n t h i
FORMAT'%U_%d';
l lo@ o us e
BACKUP DEVICE TYPEesbt
a a kb se t
l o ( cen
TAG 'db_full_incr'
b e l
CUMULATIVE li
INCREMENTAL LEVEL 1
m
kee
DATABASE FORMAT '%d_%U'
a PLUS ARCHIVELOG FORMAT '%d_%U' NOT BACKED UP;
}
4. Open a text editor and create and save a file with the following contents.
Sample:
export ORACLE_HOME=/u01/app/oracle/product/11.2.0.4/dbhome_1
export ORACLE_SID=db1124sm
export PATH=$PATH:$HOME/bin:$ORACLE_HOME/bin:$ORACLE_HOME/Opatch
export LD_LIBRARY_PATH=$ORACLE_HOME/lib:$ORACLE_HOME/rdbms/lib:/lib:/usr/lib;
a ble
export LD_LIBRARY_PATH
fe r
export LOG_TRACE_DIR=$HOME/RA_TEST/RMAN_SCRIPTS/LOG
ans
dt=`date +%y%m%d%H%M%S`
n - t r
a no
$ORACLE_HOME/bin/rman log=$LOG_TRACE_DIR/rman_bkincr_log_db1124sm_$dt.log
<<EOF
h a s
m ) ideฺ
co t Gu
CONNECT TARGET /
g ฺ
a n kn den
CONNECT CATALOG rauser/welcome1@ra-scan:1521/zdlra5:dedicated
i o nb s Stu
un e thi
BACKUP DEVICE TYPE SBT
TAG 'db_full_incr' @
b e o us 1
llo tLEVEL
ak %d_%U'
CUMULATIVE INCREMENTAL
a
DATABASE (FORMAT e n se
b
PLUS
llo lic FORMAT '%d_%U' NOT BACKED UP;
eARCHIVELOG
em
keEOF
a 5. Log in to the protected database host as a user who is a member of the OSBACKUPDBA
operating system group.
6. Open a text editor, create a file with the following contents, and save the file using the
name.crontab into your home directory. This example uses the crontab utility to
schedule the RMAN script.
Sample:
[email protected]
# MI HH DD MM DAY CMD
00 1 * * * /u01/app/oracle/product/11.2.0.4/db_home1/db_incr_daily.sh
7. In a command window, change directory to your home directory and enter the following
command to create a crontab file for this user from the contents of .crontab.
Sample:
# crontab .crontab
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Cloud b l l o
e provides
Control li reporting capabilities and also enables you to monitor and change the
k e em of protected database backup and recovery jobs.
status
a Cloud Control provides a Backup Report that displays the details of all backup and recovery
jobs that were run for a particular protected database. You can filter the data displayed in this
report depending on the time period for which you want to see the reports.
Monitoring the Protected Databases Backup - Cloud Control
To display the Backup Report for a protected database:
1. Access the home page for the protected database
2. From the Availability menu, select Backup & Recovery, then select Backup Reports.
3. Filter the data displayed in the backup report using the fields in the Search section. You
can filter by job status, type of backup job, or start time.
4. Click Go to display the backup report.
5. To view the details of a particular job execution, select the job and click View Results.
a ble
Monitor backup
fe r
job progress
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Monitor thel l o
e progressliof backups, copies, and restores by querying the V$SESSION_LONGOPS
b
k e em RMAN uses detail and aggregate rows in V$SESSION_LONGOPS. Detail rows describe
view.
a the files that are being processed by one job step. Aggregate rows describe the files that are
processed by all job steps in an RMAN command. A job step is the creation or restoration of
one backup set or datafile copy. The detail rows are updated with every buffer that is read or
written during the backup step, so their granularity of update is small. The aggregate rows are
updated when each job step is completed, so their granularity of update is large.
Monitoring the Protected Databases Backups - CLI
Note: Set the STATISTICS_LEVEL parameter to TYPICAL (the default value) or ALL to
populate the V$SESSION_LONGOPS view.
The relevant columns in V$SESSION_LONGOPS for RMAN include:
• OPNAME: A text description of the row. Detail rows include RMAN:datafile copy,
RMAN:full datafile backup, and RMAN:full datafile restore.
• Compression
– No RMAN compression is recommended. Use Database
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
Data in the Database
No RMAN Encryption and RMAN RMAN RMAN Encryption and
an s
No RMAN Compression Encryption Compression RMAN Compression
n - t r
Not Encrypted YES NO
a no YES NO
TDE Tablespace
YES a
YES
h s NO NO
Encryption
m ) ideฺ
g ฺ co t Gu
n kn dForever
Support for Incremental
a e n
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l
Compression
l o li
b
k e em recommends that you enable Compression on tables and indexes in the database,
Oracle
a and then take incremental backups as usual. Table/Index or Tablespace compression
requires the Advanced Compression Option.
Encryption
Oracle recommends that you enable TDE on table spaces in the database, and then take
incremental backups as usual. TDE requires the Advanced Security Option. The benefits of
TDE are as follows:
• TDE is transparent to applications.
• Backups of encrypted table spaces, and redo describing changes to these table spaces,
are encrypted. The TDE-encrypted data blocks are secured on the protected database,
Recovery Appliance storage, tape devices, and replicated appliances, and also when
transferred through any network connections.
• TDE on the source database reduces overhead on downstream servers.
• This technique supports an incremental-forever strategy and virtual full backups.
a a
( cen
l l o
e Backups
Long-Term li
b
k e emcustomers requiring a much higher backup and recovery retention policy when using the
For
a Recovery Appliance, please follow this solution. Longer backup and recovery retention may
be mandated by a specific business requirement, for governmental or financial regulatory
standards or for standards such as HIPPA. Regardless of the reason, setting a very large
recovery window goal (such as one year) may not be cost effective for any protected
database; so, this solution describes how to KEEP certain TAGGED backups beyond the
Recovery Appliance's recovery window goals to meet higher backup and recovery retention
requirements.
Implementing a long-term backup solution consists of the following steps:
1. Create a KEEP database backup for long-term retention: The recovery appliance
handles this type of requirement via the use of RMAN KEEP Backups. An RMAN keep
backup will backup all datafiles, as well as the control files / spfiles and archivelogs and
these are given a specific TAG. The RMAN run block looks like this:
backup device type sbt tag 'KEEP_DEC2015' database keep until time
‘to_date(sysdate+1830)';
@ u e th
b e llo to us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
( a ak nse
liceinvolving Recovery Manager (RMAN) consists of a set of backup
ello strategy
A basicbbackup
m that are scheduled to run at specified intervals. When you decide to use the Recovery
kee
scripts
a Appliance for data protection, you need to develop a plan to migrate your existing backup
strategy to one that uses the Recovery Appliance.
Your existing RMAN backup strategy typically involves storing backup metadata in an RMAN
recovery catalog. When migrating to a data protection strategy that uses the Recovery
Appliance, you need a strategy to manage existing backup metadata that is stored in an
RMAN recovery catalog.
Your existing backup strategy may store protected database backups in a local disk location
or on a shared disk. After you import the metadata for the protected database backups into
the Recovery Appliance catalog, you must migrate existing backups that are within the
recovery window goals to the Recovery Appliance storage.
a k
l o (a cens
The table
b l
eabove li
describes the changes and limitations to standard RMAN commands when
g ฺ
BACKUP INCREMENTAL LEVEL 1 DATABASE PLUS ARCHIVELOG;
}
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
An RMAN l l o li
e channel represents one stream of data to or from a backup device. The channel
b
em data from the input device, processes it, and then writes it to the output device. RMAN
ereads
ak supports the following types of channel configurations:
• DISK: backups are stored on disk.
• SBT: backups are stored on one of the following: tape using media management
software such as Oracle Secure Backup, the Recovery Appliance using the Recovery
Appliance backup module, or on the Oracle Cloud.
To back up to or restore from a Recovery Appliance, you must configure and use an SBT
channel that corresponds to the Recovery Appliance backup module installed on the
protected database host. In a Recovery Appliance environment, your backup and recovery
scripts must allocate an SBT channel with the SBT_LIBRARY parameter pointing to the
Recovery Appliance backup module.
Use the RMAN ALLOCATE command to allocate RMAN SBT channels that will be used to
back up to or recover from the Recovery Appliance. For a particular operation, you can
override the persistent configuration that was set using the CONFIGURE command by
explicitly allocating an RMAN SBT channel before the operation. Enclose the ALLOCATE
command and the other commands in a RUN block.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
Steps
1. Connect to the Protected Database and the Recovery
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
e RMANlibackup strategy typically involves storing backup metadata in an RMAN
l
Your existing
b
em catalog. When migrating to a data protection strategy that uses the Recovery
erecovery
ak Appliance, you need a strategy to manage existing backup metadata that is stored in an
RMAN recovery catalog.
Connect
To connect to a protected database (non-CDB) and Recovery Appliance:
RMAN> CONNECT TARGET sys as sysdba;
RMAN> CONNECT CATALOG ra_rman_user@ra1;
To connect to a protected database (CDB) and Recovery Appliance:
RMAN> CONNECT TARGET c##bkuser@my_cdb;
RMAN> CONNECT CATALOG ra_rman_user@ra1;
the upgraded catalog is fine and can be used to recover the protected database.
10. Start RMAN and connect as CATALOG using the rasys user. Note that rasys is the
owner of the Recovery Appliance catalog.
# rman CATALOG rasys/ra_pswd@ra-scan:1521/zdlra5
11. Import the source RMAN recovery catalog into the Recovery Appliance catalog. The
credentials of the source RMAN recovery catalog are provided by the protected database
administrator.
a ble
IMPORT CATALOG rman_cat11/rmancat11_pswd@dbrcat11 NO UNREGISTER;
fe r
IMPORT CATALOG rman_cat11/rmancat11_pswd@dbrcat11
ans
n - t r
DB_NAME 'MY_PTDB' NO UNREGISTER;
o
an
12. Verify that all the backup pieces are included in the Recovery Appliance catalog by
s
querying the RC_BACKUP_PIECE_DETAILS view.
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• Prerequisites
– Migrate existing level 0 incremental backup
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
l o
e backupli strategy may store protected database backups in a local disk location
l
Your existing
b
ak eorem
on a shared disk. After you import the metadata for the protected database backups into
the Recovery Appliance catalog you must migrate existing backups that are within the
recovery window goals to the Recovery Appliance storage. You can migrate backups for only
a subset of the protected databases contained in the imported RMAN recovery catalog.
However, for all the databases represented in the imported catalog, you can begin using the
Recovery Appliance catalog as your recovery catalog.
To begin using an incremental-forever backup strategy with Recovery Appliance, you must
first submit a level 0 incremental backup. If a recent level 0 incremental backup already exists
for a particular protected database, it might be more convenient to migrate that backup into
the Recovery Appliance, rather than take another level 0 backup from the database. After
migrating the level 0 backup and any required existing level 1 backups and archived log files,
you can then begin the incremental-forever strategy by sending level 1 incremental backups
and archived log files.
a n kn den
Backup Copies as Backup Sets ion
b S tu
u n t h i s
Back up image copies that are
l l o @ us
stored e
on local disk storage as backup sets to the Recovery
k b e
Appliance using the RMAN
t o
BACKUP AS BACKUPSET COPY OF DATABASE command. You
must configure anaSBT channel
( a e n se that corresponds to the Recovery Appliance backup module.
b ello lic
k e em
a
Databases
• Enable Real-Time Redo Transport in a Protected Database
• Enroll Protected Databases
• Take a Backup of a Protected Database
• Recover a Protected Database a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Cloud b l l o
e provides
Control li the following techniques to recover protected databases:
k e e•m Oracle Advised Recovery: Oracle Advised Recovery enables you to recover the
a protected database by using the automatic repair actions recommended by Data
Recovery Advisor. The Data Recovery Advisor automatically diagnoses data failures,
assesses their impact, reports these failures to the user, determines appropriate repair
options, and executes repairs at the user's request.
