cisco prime optical high availability installation …...installation guide for use with cisco prime...

Cisco Prime Optical 10.5 High Availability Installation GuideFor Use with Cisco Prime Optical High Availability Agent November 30, 2015

Cisco Systems, Inc.www.cisco.com

Cisco has more than 200 offices worldwide. Addresses, phone numbers, and fax numbers are listed on the Cisco website at www.cisco.com/go/offices.

http://www.cisco.com

http://www.cisco.com/go/offices

THE SPECIFICATIONS AND INFORMATION REGARDING THE PRODUCTS IN THIS MANUAL ARE SUBJECT TO CHANGE WITHOUT NOTICE. ALL STATEMENTS, INFORMATION, AND RECOMMENDATIONS IN THIS MANUAL ARE BELIEVED TO BE ACCURATE BUT ARE PRESENTED WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED. USERS MUST TAKE FULL RESPONSIBILITY FOR THEIR APPLICATION OF ANY PRODUCTS.

THE SOFTWARE LICENSE AND LIMITED WARRANTY FOR THE ACCOMPANYING PRODUCT ARE SET FORTH IN THE INFORMATION PACKET THAT SHIPPED WITH THE PRODUCT AND ARE INCORPORATED HEREIN BY THIS REFERENCE. IF YOU ARE UNABLE TO LOCATE THE SOFTWARE LICENSE OR LIMITED WARRANTY, CONTACT YOUR CISCO REPRESENTATIVE FOR A COPY.

The Cisco implementation of TCP header compression is an adaptation of a program developed by the University of California, Berkeley (UCB) as part of UCB’s public domain version of the UNIX operating system. All rights reserved. Copyright © 1981, Regents of the University of California.

NOTWITHSTANDING ANY OTHER WARRANTY HEREIN, ALL DOCUMENT FILES AND SOFTWARE OF THESE SUPPLIERS ARE PROVIDED “AS IS” WITH ALL FAULTS. CISCO AND THE ABOVE-NAMED SUPPLIERS DISCLAIM ALL WARRANTIES, EXPRESSED OR IMPLIED, INCLUDING, WITHOUT LIMITATION, THOSE OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OR ARISING FROM A COURSE OF DEALING, USAGE, OR TRADE PRACTICE.

IN NO EVENT SHALL CISCO OR ITS SUPPLIERS BE LIABLE FOR ANY INDIRECT, SPECIAL, CONSEQUENTIAL, OR INCIDENTAL DAMAGES, INCLUDING, WITHOUT LIMITATION, LOST PROFITS OR LOSS OR DAMAGE TO DATA ARISING OUT OF THE USE OR INABILITY TO USE THIS MANUAL, EVEN IF CISCO OR ITS SUPPLIERS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.

Cisco and the Cisco logo are trademarks or registered trademarks of Cisco and/or its affiliates in the U.S. and other countries. To view a list of Cisco trademarks, go to this URL: www.cisco.com/go/trademarks. Third-party trademarks mentioned are the property of their respective owners. The use of the word partner does not imply a partnership relationship between Cisco and any other company. (1110R)

Any Internet Protocol (IP) addresses and phone numbers used in this document are not intended to be actual addresses and phone numbers. Any examples, command display output, network topology diagrams, and other figures included in the document are shown for illustrative purposes only. Any use of actual IP addresses or phone numbers in illustrative content is unintentional and coincidental.

© 1999–2015 Cisco Systems, Inc. All rights reserved.

http://www.cisco.com/go/trademarks

C O N T E N T S

C H A P T E R 1 Installation Overview 1-1

Local Redundancy Configuration 1-1

Single-Node Cluster Geographical Redundancy Configuration 1-3

Prime Optical Integration with Cisco Prime Central 1-6

What’s New in Prime Optical High Availability Guide 10.5 1-7

Workflow for Installation and Upgrade 1-7

C H A P T E R 2 Installation Requirements 2-1

Hardware Requirements 2-1

Server Configuration 2-1

Generic Disk Storage Array Configuration 2-3

Additional Hardware for a Single-Node Site 2-5

Additional Hardware for a Dual-Node Site 2-6

Software and Support Requirements for a Local Redundancy Configuration 2-7

Software and Support Requirements for a Geographical Redundancy Configuration 2-8

VMware Support 2-8

Overview of Prime Optical Commands 2-9

Overview of Sudo Commands 2-10

C H A P T E R 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Configuration on a Linux Server 3-1

Geographical Redundancy Configuration Checklists 3-1

Geographical Redundancy Configuration Preinstallation Checklist 3-2

Geographical Redundancy Configuration Installation Checklist 3-2

Setting the Environment for Installation 3-4

Installing Oracle 12c (for Not Embedded Database Setup) 3-5

Installing the Prime Optical Server with Embedded Database Setup 3-5

Verifying Packages 3-9

Disabling the Prime Optical Service and Configuring SSH 3-12

Installing the HA Package (HA RPM Files) 3-13

Configuring Oracle Data Guard 3-15

iiiCisco Prime Optical 10.5 High Availability Installation Guide

Contents

Configuring RMAN (Including Flashback) 3-16

Editing the ODG Configuration File 3-17

Generating the ODG Configuration File 3-20

Configuring the Primary Server 3-21

Configuring the Mount Point to Be Backed Up 3-21

Performing a Cold Backup 3-22

Configuring the Standby Server 3-22

Configuring a Backup on the Primary and Standby Servers 3-23

Editing the Agent Configuration File 3-23

Editing the Logging Configuration File 3-25

Editing the RHCS Configuration File 3-26

Starting the Cluster 3-29

Starting RHCS Services 3-30

Starting ctm_services on the Primary Server 3-30

Verifying Cluster Status on the Standby Node 3-31

Enabling Flashback on the Standby Node 3-31

Supporting Installation Information 3-32

HA Directory Structure 3-32

HA Bin Directory 3-33

HA Configuration Directory 3-34

HA Install Directory 3-34

HA Log Directory 3-34

HA Template Directory 3-34

GEO Directory Structure 3-35

GEO Bin Directory 3-35

GEO Configuration Directory 3-35

GEO Install Directory 3-35

GEO Log Directory 3-35

Maintaining and Monitoring Cluster Activities 3-35

GEO Manager 3-36

Installing GEO Manager 3-37

Configuring GEO Manager 3-38

Configuration Parameters for Geo Manager 3-39

Starting, Stopping, and Checking the GEOService Service 3-40

Running the GEO Manager Console 3-40

ODG Console 3-42

Maintaining the Cluster 3-44

Maintaining and Optimizing Flashback 3-44

Freezing or Unfreezing the Cluster 3-46

ivCisco Prime Optical 10.5 High Availability Installation Guide

Contents

Forcing a Failover 3-46

Realigning the Servers After a Forced Failover 3-47

Backup/Restore 3-47

C H A P T E R 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Linux Server 4-1

Local Redundancy Configuration Checklists 4-1

Preinstallation Checklist 4-2

Local Redundancy Configuration Installation Checklist 4-2

Setting the Environment for Installation 4-3

Verifying Packages and yum setup 4-4

Installing Oracle 12c (for Oracle Not Embedded Database Setup) 4-6

Installing Prime Optical 4-7

Installing the HA Package 4-11

Editing the Agent Configuration File 4-12

Editing the Logging Configuration File 4-13

Editing the RHCS Configuration File 4-14

Starting the Cluster 4-16

Starting RHCS Services 4-16

Starting ctm_services on the Primary Server 4-17

Verifying Cluster Status on the Standby Node 4-17

Installing the Multicast Heartbeat Services Package 4-18

Editing the multicastConfig.ini File 4-19

Editing the logMulticast.ini File 4-19

Starting the Multicast HeartBeat Services 4-20

Configuring the MultiCast Heartbeat Services on the Secondary Server 4-20

C H A P T E R 5 Upgrading Cisco Prime Optical in an High Availability Environment Using the Data Migrator Wizard 5-1