• User Directed Recovery: This technique performs manual recovery based on the
specified criteria. You must provide information such as the objects that must be
recovered (database, datafiles, tablespaces, archived redo logs), whether to perform
complete recovery or point-in-time recovery, location to which database files must be
recovered, and so on.
a ble
fe r
an s
Recovery Appliance
n - t r
no
Backup Module
s a
h a
) ideฺ
m
co t Gu
g ฺ
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
When b l l o li Appliance for restore and recovery operations, the RMAN
e the Recovery
using
k e em
connection syntax used is the same as with a regular RMAN recovery catalog connection.
a The only difference is that you connect to the Recovery Appliance catalog and configure
RMAN channels.
Prerequisites
• The protected database must be enrolled and registered with the target Recovery
Appliance.
• Backups required to restore and recover the protected database must be stored on the
Recovery Appliance.
• The Oracle wallet containing credentials used to authenticate with the Recovery
Appliance must be configured on the protected database.
• Configure or allocate RMAN SBT channels that correspond to the Recovery Appliance
backup module.
RUN
{
RESTORE DATABASE;
RECOVER DATABASE;
ALTER DATABASE OPEN;
}
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
n i o
a.first_time AND a.next_time; i
u e t h
e l lo@ oFIRST_CHANGE#
us
DB_KEY DB_NAME
a k b
SEQUENCE#
e t NEXT_CHANGE# COMPLETION_TIME
------ --------
l o c ns ------------- ------------ ---------------
(a ----------
e
24201 be l
PTDB2 li 9911 288402086 288430116 14/07/2014 5:27:49 PM
m
a ke4.eRestore and recover the control file and the protected database.
STARTUP NOMOUNT;
RUN
{
SET UNTIL TIME "TO_DATE('2014-14-07:17:27:49','yyyy-dd-mm:hh24:mi:ss')";
RESTORE CONTROLFILE;
ALTER DATABASE MOUNT;
RESTORE DATABASE;
RECOVER DATABASE;
ALTER DATABASE OPEN RESETLOGS;
}
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a k
l o (a cens
bel li
em
ake
Objectives
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
• Optimization: Overview
• Performance Tuning Best Practices
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
5
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Monitor
Use the Recovery
4 Discuss
Appliance Capacity
Planning Reports
3
Discuss the results;
tune and finalize
Define a ble
fe r
Define the size of the Recovery
s
2 Validate
Appliance based on the collected
information
a no n - t r an
1 s
Validate the collected data
h a
) ideฺ
Collect m
o Gu
cinformation
g
n entฺ
Collect existing database’s
n k
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
el
Methodologies
b li
k e e•m Comparative Sizing (vs. an existing system): Sizing estimate based on current
a utilization; capacity and throughput of existing backup appliance(s). It considers capacity
and throughput differences based on standardization of backups leveraging an
incremental forever strategy.
• Predictive Sizing (for a new deployment): Holistic approach to centralizing and
standardizing Oracle database backups. It starts with a “clean state” to architect a new
centralized backup management strategy.
• Sizing estimates is a matter of doing the math: • Add InfiniBand-connected racks to scale
- Size and growth rate of database – Add a spine switch when connecting
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
l l o
e Appliance
The Recovery li supports up to 5.8 PB of virtual full backups in a single full rack, and
b
k e em100 PB of virtual full backups in a maximum configuration of 18 full racks.
over
a If additional capacity is required, a second base rack can be connected via high-speed
InfiniBand to the first rack. The second rack includes its own pair of compute servers that add
connectivity and processing power to the configuration. As with the first rack, storage capacity
can be easily expanded by incrementally adding storage servers. Up to 18 fully configured
racks can be connected together into a single appliance, providing more than 10 PB of usable
capacity, i.e. more than 100 PB of virtual full backups.
A single rack Recovery Appliance can achieve a sustained Delta Ingest rate of up to 12
TB/hour. In other words, it can receive 12 TB/hour of change data, and convert it into
120TB/hour of virtual backups. It is able to support a restore rate of up to 12 TB/hour.
As racks are added to the configuration, both performance and capacity increase linearly. An
18-rack Recovery Appliance achieves virtual full backup rates of up to 2 PB/hour, and 216
TB/hour of Delta Ingest and restore.
a a
( cen
l l o
e describes,
As thisbslide li optimization should focus on RMAN channel parallelism, the number
m cells, connection parallelism and the internal optimizations of the Recovery
eofestorage
ak Appliance.
• Optimization: Overview
• Performance Tuning Best Practices
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a a
( cen
The RMAN l l o li break up large files into sections, and back up and restore these
e can optionally
b
em independently. You can use this built-in RMAN feature with the Recovery Appliance
esections
ak to improve performance. You can use it to create multisection backups, which break up the
files generated for the backup set into separate files. This can be done for backup sets and
image copies.
Each file section contains a contiguous range of blocks of a file. Each file section can be
processed independently, either serially or in parallel. Backing up a file into separate sections
can improve the performance of the backup operation, and it allows large file backups to be
restarted.
A multisection backup job produces a multipiece backup set. Each piece contains one section
of the file. All sections of a multisection backup, except perhaps for the last section, are of the
same size. There is a maximum of 256 sections per file.
• Backup strategy:
– Take only one level 0 backup; ensure that indexing of this
backup has completed before taking another.
– Take a daily cumulative incremental level 1 to reduce database
recovery time.
• The most common bottlenecks are the network and the ra ble
s fe
protected database’s I/O system. r a n
o n -t
• Use block change tracking (BCT) for all protected
a n databases.
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
RMANbChannels li
k e emexample, if you have registered 15 protected databases, you should be able to view a
For
a maximum of 15 x 2 = 30 channels connected to the RAC database. Do not over-allocate
RMAN channels because doing so can result in degraded performance.
Backup Strategy
Use the following query to verify the level 0 backup has been indexed before you create
another level 0 backup:
select bp_key from rc_backup_piece where tag = ‘&tag' and
backup_type ='D' and virtual = 'NO';
Network Bottleneck
The total network throughput per database node is 80GbE/sec. Separate interface
assignments by activity (such as separate replication traffic from ingest traffic, if possible).
Usually only a handful of blocks change between incremental backups, so RMAN does a lot of
unnecessary work reading the blocks that are not required for backup. The BCT provides a way
to identify the blocks required for backup without scanning the whole datafile. After that, RMAN
needs only to read blocks that are really required for the incremental backup.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e production
Compressed li data remains compressed in the backup. So, this data will not benefit
b
k e emfurther compression during the backup (such as RMAN backup or Recovery Appliance
from
a compression). In addition to this, deduplication software cannot deduplicate compressed data.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
TDE bel li
k e emis a feature that enables encryption of individual database columns before storing them
TDE
a in the datafile, or it enables encryption of entire tablespaces. If users attempt to circumvent
the database access control mechanisms by looking inside datafiles directly with operating
system tools, TDE prevents such users from viewing sensitive information.
Network Encryption
Review the following support notes to enable TCPS protocol with SCAN Listeners.
How To Configure Scan Listeners With A TCPS Port (Doc ID 1092753.1)
Step-by-Step Guide: How to Configure SSL / TCPS on ORACLE RAC (with SCAN)(Doc ID
1448841.1)
a k
l o (a cens
el
Keep itbsimple. li
k e em
a
a a
( cen
e l l o
How tobdetermine i Recovery Appliance is keeping up with a load.
if lthe
ak eem
a k
l o (a cens
The Recovery
b li administrator creates database user accounts on the appliance.
el Appliance
em are called virtual private catalog (VPC) users.
eThese
ak These users have privileges to send and receive backups for one or more protected
databases.
Each Recovery Appliance user owns a VPC for accessing metadata in the catalog for only
those databases
The protected database’s administrator with SYSDBA or SYSBACKUP privilege is associated
with the Recovery Appliance user.
Authentication credentials of the Recovery Appliance user are stored securely in an Oracle
Wallet on the protected database host.
The database administrator connects to the Recovery Appliance VPC user using the catalog
role.
a a
( cen
l l o li
m be
e
ake
a a
( cen
Extreme l l o li should NEVER occur!
espace pressure
b
k e e•m This could only happen if the Recovery Appliance administrator IGNORES the
a numerous alerts and warnings.
The backup copy policy setting in the protection policy defines prior if extreme space pressure
occurs.
• Always accept new, incoming backups (default)
• The oldest backup within the recovery window could be purged to make room for
incoming backups.
• Refuse new, incoming backups if there is not enough space to meet a recovery window.
Backups within recovery window could only be purged on disk if they had been copied to tape
or replicated.
If this policy is enabled, backups on the Recovery Appliance will never have more than their
reserved space setting.
a k
l o (a cens
Extreme
b espace li should NEVER occur!
l pressure
k e e•m Could only happen if the Recovery Appliance administrator IGNORES the numerous
a alerts and warning
The Backup Copy Policy setting in the Protection Policy defines prior IF extreme space
pressure occurs:
• Always accept new, incoming backups (default).
• The oldest backup within the recovery window could be purged to make room for
incoming backups.
• Refuse new, incoming backups if not enough space is available to meet the recovery
window.
Backups within the recovery window could only be purged on disk, if they’d been copied to
tape or replicated.
If this policy is enabled, backups on the Recovery Appliance will never have more than their
reserved space setting.
• Optimization: Overview
• Performance Tuning Best Practices
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a k
l o (a cens
The Recovery
b li ships with two utilities, rastat.pl and
el Appliance
em
enetwork_throughput_test.sh, that can assist you in evaluating the performance of your
ak system.
rastat.pl is a command line utility that runs tests against the Recovery Appliance to gather
performance statistics that can help you identify system bottlenecks.
The tests can generate statistics on:
• Backup data sent to the Recovery Appliance over the network
• Restore data received from the Recovery Appliance over the network
• Recovery Appliance ASM disk group read or write I/O
• Recovery Appliance container file read or write I/O
• Recovery Appliance container file allocation rate
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• Examples
>$ORACLE_HOME/perl/bin/perl rastat.pl --test=NETBACKUP --filesize=2048M
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
--catalog=rman/rman@inst2 --parms='SBT_LIBRARY=/u01/oracle/lib/libra.so,
ENV=(RA_WALLET=location=file:/u01/oracle/dbs/ra_wallet
credential_alias=ra-scan:1521/zdlra5:dedicated)'
a ble
Disk Group: /:BLOCK_POOL
fe r
>$ORACLE_HOME/perl/bin/perl rastat.pl --test=ASMREAD --filesize=2048M
an s
--rasys=admin/admin@inst2 --diskgroup=+RCVAREA
2048 MB, 9.55s write IO time, 3.50s CPU time, 214.35 MB/s, 36.60% CPU usage
n - t r
RECOVERY APPLIANCE READ IO TEST FROM DISK
PL/SQL procedure successfully completed.
a no
Disk Group: +RCVAREA
h a s
2048 MB, 6.06s read IO time, .65s CPU time, 337.99 MB/s, 10.79% CPU usage m
) ideฺ
g ฺ co t Gu
PL/SQL procedure successfully completed.
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
To runb the l l o
e rastat.pl li utility:
ak ee1.m Ensure that the system from which you are running the utility meets the requirements.
2. Open a command prompt window.
3. Enter the applicable command syntax for the tests you want to run, and press Enter.
• Is a shell script
• Uses the QPERF utility part of the Exadata Engineered
System
• Sets up trusted SSH connectivity between the execution node
and send/receive nodes le
r a b
s fe
- t r an
no n
s a
h a
) ideฺ
$ ./network_throughput_test.sh -s sending_nodes -r receiving_nodes -i bondeth0
m
Using 'scam09db01 scam09db02 scam09db03 scam09db04 scam09db05 scam09db06 scam09db07 scam09db08' hosts for sending nodes
co t Gu
ฺ
Using 'scas10adm01 scas10adm02 scas10adm03 scas10adm04' hosts for receiving nodes
g
Validating Trusted SSH to sending nodes scam09db01 scam09db02 scam09db03 scam09db04 scam09db05 scam09db06 scam09db07 scam09db08 ... OK
kn den
Validating Trusted SSH to receiving nodes scas10adm01 scas10adm02 scas10adm03 scas10adm04 ... OK
n
Total Network Bandwidth 8,372,597,121 bytes/sec
a
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l
Assumptions
l o li
b
k e emscript uses the QPERF utility that is provided as a part of an Exadata Engineered System
This
a as well as a part of the Zero Data Loss Recovery Appliance.
This script may be executed from any Linux server with a BASH shell, and requires trusted
SSH to be established between the execution node and the two sets of nodes that are being
tested.
The Execution node may be one of the nodes being tested.
The O/S user used for trusted SSH is the O/S user running the script. There is no support for
the script to use a different O/S user for logging into the nodes being tested.
Installation
If the script is not already available under /opt/oracle.RecoveryAppliance/client, then
download it as instructed in the MOS Note given later in the lesson.
host name per line, similar to the group file used by DCLI utility
The interface defaults to bondeth0 but can be any interface on the receiving nodes, such
as bondeth1 if testing network throughput to a downstream replica.
MOS Note: Zero Data Loss Recovery Appliance Network Test Throughput script (Doc ID
2022086.1)
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• Optimization: Overview
• Performance Tuning Best Practices
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Appliance
The Recovery li home page is a virtual command center providing an overview of all
b
em of centralized management of the environment including storage, performance,
easpects
ak recent activity, issues that may need attention, and a summary of the overall environment.
The Home Page is divided into sections, with each section providing a snapshot of the
relevant information.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
ake
me
bel
l o
a k
li
l l o
(a cens
u
be e to
n
@ use
b a n
nio this
k
S t
ฺ c
ud
ng ent
)
s
om Guid
ha eฺ
an
o n - t r an
sfe r a b
le
8
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a k
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
Remote Appliance.
s a
h a ฺ
) centers/offices
m
Hub and Spoke • Multiple remote data
i d e
g ฺ
• Backups arecoaggregated
t G uone central
to
a n kn den
location
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Topologies
Replication li
b
k e em
Replication topologies can be as simple or complex as required. The primary variables are as
a follows:
• The total number of Recovery Appliances in the replication environment, and their
relationships to one another
• The protection policies on the upstream Recovery Appliance that manage the outgoing
replicated backups, and the policies on the downstream Recovery Appliance that
manage the incoming replicated backups
• The replication server configurations that exist on each Recovery Appliance in the
replication environment
• The association between a replication server configuration and a protection policy
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Conceptse l l o li
b
k e em
To protect against server or site outage, a Recovery Appliance can replicate backups to a
a different Recovery Appliance.
When the upstream Recovery Appliance sends the incremental backup (full level 0 for the first
time and incremental level 1 subsequently) to the downstream Recovery Appliance, it creates
a virtual full backup as normal.
The downstream Recovery Appliance creates backup records in its recovery catalog. When
the upstream Recovery Appliance requests the records, the downstream Recovery Appliance
propagates the records back.
If the local Recovery Appliance cannot satisfy virtual full backup requests, then it
automatically forwards them to the downstream Recovery Appliance, which sends virtual full
backups to the protected database. DBAs use RMAN as normal, without needing to
understand where or how the backup sets are stored.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a a
( cen
l o
e Processli
l
Replication
m b
eAeprotected database is associated with each Recovery Appliance catalog where its backups
a k reside. In this way, a backup could be restored directly from a downstream Recovery
Appliance by connecting to that Recovery Appliance’s catalog. Furthermore, the replicated
backup information is also stored on the upstream Recovery Appliance. So, if the backup has
been expired from the upstream, it will be automatically retrieved from the downstream
Recovery Appliance.
An initial full backup (original incremental level 0 or the most recent virtual full) is replicated,
and then follow-on incremental backups and archived logs are replicated. The backup would
be ingested on both the upstream and downstream Recovery Appliances as each is
independent of the other.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
User Interfaces
• Oracle Enterprise Manager Cloud Control
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
le
PROXY_URL VARCHAR2 IN DEFAULT
b
PROXY_PORT NUMBER IN DEFAULT
…
HTTP_TIMEOUT NUMBER IN DEFAULT
fe r a
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Interfaces
Replication li
b
k e emReplication page in Cloud Control is the recommended interface for configuring the
The
a Recovery Appliance replication, but you can use also the DBMS_RA package to create and
manage replication.
The previous examples show how easy it is to create a Replication Server, add a Protection
Policy to the Replication Server, and a description of the parameters to pass to the
DBMS_RA.CREATE_REPLICATION_SERVER procedure to create a Replication Server.
Replication
Enabled
Replication
Summary
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
Replication
n k ng ent
Destination
n b a
S t ud
u nio this 11
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
el
OraclebEnterprise li
Manager Cloud Control administration web pages
k e em Page
Home
a • Summary: In this section, use the Current Activity tab to display information about
Backup, Copy to Tape, Replication, and Restore operations.
• Replication: Recovery Appliance can replicate backup data to another Recovery
Appliance. This section contains a table that lists other Recovery Appliances that are
replicating to this Recovery Appliance or that are being replicated to or from this
appliance. If replication has not been configured, a link will appear that allows you to
configure replication.
Replication Page
Use this page to configure replication of the backups associated with a protection policy. A
Recovery Appliance can replicate backup data to other local or remote Recovery Appliances.
The Protection Policies table displays the protection policy Name, Recovery Window Goal
(days), Storage Location, Backup Polling Location, and Copy-to-Tape and Replication
Settings.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a a
( cen
e l l o
Beforebconfiguring li
replication, protected databases for which backups are to be replicated
database user credentials for a virtual private catalog user on the downstream
Recovery Appliance who has been granted permission to manage the protected
databases for which backups will be replicated.
d. Use the Upstream Recovery Appliance Host Credentials field to specify credentials
for the operating system user who owns the Recovery Appliance database
installation.
4. Click OK.
After that you should be able to add Protection Policies and Pause or Resume Replication fora ble
maintenance tasks. fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
REPLICATION_SERVER_NAME REPLICATION_SERVER_ST
ZDLRA1_REP RUNNING
SQL>
a ble
Adds a replication server configuration to the protection
fe r
ADD_REPLICATION_SERVER policy that was created by the
CREATE_REPLICATION_SERVER procedure
an s
n - t r
REMOVE_REPLICATION_SERVER a no
Removes a replication server configuration from the
protection policy that was created by the
a s
CREATE_REPLICATION_SERVER procedure
h
m ) ideฺ
PAUSE_REPLICATION_SERVER or ฺ co pauses
Thisgprocedure
t G uor resumes replication to the
RESUME_REPLICATION_SERVER
n n downstream
kspecified e n Recovery Appliance
a d
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l
Command-Line
o li
Interface
b
k e emcan use the DBMS_RA package to create and manage replication.
You
a The Recovery Appliance permits in-progress replication of backup pieces to complete. If the
Recovery Appliance queued backup pieces for replication through this replication server
configuration but did not replicate them, then the Recovery Appliance holds the backup
pieces until you call RESUME_REPLICATION_SERVER. No replication tasks that run against
this Recovery Appliance can execute until you resume replication to the downstream
Recovery Appliance.
REPLICATION_SERVER_NAME REPLICATION_SERVER_STATE
------------------------------ ------------------------------
ZDLRA1_REP RUNNING
SQL>
a b le
fe r
ans
SQL> SELECT DISTINCT REPLICATION_SERVER_NAME,
n - t r
REPLICATION_SERVER_STATE FROM RASYS.RA_REPLICATION_SERVER;
a no
h a s
REPLICATION_SERVER_NAME
m ) ideฺ
REPLICATION_SERVER_STATE
g ฺ co t Gu
------------------------------ ------------------------------
a n kn den
nb s Stu
ZDLRA1_REP PAUSED
n i o i
u e t h
e l lo@ o us
SQL> EXEC RASYS.DBMS_RA.RESUME_REPLICATION_SERVER( 'ZDLRA1_REP' );
a a kb se t
l o ( censuccessfully completed.
PL/SQL procedure
b e l li
m
a keeSQL> SELECT DISTINCT REPLICATION_SERVER_NAME,
REPLICATION_SERVER_STATE FROM RASYS.RA_REPLICATION_SERVER;
REPLICATION_SERVER_NAME REPLICATION_SERVER_STATE
------------------------------ ------------------------------
ZDLRA1_REP RUNNING
SQL>
CLI Views
Different data dictionary views related to Replication functionality:
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
View Description
This view describes the downstream Recovery
RA_REPLICATION_SERVER Appliances that are directly receiving replicated
backups from this particular Recovery Appliance.
The POLICY_NAME column of this view lists the
a ble
protection policy used by this protected
fe r
RA_DATABASE database. The REPLICATION_USAGE column
an s
shows the cumulative amount of disk space (in
n - t r
a no
GB) replicated for this database.
This view describes the defined protection
RA_PROTECTION_POLICY policies. h a s
m ) ideฺ
RC_BACKUP_PIECE_DETAILS ฺ c o Gu about all available
Lists detailed information
n b an tude
u nio this S
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
el
CLI Views
b li
k e emcan monitor replication using the Recovery Appliance catalog views.
You
a
REPLICATION_SERVER_NAME REPLICATION_SERVER_STATE
PROTECTION_POLICY
a ble
------------------------------ ------------------------------ ------
fe r
------------------------
an s
- t r
non
ZDLRA1_REP RUNNING BRONZE
ZDLRA1_REP RUNNING
h a sa GOLD
ZDLRA1_REP RUNNING
m ) ideฺ SILVER
g ฺ co t Gu
SQL>
a n kn den
i o nb s Stu
u n t h i
l l o @ us
Display cumulative GB replicated for e
several databases …
a k be e to
SQL>
l o (a cens
SQL>be
l li
SELECT DB_UNIQUE_NAME, REPLICATION_USAGE FROM
m
kee
RASYS.RA_DATABASE;
a
DB_UNIQUE_NAME REPLICATION_USAGE
-------------------------------- -----------------
CUST15 0
DBNZDL 0
CUST05 0
ODA1RAC 6601.02258
DB1116 6240.13765
ODA2DBLG 478.316072
DB1124 3314.12441
CUST02 0
ODA2RAC 5269.81718
DB1025 365.495429
CUST09 0
… Zero Data Loss Recovery Appliance Overview 8 - 19
Summary
Appliance
• Describe the replication process for the Recovery Appliance
• Configure the Recovery Appliance replication process
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a k
l o (a cens
bel li
em
ake
Objectives
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a a kb se t
l o ( cen
A robust
b l li protects data against intentional attacks, unintentional user errors
ebackup strategy
m as file deletions), and software or hardware malfunctions. Tape libraries provide
kee
(such
a effective protection against these possibilities.
The fundamental difference in the two approaches is that the Recovery Appliance backs up to
tape, and not to the protected databases. The Recovery Appliance comes with preinstalled
Media Manager software, and supports optional Fibre Channel cards. Thus, installation of a
media manager is not necessary on the protected database hosts.
When the Recovery Appliance executes a copy-to-tape job for a virtual full backup, it
constructs the physical backup sets, and copies them to tape, and then writes the metadata to
the recovery catalog. If you want, the Recovery Appliance can also copy successive
incremental backups and archived redo log file backups to tape. Whereas the backup on the
Recovery Appliance is virtual, the backup on tape is a non-virtual, full physical backup. The
Recovery Appliance automatically handles requests to restore backups from tape, with no
need for administrator intervention.
media managers.
• Tape drives and tape libraries function more efficiently because the Recovery Appliance
is a single large centralized system with complete control over them. In other tape
solutions, hundreds or thousands of databases can contend for tape resources in an
uncoordinated manner.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• Red Stack
• The Recovery Appliance uses Integrated Media Manager
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
software
• Components
– 1 x QLogic QLE8362 Converge Network Adapter (configured for FC)
per node
– 2 x Brocade 6510 Switches (16GB x 48 port switches) for HA
– Oracle StorageTek Modular Library System:
a ble
FC-SCSI: SL3000 and SL150 fe r
—
an s
— ACSLS API: SL8500 and SL3000
n - t r
— ACSLS Logical Libraries: SL8500 and SL3000
a no
• Encryption
h a s
– Oracle Key Manager m ) ideฺ
• Compression g ฺ co t Gu
a n kn den
– Tape Library compression
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
e Applianceli Media Manager
Recovery l
b
k e emRecovery Appliance Media Manager provides reliable, centralized tape management by
The
a protecting file-system data and Oracle Database files for multiple environments. This Media
Manager is the tape management component for the Recovery Appliance. It is installed with
its components on the Recovery Appliance during its configuration.
Components
• QLogic QLE8362 Converge Network Adapter provides industry-leading native FC
performance with up to 16 Gb throughput at extremely low CPU usage and full hardware
offloads. It also supports 10 Gb FCoE functionality but the Recovery Appliance
implements only native Fibre Channel connectivity. There are two HBAs, one per
compute node installed. The HBA is configured via the ra_install script during
installation. For additional information refer to the following PDF file:
http://www.qlogic.com/OEMPartnerships/Oracle/Documents/Storage/ds_16GbFC-PCIe-
universalHBA.pdf
different tap drives from Oracle and third-party providers. For additional information refer
to the following web page: http://www.oracle.com/technetwork/documentation/tape-
storage-187555.html
Encryption
Copies to tape could be encrypted by leveraging Integrated Media Manager host-based
encryption or by support of tape drive encryption such as Oracle Key Manager software.
Media Manager manages the encryption keys for encrypted backups on tape or leverages on
a ble
Oracle Key Manager software. ZDLRA recommends using the Oracle Key Manager for
fe r
s
enterprise key encryption needs. This should be configured and functional prior to ZDLRA
an
deployment. Tape drives should already be registered, and encryption enabled.
n - t r
Compression a no
Copies to tape would not be compressed by the appliance as h a
the
sbackups would benefit from
standard tape drive compression. m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
a a
( cen
e l l o li
b
k e em
a
preconfigured software.
– The Recovery Appliance has a dedicated library (partition).
– Each tape drive is accessed via one or two paths (HA).
– Redundant FC SAN is preferred (connected to two different SAN
switches).
• Third-party media management vendor (MMV) le
r a b
– The Recovery Appliance does not connect directly to the tape library
sfe
or the FC SAN infrastructure.
tr a n
n -
– The Recovery Appliance uses third-party MMV.
a no
– Media servers are supplied by the customer and
h a sconnect to the FC
SAN infrastructure.
m ) ideฺ
– Use 10 GigE to send backups fromgthe ฺ coRecovery
t G u Appliance Client to
media servers.
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
Recommendations li
b
emlibrary should have dedicated resources for the Recovery Appliance via either a
keThe
a physically dedicated library or virtualization like partitions and logical libraries. We recommend
using hardware-based drive cleaning with the Recovery Appliance. The tape library should be
online and ready to present to the Recovery Appliance. One or more tape drives should be
deployed. The drives must be dedicated to the Recovery Appliance.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
server
Maintains configuration
settings and backup OSB catalog
history catalog N
E
• Media server: Transfers T Media
data to or from attached W server
O
R a ble
devices e r
K s f
• Client: Accesses data to - t r an
Storage device
be backed up no n
s
Client a
h a
) ideฺ
Administrative
m
ฺ co t Gu
Domain
g
n k n en Data to backup
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
el
An administrative
b li in Integrated Media Manager is a group of machines on your
domain
em that you manage as a common unit to perform backup and restore operations. An
enetwork
ak administrative domain has one administrative server, one or more clients, and one or more
media servers. In the Recovery Appliance, the two Compute Nodes support all of these
components or host roles.
• The administrative server is one of the two machines in your Recovery Appliance
administrative domain. The Recovery Appliance contains a full installation of Integrated
Media Manager software in a cluster configuration. This host maintains the backup
catalog files and other files for configuration settings and administrative data. The
administrative server runs the scheduler, which starts and monitors jobs within the
administrative domain.
• Each compute node has a media server in your Recovery Appliance that has one or
more secondary storage devices, such as a tape library, connected to it. A media server
transfers data to and from its attached storage devices. The Recovery Appliance
compute servers are media servers for direct attached tape devices.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• Concepts
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Media family
Volume set
Volume Volume
Backup image
a ble
Product of a backup operation fe r
an s
Product of a backup operation
n - t r
Physical tape Physical tape a no
h a s
m )images d e ฺ
Set of tapes containing backup
ฺ co t Gu i
Set of volumes with common
g
kn dcharacteristics
n
a n e
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li organizes the backups it creates in a simple hierarchy, which
e Media Manager
Integrated
b
m the following logical concepts:
kee
comprises
a • A “backup image” is the product of a backup operation. A backup image is a file that
consists of one or more backup sections. A backup image can be contained within a
single volume or it can span multiple volumes. The part of a backup image that fits on a
single volume is called a “backup section.”
• A “volume” is a single unit of media, such as an LTO-5 tape. It can store many backup
images.
• A “volume set” is a group of one or more volumes that contain a complete backup
image.
• A “media family” or “tape pool” is a named classification of volumes that share some
common attributes, such as the volume naming method, the policies used for writing
data to volumes, and the length of time volumes are retained in the media family.
When you back up files by using Integrated Media Manager, you generate a volume set that
has some common characteristics defined by the media family that you have associated with
your backup operation.
CLI
• obtool
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
BUI a ble
fe r
• Web tool an s
n - t r
• Enterprise Manager
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e
Interfaces l l o li
b
k e e•m Command Line Tool - obtool: Use this tool to perform all Integrated Media Manager
a configuration, backup and restore, maintenance, and monitoring operations.
• Web tool: This is a browser-based GUI that enables you to configure an administrative
domain, manage backup and restore operations, and browse the backup catalog.
• Oracle Enterprise Manager: Use this tool to perform most Media Manager tasks,
including administrative domain and hardware configuration, managing your media, and
backing up and restoring databases. Oracle Enterprise Manager is the preferred Web
interface for Media Manager tasks.
– Key Components
— CRS stack
— RDBMS Media Manager Catalog
— DBFS shared file system
— Media Manager software
• Oracle Clusterware software a ble
fe r
– Use crsctl {status|check|start|stop} res {osbadmin|osbmedia}
a n s
-t r
– During a relocation, running jobs are impacted. non
• Copy-to-Tape Services h a sa
m )Media d e ฺ
– For Copy To Tape to work, we need all
ฺco t Gu i Manager
services up and healthy. kng n
a n d e
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l
Implementation
o li
b
k e em Media Manager is deployed as a clusterware software in the Recovery Appliance.
Integrated
a All services must be up for the Media Manager to start. The CRS service will pull up the
required services.
It has a cluster service for the admin role, which uses the Oracle RAC database as a Media
Manager catalog. Media server and client services are also managed by the clusterware
software.
Oracle Clusterware software commands are used to manage the Media Manager services.
Standard Media Manager commands still work. During a relocation, the job may restart in
Media Manager; otherwise, it will drop back into the Recovery Appliance retry logic and either
fail (if it has exhausted its retries) or re-execute. After the relocation, services and jobs resume
on their own.
To restart services or check their status, use crsctl as root.
For Copy to Tape to work, the Recovery Appliance server must be up and the SBT_LIBRARY
must be “READY”:
LIB_NAME STATU
------------------------------ -----
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
ROBOT0 READY
SQL>
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
[oracle@host1 ~]$
fe r
an s
n - t r
[oracle@host1 ~]$ crsctl stat res osbmedia
NAME=osbmedia no
Media Manager
a
TYPE=local_resource
h a s
Media services
) ideฺ
TARGET=ONLINE , ONLINE
STATE=ONLINE on host1, ONLINE on host2
m
co t Gu
[oracle@host1 ~]$
g ฺ
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
OraclebClusterware liMedia Manager Services
k e emcrsctl to check the status, start, and stop the Media Manager Admin, Media, and Client
Use
a Services. The Admin service will rotate between the nodes due to switchover or failover
maintenance operations. Each compute node has a media server and a client.
obtool commands
In addition to this you should be able to use Media Manager standard commands.
For example, to list host roles in the Media Manager implementation:
[root@host1 ~]# obtool lsh
host1 mediaserver,client (via OB) in service
host2 mediaserver,client (via OB) in service
host1-avip3 admin,mediaserver,client (via OB) in service
[root@host1 ~]#
h a s obtool utility
) ideฺ
-rwxr-xr-x 1 root root 78589 Dec 14 2015 obcm
-rwsr-xr-x 1 root root
-rwsr-xr-x 1 root root
150403 Dec 14 2015 obt
777101 Dec 14 2015 obtar m
co t Gu
g ฺ
-rwsr-xr-x 1 root root 1481839 Dec 14 2015 obtool
-rwxr-xr-x 1 root root
[root@host1 bin]#
a n kn den
70192 Dec 14 2015 php.ini
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
OraclebClusterware liMedia Manager Directories
k e em Media Manager is installed on a DBFS file system in the Recovery Appliance to be
Integrated
a accessible by the two compute nodes.
Configuration files and binaries are located in this DBFS file system:
[root@host1 OSB]# pwd
/dbfs_obdbfs/OSB
[root@host1 OSB]# ls -l etc
total 1
drwxr----- 5 root oinstall 0 Jan 28 2016 ob
-rw-r--r-- 1 root root 186 Jan 27 2016 obconfig
[root@host1 OSB]#
[root@host1 OSB]# ls -l etc/ob
total 1
drwxr-xr-x 2 root root 0 Aug 31 06:00 mail
drwxr-xr-x 2 root root 0 Nov 17 2015 report
drwxr----- 2 root oinstall 0 Jul 28 08:52 wallet
lrwxrwxrwx 1 root root 15 Nov 17 2015 xcr -> /usr/etc/ob/xcr
[root@host1 OSB]#
Zero Data Loss Recovery Appliance Overview 9 - 17
Copy-to-Tape Backup Operations in the Recovery
Appliance
• Defines the recovery window on tape • Associated with a Media Manager Library
• Defines the number of streams to use for the
backup operation
Copy-to-Tape Job Templates
• Media management parameters (optional)
Define: – These override any parameters defined in
• Specific database or all databases in the Media Manager Library setting or via
a protection policy the Media Manager
Attribute Set h a s
• Sets maximum number of RMAN channels
m ) ideฺ
that can be used concurrently (based on the
number of available tape drives)
ฺco thetRMAN
• Copy-to-Tape schedule:one time or
recurring g
•nDefines G u media parameters to be
a n k usedden
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e
Operationsl l o li
b
k e em are the main categories of operations required to configure the Recovery Appliance to
These
a execute Copy-to-Tape Jobs:
• Media Manager Library
• Attribute Set(s)
• Copy-to-Tape Job Templates
• Recovery Appliance Protection Policy
le
library robot0 in service
drive 1 robot0_tape01 in service
r a b
e
drive 2 robot0_tape02 in service
ob>
s f
• Volume Sets and Volumes - t r an
no n
ob> lsvol -L robot0
s a
Inventory of library robot0:
in 1: h a
) ideฺ
volume host1_db-000036, barcode DEMO10,
m
1121196032 kb remaining, content manages reuse, mediainfo hw
co t Gu
encryptable
in 6:
g ฺ
volume host1_db-000035, barcode DEMO19,
encryptable
a n kn den
1093332992 kb remaining, content manages reuse, mediainfo hw
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Media b e l l
Library
o li
k e em its installation, while the Recovery Appliance is being configured, the Integrated Media
During
a Manager creates a media manager library with default parameters, such as the following:
• Library name (ROBOT0)
• Maximum number of accessible tape drives
• Number of restore drives
• Media manager location (robot0)
Attribute Set
An SBT attribute set helps you to further customize and categorize your backups while
copying them to tape. An attribute set is created for each tape drive associated with its media
manager library. It helps classify backups while storing them on tape by specifying
parameters such as the media pool number, streams, and media manager commands
needed to perform the tape backup operation.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• Administration Tasks
– Media Manager Library ROBOT0 and the Attribute Sets for
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
ฺ o Gu
cAttribute
Media Manager
k g
n enTasks t Set
Library Tasks
b a n ud
n S t
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
el Tasksli
Administration
b
k e em
To copy backups to tape by using the Recovery Appliance, you should first be able to create a
a media manager library for your media management software to manage all your tape backup
jobs by adding parameters that apply to a set of jobs. You should also be able to create a
media manager attribute set that helps you control your tape backup jobs further by adding
more job-specific parameters and commands for your media manager software.
In the slide, we illustrate how to execute some administration tasks related to create, edit, and
delete media manager libraries and attribute sets assigned to media manager libraries.
Remember that the Recovery Appliance comes with Integrated Media Manager preconfigured
as its media manager. Integrated Media Manager components, which include a media
manager library and attribute sets for each of its tape drives, are also preconfigured.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
Job Infrastructure li
b
em created by RMAN, backing up an Oracle database results in a backup set (an RMAN-
eWhen
ak specific logical structure), which contains at least one backup piece (an RMAN-specific
physical file containing the backed up data). In the Recovery Appliance, the Oracle Database
backups are stored as virtual full backups. So, when the Recovery Appliance executes a
copy-to-tape job for a virtual full backup, it constructs the physical backup sets, and copies
them to tape.
Integrated Media Manager backs up and maintains backup metadata for each RMAN backup
piece written to tape within its own catalog. You can browse backup pieces with the obtool
command line, or with Integrated Media Manager Web tool by using Oracle Enterprise
Manager Cloud Control.
Use the Copy-to-Tape Job Templates page to specify copy-to-tape job templates for a
selected database or databases associated with a protection policy. You can schedule the job
to define when the backups are copied to tape. A copy-to-tape job template also specifies a
media manager attribute set to control the copy operation.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
e Stateli
l
Maintenance
b
k e em performing tape hardware maintenance, you should be able to disable the
Before
a SBT_LIBRARY and ensure that no more jobs are active. You can choose to pause Media
Manager operations by using Cloud Control, and then resume them when required.
In addition to using Enterprise Manager, you can use DBMS_RA PL/SQL package
procedures.
When the Media Library is running, the query above returns READY.
a ble
fe r
LIB_NAME STATU
ans
---------------------------------------- -----
n - t r
o
ROBOT0 PAUSE
s an
) ha eฺ
SQL>
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
b el li
k e em
a
• Job Monitoring
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
TASK_ID STATE DB_UNIQUE_NAME ERROR_TEXT BS_KEY PIECE#
fe r
----------
2051036
----------
COMPLETED
-------------------- -------------------- ---------- ----------
ODA1RAC 1049767
an s
1
2051038 COMPLETED ODA1RAC 1049780
n - t r 1
no
2051040 COMPLETED ODA1RAC 1049800 1
2051042
2051044
COMPLETED
COMPLETED
ODA1RAC
ODA1RAC
s a 1049824
1049832
1
1
2051046 COMPLETED ODA1RAC
h a
) ideฺ
1049921 1
2051047 COMPLETED ODA1RAC
m
co t Gu
1049979 1
…
g ฺ
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
Job Monitoring li
b
em an SBT job runs, it generates and queues a task for each backup piece to be copied.
eWhen
ak RA_SBT_TASK lists these tasks and their completion states.
You can determine whether SBT jobs run according to your defined schedule by querying the
RA_SBT_JOB.LAST_SCHEDULE_TIME column. This column indicates the last time that the
SBT job was scheduled to run.
TEMPLATE_NAME BACKUP_TYPE
LAST_SCHEDULE_TIME
---------------------------------------- -----------------------------
- ----------------------------------------
a ble
SF_TAPE_AL ARCH 02-SEP-16
fe r
12.26.18.772028 AM -06:00
a n s
n -t r
SF_TAPE_SILVER FULL o
29-AUG-16
10.03.13.511986 AM -06:00
s an
SF_TAPE_SILVER_DAILY_INCR INCR ) ha eฺ 30-JUL-16
10.02.56.870003 AM -06:00
ฺ c om Guid
SF_TAPE_GOLD k
FULL
n ng ent 29-AUG-16
10.03.13.841137 AM -06:00 b a tu d
i o n s S
…
u n t h i
l lo@ o us e
e
kb se t
a a
( cen
e l l o li
b
k e em
a
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a k
l o (a cens
bel li
em
ake
Objectives
a k
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
Patching process
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
1 2 3 4 5 Complete
fe r
Complete
an s
n - t r
Unpack the Patch File Patching a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn Versions
Zero Data Loss Recovery Appliance Supported
d e n (Doc ID 1927416.1)
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
General l l o i
eInformationlAbout the Upgrade
b
k e emupgrade process should only be completed on clean and functioning Recovery
The
a Appliances. Attempting an upgrade while an appliance is in an unknown service state could
cause stability or upgrade issues.
• Read the full instructions in the readme file included in the patch file.
• The automation tool /opt/oracle.RecoveryAppliance/bin/ra_update.pl is
delivered in the ra_automation rpm. You should ensure that you have the latest
version of ra_automation RPM.
• The upgrade requires access to the compute nodes as root, and access to the rasys
user.
- This is done by setting up upgrade root keys that are password protected by the
upgrade administrator, and by disabling the rasys user to customers during the
install.
- At no time does the upgrade automation need to know the customer set
passwords for the system. Also, at no time are default passwords set.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
ab le
• The upgrade is non-rolling, and takes about a little more than an hour to fer
apply. a n s
n -t r
• Do not schedule backup jobs; Redo Shipping might have errors. o
• Upstream replication servers should be paused. s an
) ha eฺ
• Enterprise Manager agents will be stopped onm id servers during
the compute
ฺ c o u
the upgrade.
k n g nt G
b n and
• If Tape was used, pause the tapealibrary
t u dethen wait for the
n
nio this S
BACKUP_SBT tasks to complete.
u
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Plan the
b l
eUpgrade li
em there will be downtime, plan to do the upgrade at a time when it will least impact
eBecause
ak users. In addition, it might be useful to notify users when the system will be unavailable.
At the start of the patching process, the ra_update.pl script will stop all services.
Individual services may start and stop throughout the process, but ra_server itself will
remain unavailable.
Required Files
Obtain the required files:
• Get the latest Recovery Appliance patch set from here: Zero Data Loss Recovery
Appliance Supported Versions (Doc ID 1927416.1)
• Example contents of an RA Patch: (RA Patch Example is using Patch 22552579: ZERO
DATA LOSS RECOVERY APPLIANCE: 12.1.1.1.7
PATCHSET: p22552579_121110_Linux-x86-64.zip)
- p6880880_121010_Linux-x86-64.zip
- ra_automation-12.1.1.1.7-RELEASE.x86_64.rpm
- README.txt
12.1.0.2.160419 (APR2016)
- Ensure that the patches are copied over as the OS user oracle.
- Do NOT unpack the patch zips or create extra directories.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
h a s
m ) ideฺ
• If phase 1 fails, you should correctฺcthe o issues
u it identified.
g t G
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Validations
Pre-Patch li
b
k e empre-install script returns “FOUND OLD PATCHES RUNNING THROUGH TO REPLACE”
The
a when the Recovery Appliance system needs to be patched.
After the pre-install verification, update your RPM on all compute nodes by using the following
commands as root:
[root@host ~]# rpm -e ra_automation;
[root@host ~]# rpm -ivh /radump/ra_automation-12.1.1.1.7-
RELEASE.x86_64.rpm
If root authentication between compute nodes isn't configured, the upgrade script sets that
up for the upgrade. This authentication requires entering the root password once for each
compute node.
Open incidents should be reviewed, and if possible resolved prior to patching. Patching won't
generally resolve open incidents, and could actually have issues because of them.
The upgrade script validates that all services are up. This is done to start from a known
working state.
Welcome to RA Update.
Thu Jan 14 13:08:13 2016: Welcome to RA Update.
Thu Jan 14 13:08:13 2016:
RA Update Start
Thu Jan 14 13:08:13 2016: Patching Start
Thu Jan 14 13:08:16 2016: Pre Patch: Validating Recovery Appliance stack is
online
a ble
Thu Jan 14 13:08:16 2016: Check service status [crs]: Online.
fe r
ans
Thu Jan 14 13:08:17 2016: Check service status [db]: Online.
n - t r
Thu Jan 14 13:08:17 2016: Check service status [rep_dbfs]: Online. o
… s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
i
!!! NOTICE NOTICE NOTICE !!!
n o i
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
Patch b e l
Installation li
k e em patch installation, the process creates a temporary password for the rasys user. This
During
a is done so that you can access the system as rasys, and also to block others from changing
the system during the upgrade.
All services stop because the patching processing requires a complete service outage. If
there is any conflict, the new patches roll back. After that the upgrade script applies and
configures the database bundle patch and the Recovery Appliance patch and updates.
n
Thu Jan 14 13:54:35 2016: Relockedi o i
u
@ us e t h
Thu Jan 14 13:54:35 2016:
l l o Installing the Recovery Appliance Patch
k
Thu Jan 14 13:54:58
a be2016: e o
tInstalled the Recovery Appliance Patch
Thu Jan 14 13:54:58
l o ns Patching Complete
(a ce2016:
b
Thu Jane14 li 2016: Enabling zdlra-name
l 13:54:58
ak e…em
Thu Jan 14 14:09:28 2016: Validating post patch CRS environment
Thu Jan 14 14:10:24 2016: Validate post patch CRS environment: Compete
Thu Jan 14 14:10:25 2016: Post Patch: Checking for EM Agents :
Thu Jan 14 14:10:25 2016: Post Patch: EM Agent Checks: Complete
Thu Jan 14 14:10:25 2016: End RA Update: OK
[root@host1 bin]#
If phase 2 fails carefully review all current logs, correct the issue, and restart the script using –
restart: /opt/oracle.RecoveryAppliance/bin/ra_update.pl –restart –
patch_type=upgrade;
• Phase 3
– Restarts Services
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
STATE
------------
ON
SQL>
a ble
fe r
• The user performing the installation should then run the a n s
n -t r
cleanup steps. o
a nyou paused.
• Re-enable any upstream replication serversasthat
) h eฺ
com uid
• Re-enable any tape job that you paused.
g ฺ t G
k n n
n b an tude
u nio this S
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
el
Post Upgrade
b li
k e empatch process automatically restarts Recovery Appliance services. The user performing
The
a the installation should then validate the system and service access, and confirm that services
are functioning.
You should also verify that the Recovery Appliance system is running by querying the
RASYS.RA_SERVER table and verifying that the STATE column value is ON.
On each compute node, clean up the patch files used during the upgrade by using the
following example commands:
[root@host1 ~]# rm -rf /radump/patch
[root@host1 ~]# rm -rf /opt/oracle.RecoveryAppliance/data/upgrade
[root@host1 ~]# rm /radump/ra_automation-12.1.1.1.4-
RELEASE.x86_64.rpm
On the first compute node, clean up the zip files by using the following example commands:
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a k
l o (a cens
Patching
b l
eOracle li
Exadata Storage Servers
k e em for Exadata Storage Server are provided as complete software images that contain
Patches
a updates to the Linux operating system, cell server software, InfiniBand software, and
component firmware. This ensures that all the software and firmware components on an
Exadata Storage Server remain consistent with each other.
Exadata Storage Server patches are supplied independent of Recovery Appliance Database
Server patches. However, an Exadata Storage Server patch may require a specific Oracle
Database Server patch level, or database server firmware or operating system version.
Details are provided in the patch documentation and in My Oracle Support note 888828.1.
Exadata Storage Server patches can generally be applied in a rolling manner while databases
continue running, but for the Recovery Appliance, they currently cannot be applied so.
Patches may also include instructions for parallel (non-rolling) installation on multiple Exadata
Storage Servers.
Exadata Storage Server automatically maintains firmware within the server. Firmware levels
are periodically checked while Exadata Storage Server is running, and the correct firmware is
automatically applied to components when the server restarts. Disk firmware is also
automatically updated when a disk is replaced.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
# ./patchmgr –cells <cell_group> -patch [-rolling]
fe r
Non-Rolling an s
1. Patch n - t r
Cell 1
a no
patchmgr
1. Patch h a s
Cell 2 m ) ideฺ
1. Patch ng
ฺco t Gu
Cell nnk d e n
a tu (Doc ID 2124734.1)
b patchS(24306177)
nand
i o
Exadata 12.1.2.3.1 release
n i s
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
e l
OraclebExadata li Server Patch
Storage
k e em Storage Server patches are typically installed by using a script called patchmgr,
Exadata
a which is supplied with the patch. They may be installed in only one way: non-rolling.
A non-rolling patch is applied to all storage servers simultaneously with databases offline.
This approach requires down time, but may be preferred if a scheduled maintenance window
exists for the system. Because all of the storage servers are patched in parallel, the whole
patch may take approximately two hours to complete.
In either mode, a single invocation of patchmgr can be used to patch all of the Exadata
Storage Servers.
In all cases, ensure that you follow the patch-specific instructions for using patchmgr.
a a
( cen
Other b e l l o
Components liPatching
k e emsoftware and firmware on all components must always be maintained in accordance with
The
a the guidelines published in My Oracle Support note 888828.1. In particular, this is critical for
the InfiniBand switches. For other components, there may be less urgency to keep up with the
latest patches.
If the Tape option is included in Recovery Appliance, follow the QLogic Driver Download
Instructions from this MOS support note: Zero Data Loss Recovery Appliance Supported
Versions (Doc ID 1927416.1)
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a k
l o (a cens
el
OraclebEnterprise li
Manager Plug-In
k e emOracle Recovery Appliance plug-in for Enterprise Manager Cloud Control features
The
a monitoring dashboards and streamlined target discovery.
See Doc ID 1929507.1 for information about the plug-ins and patches required for Zero Data
Loss Recovery Appliance management, monitoring, and protected database management.
Main Pane
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Navigation
Pane
a ble
fe r
an s
n - t r
OEM 12c
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den OEM 13c
b t u
n i on is S
@ u e th
b e llo to us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
( a ak nse
ello
OraclebEnterprise lice Recovery Appliance Dashboard
Manager
k e emdashboard for the Current Status of the Backup Environment includes these sections:
The
a • Summary of current and recent activity
• Protected database backup/recovery issues
• Daily data sent / received
• Performance
• Media Manager status
• Replication status to / from
• Storage Location status
• Incidents and Events
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
Power-off sequence:
1. Check the status of Recovery Appliance services and shut them down.
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
3. Stop DBFS.
STATE=ONLINE on host1
$ $GRID_HOME/bin/crsctl status res ob_dbfs rep_dbfs $ $GRID_HOME/bin/crsctl stop res ob_dbfs rep_dbfs
NAME=ob_dbfs CRS-2673: Attempting to stop 'ob_dbfs' on 'host2'
TYPE=local_resource CRS-2673: Attempting to stop 'rep_dbfs' on 'host1'
TARGET=ONLINE , ONLINE CRS-2673: Attempting to stop 'ob_dbfs' on 'host1'
le
STATE=ONLINE on host1, ONLINE on host2 NAME=rep_dbfs CRS-2673: Attempting to stop 'rep_dbfs' on 'host2'
b
TYPE=local_resource CRS-2677: Stop of 'rep_dbfs' on 'host1' succeeded
a
TARGET=ONLINE , ONLINE CRS-2677: Stop of 'ob_dbfs' on 'host1' succeeded
STATE=ONLINE on host1, ONLINE on host2 CRS-2677: Stop of 'rep_dbfs' on 'host2' succeeded
CRS-2677: Stop of 'ob_dbfs' on 'host2' succeeded
fe r
4. Stop the Oracle Cluster Database.
an s
srvctl status database -d zdlradb
n - t r
no
Instance zdlradb1 is running on node host1
Instance zdlradb2 is running on node host2 $ srvctl stop database -d zdlradb
s a
5. Stop the Oracle Clusterware Stack.
h a
) ideฺ
m
# $GRID_HOME/bin/crsctl stop cluster -all
co t Gu
CRS-2673: Attempting to stop 'ora.crsd' on 'host1'
ฺ
CRS-2790: Starting shutdown of Cluster Ready Services-managed resources on ‘host1’
g
…
a n kn (CRS).
6. On each computer, stop the Cluster Ready Services
d e n
nb s Stu
# $GRID_HOME/bin/crsctl stop crs
o
CRS-2791: Starting shutdown of Oracle High Availability Services-managed resources on 'host2'
u n i h i
CRS-2673: Attempting to stop 'ora.crf' on 'host2'
CRS-2673: Attempting to stop 'ora.mdnsd' on ‘host2'
…
e t
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
e l
OraclebRecovery li
Appliance Power Sequence
k e emslide lists the recommended sequence for powering the Recovery Appliance off and on in
The
a non-emergency situations.
When performing either sequence, ensure that each step is completely finished before
moving to the next step. Failure to execute steps in the proper order may result in the
Recovery Appliance not functioning correctly.
When powering on the Exadata Storage Servers and the Recovery Appliance database
servers, power can be applied by pressing the power button at the front of each server, or
remotely using the ILOM interface for each server. The ILOM can be accessed by using the
web console, the command line interface (CLI), IPMI, or SNMP. For example, you can start a
server from the ILOM CLI by executing the start /SYS command.
To power the rack on or off, use the switches located on the PDUs, which are located at the
rear of the rack.
# shutdown -h -y now
a a
( cen
e l l o
OraclebRecovery li
Appliance Power Sequence
k e em Use the dcli utility to run the shutdown command on multiple servers simultaneously.
Note:
a Do not run dcli from a server that will be powered off by the command.
Power-on sequence:
• Rack, including network switches: Apply power for a few
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
OraclebRecovery li
Appliance Power Sequence
k e emcan use the Oracle ILOM interface to power on the Recovery Appliance servers remotely.
You
a To access Oracle ILOM, use the web console, the CLI, intelligent platform management
interface (IPMI), or simple network management protocol (SNMP).
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a n
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e a Damaged
Replacing li Physical Disk
b
k e em a physical disk, either a hard drive on a high-capacity cell or a flash drive on an
Replacing
a Extreme Flash cell, due to a problem or failure is probably the most likely hardware
maintenance operation that an Exadata Storage Server may require. Because Exadata uses
ASM redundancy, the procedure to replace a problem disk is quite simple.
The first step requires that you identify the problem disk. This could occur in several ways, as
given below:
• For a failed physical disk or a physical disk with poor performance, the blue ready-to-
remove LED is automatically switched on. For a physical disk with predictive failure
status, the blue ready-to-remove LED is automatically switched on after all grid disks
present on the physical disk are dropped from ASM and redundancy is fully restored. In
both cases, the blue LED indicates that it is safe to remove and replace the affected
disk.
• Hardware monitoring using ILOM may report a problem disk.
When a failed disk is detected, the Oracle ASM disks associated with the grid disks on the
physical disk are automatically dropped with the FORCE option, and an Oracle ASM rebalance
follows to restore the data redundancy. This process is known as proactive disk quarantine.
After you have identified the problem disk, you can replace it. When you remove the disk, you
will get an alert. When you replace a physical disk, the disk must be acknowledged by the RAID
controller before it can be used. This does not take a long time, and you can use the LIST
PHYSICALDISK command to monitor the status until it returns to NORMAL.
a b
The grid disks and cell disks that existed on the previous disk in the slot will be automatically re-
le
fe r
created on the new disk. If these grid disks were a part of an Oracle ASM disk group with
ans
t r
NORMAL or HIGH redundancy, they will be added back to the disk group and the data will be
n -
no
rebalanced based on disk group redundancy and the asm_power_limit parameter.
a
Recreating the ASM disk and rebalancing the data may take some
h a s time to complete. You can
) monitor
monitor the progress of these operations within ASM. You can
m d e ฺthe status of the disk as
co t You
reported by V$ASM_DISK.STATE until it returns to NORMAL.
ฺ i
ucan also monitor the
g G
rebalance progress by using GV$ASM_OPERATION.
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
a a
( cen
e l l o li
m b
a kee
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l o
e Removalli
Safe Disk l
b
em
eSometimes a disk must be replaced proactively, even though its blue ready-to-remove LED is
ak switched off. For example, the CALIBRATE command may identify a disk delivering
abnormally low throughput or I/Os per second (IOPS).
Nonetheless, removing the disk could lead to a disk group force dismount if the removed disk
has the only copy of some pieces of data, or a cell crashes due to loss of its OS if the disk is
the only system disk.
To mitigate this, Exadata release 11.2.3.3.0 adds a new command.
ALTER PHYSICALDISK ... DROP FOR REPLACEMENT checks whether the specified
physical disks can be removed safely, and if so, the disks are disabled and the blue ready-to-
remove LEDs are switched on. This command replaces checks that were previously required
to ensure that disks could be removed safely. By using this command, disk replacements are
simplified and problems are avoided.
In addition, ALTER PHYSICALDISK ... REENABLE FORCE re-enables normal physical disks.
This command can be used to re-enable physical disks in situations where
ALTER PHYSICALDISK ... DROP FOR REPLACEMENT was executed, but the administrator
does not want to remove the disks.
a ble
fe r
2 Power off 3 Replace the 4 Power on
an s
the cell. flash card.
n - t r
the cell.
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Replacing l l o
e a Damaged li Flash Card
b
k e em a damaged flash module is similar to identifying a damaged physical disk.
Identifying
a Hardware monitoring using ILOM or a drop in performance indicated by the CALIBRATE
command may indicate a problem. If a failed FDOM is detected, an alert is generated.
A bad flash module results in a decreased amount of flash memory on the cell. The
performance of the cell is affected proportional to the size of the flash memory lost, but the
database and applications are not at risk of failure.
As shown in the slide, a damaged flash module can also be reported by using the LIST
PHYSICALDISK DETAIL command. The slotNumber attribute shows the PCI slot and the
FDOM number. In this example, the status attribute indicates a device failure.
Although the PCI slots in an Exadata Storage Server support hot replacement, it is
recommended to power off the cell while replacing a damaged flash card.
After replacing the card and powering on the cell, no additional steps are required to re-create
the Smart Flash Cache and Smart Flash Log areas on the new flash modules.
a a
( cen
Moving l l o li One Cell to Another
eAll Disks from
b
k e emmay need to move all drives from one storage server to another. This may be necessary
You
a when there is a chassis-level component failure, or when troubleshooting a hardware
problem. To move the drives, perform the following steps:
1. If possible, back up /etc/hosts, /etc/modprobe.conf, and the files in
/etc/sysconfig/network and /etc/sysconfig/network-scripts. This is
mainly a precautionary step and is also useful if you want to move the disks back to the
original chassis.
2. If possible, deactivate the grid disks by using ALTER GRIDDISK ALL INACTIVE.
Also, ensure that the Oracle ASM disk_repair_time attribute is set sufficiently long
enough so ASM does not drop the disks before the grid disks can be activated in
another Exadata Storage Server.
3. Shut down the original server.
a a kb se t
l o ( cen
l
e Rescue
Cell Software
b li Procedure
em Storage Server maintains mirrored system areas on separate physical disks. If one
eExadata
ak system area becomes corrupt or unavailable, the mirrored copy is used to recover.
In the rare event that both system disks fail simultaneously, you must use the rescue
functionality provided on the CELLBOOT USB flash drive that is built into every Exadata
Storage Server. The rescue procedure may also be required to recover from a file system
corruption or a corrupt boot area.
It is important to note the following when using the rescue procedure:
• Use extreme caution when using this procedure, and pay attention to the prompts. The
rescue procedure can rewrite some or all of the disks in the cell. If this happens, you can
lose the contents of those disks. You should use the rescue procedure only with
assistance from Oracle Support Services.
• The rescue procedure does not destroy the contents of the data disks or the contents of
the data partitions on the system disks unless you explicitly choose to do so during the
rescue procedure.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
ake
me
bel
l o
a k
li
l l o
(a cens
u
be e to
n
@ use
b a n
nio this
k
S t
ฺ c
ud
ng ent
)
s
om Guid
ha eฺ
an
o n - t r an
sfe r a b
le
11
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Diagnostics and
Troubleshooting r a ble
s fe
- t r an
o n
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
Objectives
tasks
• Describe the main areas for troubleshooting
• Use diagnostic and troubleshooting utilities
• Describe Automatic Service Requests (ASRs)
• List Oracle Enterprise Manager (OEM) plug-in monitoring
r a ble
capabilities fe
n s
• Describe replication issues
n - tra
o
• List Media Manager issues an s
h a
) ideฺ
m
co t Gu
g ฺ
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li
m be
e
ake
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a a
( cen
l l o i areas where a performance issues can be identified.
1.
m be are fourlmain
There
e 2. It is important to identify the area where the problem was initially found and any
ake additional errors across the four main areas.
3. Having an original baseline to compare with the suspicious situation will help
performance investigations.
4. Record the sequence of the steps executed to reproduce the issue and their outcome:
Elapsed time, messages (warning, error, and so on).
5. Provide the information about end-to-end architecture and network topology.
Protected Database
• Connectivity to the Recovery Appliance (Setup, Configuration, and so on).
• Server configuration and capacity planning report (Exawatcher, Oswatcher).
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
em
ake
• Questions to formulate:
– Third-Party products?
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a a
( cen
l l o li questions:
e the following
Formulate
b
k e em a detailed problem description and describe the impact of the problem.
Provide
a • What Recovery Applinace version are you running?
• Do you have any third-party products installed on the Recovery Appliance? What are
they?
• Are there any network security attributes in your configuration: firewalls, ACL, etc.?
What are they? Who is the vendor? What is the configuration?
• Are you using any RMAN compression?
• Are you using any encryption? Are you using RMAN, TDE, or other vendor encryption
product? What is the configuration?
• Are you using Oracle *Net encryption? Are you using TCPS native or any external
authentication?
• Are you using tape backups and which product? If not OSB, did you engage the third-
party vendor?
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
le
… INFO : 0
b
2016/09/02 15:05:22 EDT: host2:Completed Collection ERROR : 0
a
2016/09/02 15:05:22 EDT: Completed collection of zip files. WARNING : 1
Logs are being collected to: $TFA_HOME/repository/collection_Wed_Sep_2_15_04_19_EDT_2016_node_all
$TFA_HOME/repository/collection_Wed_Sep_2_15_04_19_EDT_2016_node_all/host1.tfa_Wed_Sep_2_15_04_19_EDT_2016.zip
tfactl>
fe r
$TFA_HOME/repository/collection_Wed_Sep_2_15_04_19_EDT_2016_node_all/host2.tfa_Wed_Sep_2_15_04_19_EDT_2016.zip
[oracle@host1 ~]#
an s
• Diagnostic Assistant (DA) n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Collection
Diagnostic li
b
k e emTFA and the DA provide the official methods to gather information from ZDLRA for
The
a troubleshooting any service request.
The TFA is a diagnostic collection utility used to simplify diagnostic data collection on Oracle
Clusterware/Grid Infrastructure, RAC, and single-instance database systems.
TFA Collector Features:
• It simplifies diagnostic data collections.
• A single command performs clusterwide diagnostic collection for ALL components.
• Diagnostic files “trimmed” around the incident time.
• Collected diagnostics are consolidated on a single node.
• Increased efficiency of admin staff.
Oracle.
DA benefits:
• Multiple execution methods
• Menu-driven command line interface
• Web User Interface
• Direct command execution
• Enterprise Manager's Support Workbench
a b le
• Knowledge of the syntax of the various data collectors NOT needed fe r
ans
• Proactively troubleshoot by problem type, system components, common issues
n - t r
• Limited prompting (common values stored for re-use)
a no
a s
• Deployed via a single zip as well as being delivered within many product lines
h
m ) ideฺ
• Create a draft service request directly from the DA, without the need for Oracle
Configuration Manager (OCM) g ฺ co t Gu
n kn den
• Package the diagnostic collections and upload directly to an existing or draft service
a
request
i o nb s Stu
u n t h i
• Immediate access to diagnostic menus attached to a MOS Knowledge Article
l @ us
losupport e
k e
Please refer to the following
bGathering to(Docnotes for additional information: TFA Collector - Tool for
Enhanced Diagnostic
( a a n s e ID 1513912.2) and Diagnostic Assistant: Information
e
elloID 201804.2)
Center (Doc
b lic
k e em
a
a k
l o (a cens
el dataliregarding key software, hardware, and firmware versions, along with
Exachkbcollects
k e em
information about Recovery Appliance-specific configuration best practices.
a The output assists Recovery Appliance administrators to periodically review and cross-
reference the current data for key Recovery Appliance components against supported version
levels and recommended best practices.
Exachk is preinstalled on new Recovery Appliance systems. The latest updated version of
Exachk is available for download from My Oracle Support note 1070954.1. It can be executed
as desired and should be executed regularly as a part of the maintenance program for
Recovery Appliance.
Exachk is not a continuous monitoring utility and should not be considered as a replacement
for other monitoring or alerting tools, such as Enterprise Manager or ILOM.
It is non-intrusive and does not change anything on the Recovery Appliance apart from writing
output files and some small temporary files, which are deleted after they are used. It also
offers to configure SSH user equivalence if it is not configured.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
Exachkbproduces anliHTML report of findings, with the most important exceptions listed first by
k e em run.The
component. report is contained in a date-stamped zip file that is produced during each
Typically, administrators will transfer the exachk output ZIP file to a desktop
a exachk
computer and expand it there to review the exachk findings.
The exachk HTML report contains a summary of findings, along with relevant details. The
detailed entries typically describe the issues in detail and provide recommendations for
remediation. A link to documentation or a link to a note in My Oracle Support may also be
included to provide further information, especially if the recommendations are not
straightforward.
The slide shows an example of the output from the exachk HTML report.
For additional information please refer to the following support note: Oracle Exadata
Database Machine exachk or HealthCheck (Doc ID 1070954.1)
a k
l o (a cens
ExaWatcher
b li
el is a collection of perl and shell scripts designed to collect and archive OS and
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• OEM Diagnostic
Copyright (c) 2016, Oracle Corporation. All rights reserved.
Appliance plug-ins
SQL> SET LINESIZE 120
SQL> SELECT INCIDENT_ID, DB_UNIQUE_NAME, ERROR_CODE, ERROR_TEXT
2 FROM RASYS.RA_INCIDENT_LOG
3 WHERE STATUS = 'ACTIVE' AND SEVERITY = 'INTERNAL';
• Protected Databases
Diagnostic a b le
e r
$ rman target / catalog rauser11/welcome1@scao09ingest-scan3:1521/zdlra8 log=/tmp/rman_backup.log debug
f
s
trace=/tmp/rman_backup.trc
an
RMAN>
• Recovery Appliance t r
Diagnostic
$ sqlplus rasys/*****
SQL> EXEC DBMS_RA.DUMP(do_dpdump => TRUE);
o n -
an
$ ls -ltr /radump/*dat
-rw-r--r-- 1 oracle dba 24479534 Oct 2 07:47 /radump/radump_time_074708__ospid_112711_RADUMP.dat
• Tape Diagnostic s
ha eฺ
)
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
Items to
b l
eCollect: li
k e e1.m If the OEM is being used:
a a. From the Recovery Appliance and the protected database, get the patch level of
the agents. This can be obtained by running optach from the database home with
the lsinventory option and pointing to the EM agent plug-in directory.
b. Gather all internal incidents from the Recovery Appliance incident log querying the
RASYS.RA_INCIDENT_LOG table.
c. The patch level of the database plug-in and the Recovery Appliance plug-in can be
obtained by using opatch from the OMS home with the lsinventory option and
pointing to the OMS plug-ins directory.
2. From the Protected Database side, if the issue is between the protected database and
the ZDLRA:
a. Capture the output of the netstat -rna command
b. Capture the output of the opatch lsinventory -details for the both
database home and grid home (if the customer is using the GI)
to monitor the system’s resource utilization: CPU, Memory, IO, Network, etc.
g. If this is a performance problem, provide AWR report from the protected database
during the problematic period, this can be obtained through OEM or manually.
h. The spool files from querying the following views: gv$backup_async_io and
gv$backup_sync_io.
i. Network Performance test throughput result. This test can be performed by using
the qperf utility.
3. From the Recovery Appliance side (repeat the steps below on replica appliance if the a ble
fe r
problem is related to replication):
ans
a. Capture the output of the netstat -rna command.
n - t r
no
b. The Recovery Appliance version the customer is running, this can be obtained by
a
h a s
executing the following on any ZDLRA database node as the oracle user:
) ideฺ
$ORACLE_HOME/OPatch/opatch lspatches ( for the ZDLRA database & GI)
m
ฺ co t Gu
c. Capture the output of: cat $ORACLE_HOME/rdbms/install/zdlra/zdlra-
g
software-id
a n kn den
i o nb s Stu
d. As the root user, run: rpm -qa ra_automation
e. AWR dump from the u n t
problematic h i
period that will help to identify any issue on the
l lo@Database
Recovery Appliance
e
us side.
b e t o
( a ak noutput.
f. An radumper s e On the ZDLRA database nodes execute the
e l o
lDBMS_RA.DUMP
l i ce procedure.
m bg. Recovery Appliance performance reports from the problematic period. These can
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
through to resolution
– It leverages the EM incident and event notification system.
• Default and user-defined metrics trigger warnings or error messages based on
thresholds.
– EM collects metrics on key information, providing out-of-the box monitoring and alerting.
Protected Database Protected
Activity Summary Database
Recovery Status
a ble
Summary
fe r
Filter by an s
Backup Volume and
Incoming Data Rate n - t r
Performance Charts
a no
h a s
m ) ideฺ
g ฺ co t Gu Protected
n n
k de n Database
a
nb s Stu Status Filter
n i o i
u e t h
e l lo@ o us Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a kb se t
l o ( cen
l
e sections
The following
b li of the Home page show monitoring information:
ak ee•m Summary: This section shows the number of databases with no issues, with alerts, and
with warnings. In Cloud Control, an alert is an indicator that a particular metric condition
has been encountered. For example, an alert might indicate that a metric threshold has
been reached.
• Media Managers and Replication: These sections show the status of copy-to-tape and
Recovery Appliance replication services.
• Protected Database Issues: This section summarizes the backup status for protected
databases, and provides a category filter so you can view which databases are affected.
• Incidents and Events: This section displays incidents and events reported for the
Recovery Appliance and all associated targets. You can filter by target and category.
You can click the Summary link to drill down to the Incident Manager to view detailed
information about the incident.
a ble
Used Space: 2.5 TB Current RPO: < 1 sec !
fe r
an s
n - t r
a no
h a s Deduplication Ratio: 10:1
m ) ideฺ
Reserved Space: 7.9 TB
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Databases
The Protected li table displays information related to the database name, version,
b
em policy, database size, recovery window information, unprotected data window
eprotection
ak information, near-Zero Data Loss Recovery, errors/warnings, last complete backup,
replication and Copy-to-Tape.
You can view detailed information about the currently selected protected database by using
the tabs in the Protected Database Detail section below the Protected Databases table. The
table displays charts showing the status of each protected database on the Status tab, and a
list of Recovery Appliance incidents on the Incidents and Events tab.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e and Events
The Incidents li section shows all incidents, events, and warnings for a Recovery
b
em
eAppliance. Click any incident to open the Incident Manager page. Incident Manager provides,
ak in one location, the ability to search, view, manage, and resolve incidents and problems
affecting your environment.
[root@host1 ~]#
le
Imaging mode : fresh
Imaging status : success
r a b
Version
Image activation date
:
:
12.1.2.2.0.150917
2015-11-07 13:46:21 -0700
s fe
Imaging mode
Imaging status
:
:
patch
success
- t r an
[root@host1 ~]#
no n
s a
h a
) ideฺ
m
co t Gu
g ฺ
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Software l l o li
eis initially installed on Exadata storage and Recovery Appliance database servers by
b
k e em a Recovery Appliance-specific imaging process. The Exadata Storage Server patch
using
a bundles are also provided as software images.
Two utilities are provided to monitor the Recovery Appliance and Exadata Storage images:
• imageinfo: Displays information relating to the images currently installed on a server
• imagehistory: Displays historical information about all the images installed on the
server
The slide shows examples of the output displayed by imageinfo on a storage server and on
a database server.
There is significantly less output associated with a database server because database
servers have no concept of active and inactive images, and they do not contain a cell boot
USB device.
The behaviour and output of imageinfo can be modified by using a series of optional
parameters. Execute imageinfo –h to obtain a list of the available options.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a k
l o (a cens
el Request
Auto Service
b li (ASR) is a secure, scalable, and customer-installable software solution
emis available as a feature of your Oracle hardware warranty and Oracle Premier Support
ethat
ak for Systems. The ASR software helps to resolve problems faster by using auto-case
generation for server and storage systems when specific hardware faults occur.
The ASR functionality is available for Recovery Appliance in addition to all other Engineered
Systems. ASR automatically opens service requests with Oracle Support when specific
hardware faults occur either in the Exadata Storage Servers or in the Recovery Appliance
database servers. The ASR covers the following server components: CPUs, disk controllers,
disks, flash cards, flash modules, InfiniBand cards, memory modules, system boards, power
supplies, and fans.
FRU replaced by
Field Engineer
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
Oracle Support
Engineer
an
Management s
n - t r
System
a no
ASR
Manager
h a sService request (SR)
ASR Service
SNMP trap sent to
ASR Manager
Fault telemetry
securely m ) ideฺ created
transmitted to co u
Oracle gฺ t G
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li illustrates the ASR process. It shows how a service request is
e in the slide
The diagram
b
em
eautomatically opened by the ASR Manager after it receives an SNMP trap that is triggered by
ak a server hardware fault. Within Oracle Support Services, the ASR is acknowledged via email
and serviced according to normal procedures.
Customers must note that there are occasions when an SR may not be automatically filed.
For example, this can happen due to loss of connectivity to the ASR Manager. Customers can
continue to monitor their systems for faults and engage Oracle Support if they notice a fault
but do not receive notice that a service request has been automatically filed.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
• Benefits include:
– Efficient problem diagnosis
– Easier service request reporting
– Configuration information uploaded to Oracle either
automatically or manually
a ble
– Configuration information remains confidential. fe r
anthe s
• Oracle recommends the use of OCM in conjunction with
- t r
Recovery Appliance. non
h a sa
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
OCM automatically li
collects configuration information from your environment at regular
b
k e em This configuration information can be uploaded to My Oracle Support. This helps
intervals.
a Oracle maintain up-to-date information about your environment, diagnose support issues
more efficiently, and offer consistently better support outcomes. OCM is recommended for
use in conjunction with Recovery Appliance.
Oracle recommends that OCM is set up so that configuration information can be automatically
collected and uploaded to the Oracle Customer Configuration Repository and My Oracle
Support. Alternatively, information can be manually uploaded by an administrator.
No business or personal information is collected and uploaded, except for contact information,
which is used for handling transmission problems. All information is kept strictly confidential.
Note that OCM is separate and distinct from the Enterprise Manager Configuration
Management Pack.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li of OCM is typically performed as a part of the initial installation and
e configuration
Like ASR,
b
em
econfiguration of the Recovery Appliance. Customers can opt out of configuring OCM.
ak However, this is not recommended by Oracle.
For customers who want to configure OCM against an existing Recovery Appliance system,
the easiest way is to use the resources on the Collector tab in My Oracle Support. The slide
shows the interface presented to users, which they can use to get started with OCM.
The OCM collector is installed on the Recovery Appliance database servers. No OCM
components are installed on the Exadata Storage Servers.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
ake
me
bel
l o
a k
li
l l o
(a cens
u
be e to
n
@ use
b a n
nio this
k
S t
ฺ c
ud
ng ent
)
s
om Guid
ha eฺ
an
o n - t r an
sfe r a b
le
12
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a k
l o (a cens
bel li
em
ake
Objectives
Manager
• Use different Oracle-provided reports for the Recovery
Appliance
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
• Introduction to BI Publisher
• Using Oracle-Provided Reports
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a k
l o (a cens
BI Publisher,
b li Manager's reporting framework, makes information about your
el Enterprise
em environment available to audiences across your enterprise.
emanaged
ak Strategically, reports are used to present a view of enterprise monitoring information for
business intelligence purposes, but can also serve an administrative role by showing activity,
resource utilization, and configuration of managed targets.
IT managers can use reports to show availability of sets of managed systems. Executives can
view reports on availability of applications (such as corporate email) over a period of time.
Planning Ongoing
Setup and Configuration
Phase Maintenance
a k
l o (a cens
l perform
eyou
Typically,
b li reporting tasks in the following sequence:
k e e1.m During the planning phase, familiarize yourself with the monitoring and reporting tools
a available through Cloud Control.
2. Setup and Configuration tasks are automated when installing Recovery Appliance
plugins in Enterprise Manager Cloud Control. Pre-configured reports for Recovery
Appliance are available and ready to be used.
3. During the ongoing maintenance phase, review the reports, as needed.
- Review the Storage Capacity Planning Summary every week, using the Capacity
Planning Details to get more detailed information.
- Review the protected database reports as needed.
• Introduction to BI Publisher
• Using Oracle-Provided Reports
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a a
( cen
Through l l o li the Recovery Appliance provides pre-created BI Publisher reports that
eBI Publisher,
b
k e em you to meet the following goals:
enable
a • Ensure that the Recovery Appliance has sufficient storage space for its needs: By using
the capacity reports, you can plan for additional storage, reduce the number of new
protected databases added to the Recovery Appliance, or adjust protection policies so
that the aggregate recovery window space decreases.
• Ensure that the network is not overloaded: The capacity reports also indicate whether
the Recovery Appliance has maximized network capacity. In some cases, you can
reduce network load by redistributing network traffic more evenly throughout the day. If
network traffic is not distributed, and if network peaks are close to maximizing network
bandwidth, then you may need to adjust the backup window times of some protected
databases.
• Provide a good view of system performance and activity for service requests
Enterprise manager allows you to view reports interactively and/or schedule generation of
reports on a flexible schedule. You can choose for scheduled reports to be emailed to any
number of recipients.
Note: BI Publisher software included with Cloud Control can only access data in the
Enterprise Manager Repository, not data in the Recovery Appliance schema. Recovery
Appliance schema data is accessible through SQL queries of V$ and recovery catalog views.
a ble
fe r
ans
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
a k be e to
l o (a cens
bel li
e m
ake
• Introduction to BI Publisher
• Using Oracle-Provided Reports
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o i
e StoragelCapacity
Reviewbthe Planning Summary section to determine the storage growth rate
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e provideslan
This report i even finer granularity of information than the capacity planning
b
em Unlike the capacity planning report, the detailed report also has memory and IOPS
esummary.
ak information.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o li finer granularity of information than the capacity planning summary,
e also provides
This report
b
emincludes memory and IOPS information.
eand
ak The pie chart shows that over 18% of the databases are not meeting their service level
agreements. The table shows that out of 27 total databases, one is not meeting its recovery
window goal. Also, four databases are not within their unprotected data threshold.
• Introduction to BI Publisher
• Using Oracle-provided reports
Unauthorized reproduction or distribution prohibitedฺ Copyright© 2019, Oracle and/or its affiliatesฺ
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
l l o
e Databases
The Protected li section in the Recovery Appliance table displays charts showing
b
emstatus of each protected database on the Status tab.
ethe
ak You have two links to navigate to the Protected Database Report and to the Backup Report.
For example, the chart shows the Summary details and the data sent and received over the
past week by CUST01 database after making click the Protected Database Report link for the
CUST01 database.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
e l l o
Top 10bProtected li
Database reports use multiple dimensions: Top 10 Protected Databases by
emTransfer, Top 10 Databases by Backup Data, Top 10 Databases by Replication Data,
eData
ak and Top 10 Databases by Copy-to-Tape Data.
a ble
fe r
an s
n - t r
a no
h a s
m ) ideฺ
g ฺ co t Gu
a n kn den
i o nb s Stu
u n t h i
l lo@ o us e
e
kb se t
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a a
( cen
Use the
l o
eView Backupli Report page to:
l
b
ak ee•m Display lists of backup jobs that are known to the database through the information
recorded about them in the database control file
• Filter the backup jobs that display by Status, Start Time, and Type
• Link to Job Details and Job Summaries
You can drill down to a summary page of the backup job by clicking on the Status of the job in
the Result table where you can view the contents of the output log.
a ble
fe r
an s
n - t r
o
s an
) ha eฺ
ฺ c om Guid
n k ng ent
n b a
S t ud
u nio this
l l o @ use
be e to
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.
a k
l o (a cens
bel li
em
ake
ake
me
bel
l o
a k
li
l l o
(a cens
u
be e to
n
@ use
b a n
nio this
k
S t
ฺ c
ud
ng ent
)
s
om Guid
ha eฺ
an
o n - t r an
sfe r a b
le