Overview of the Upgrade Process Using the Data Migrator Wizard 5-2

Installing the Data Migrator Wizard 5-2

Using the Data Migrator Wizard in a Geographic Redundancy, HA Environment on Linux 5-4

Using the Data Migrator Wizard in a Dual-Node Cluster, Local Redundancy, HA Environment on Linux 5-4

Using the Data Migrator Wizard to Export the Database 5-5

Using the Data Migrator Wizard to Import Your Data 5-6

Uninstalling the Data Migrator Wizard 5-9

vCisco Prime Optical 10.5 High Availability Installation Guide

Contents

A P P E N D I X A Installing and Patching Oracle A-1

Oracle Prerequisites A-1

Creating a New Oracle User A-1

Updating an Existing Oracle User A-2

Setting Shell Limits for an Oracle User A-2

Installing the Oracle 12c Software with the Response File (*.rsp) Provided by Cisco A-3

Downloading and Installing the Required Oracle 12c Patch for 64-Bit Linux Platforms A-5

viCisco Prime Optical 10.5 High Availability Installation Guide

Cisco Pri

C H A P T E R 1
Installation Overview
The HA solution for Cisco Prime Optical is available in local redundancy and geographical redundancy configurations for Linux. For information about the different HA configurations available for Prime Optical, including hardware configuration and network diagrams, see the following sections:

• Local Redundancy Configuration, page 1-1

• Single-Node Cluster Geographical Redundancy Configuration, page 1-3

• Prime Optical Integration with Cisco Prime Central, page 1-6

• What’s New in Prime Optical High Availability Guide 10.5, page 1-7

• Workflow for Installation and Upgrade, page 1-7

Note The servers in the following figures contain mirrored internal root disks. The servers must be the same make, model, and storage capacity. The 100BASE-T heartbeat cables between the primary and secondary servers in the dual-node configuration are cross-over (reversing) Ethernet cables. We recommend that you use a fault-tolerant, switched/routed network for communication with the HA servers.

Local Redundancy ConfigurationThe local redundancy configuration provides an automatic failover solution to specific software and single hardware failures without the need to reconfigure IP addresses on your switched/routed network.

The following figures illustrate a local redundancy HA configuration on Prime Optical for Linux environments.

• Figure 1-1 is a network diagram of Prime Optical in a local redundancy HA configuration.

• Figure 1-2 illustrates the hardware configuration of Prime Optical in a local redundancy HA configuration.

1-1me Optical 10.5 High Availability Installation Guide

Chapter 1 Installation Overview Local Redundancy Configuration

Figure 1-1 Prime Optical in a Local Redundancy HA Configuration

In a local redundancy configuration (Figure 1-1), there is a single cluster consisting of a primary and a secondary server.

Prime Opticalclients

PrimaryserverFC-AL

Secondaryserver

Disk array

Disk arraysubsystem

Disk array

100BASE-T

100BASE-T100BASE-T

100BASE-THeartbeat

Serviceprovider's OSS

Switch/routernetwork

ONS 15305

ONS 15327 ONS 15327

ONS 15600SONET

ONS 15454SONET

ONS 15454SONET

3445

45

ONS 155xx

1-2Cisco Prime Optical 10.5 High Availability Installation Guide

Chapter 1 Installation Overview Single-Node Cluster Geographical Redundancy Configuration

Figure 1-2 Local Redundancy

In a local redundancy configuration, the primary and secondary servers share a common disk array (that is, the servers are both connected to a single disk array). This single disk array is where the Prime Optical application, Oracle application, and Prime Optical database are installed.

The heartbeat link connects the primary and secondary servers to monitor the health of both servers. The heartbeat link is achieved between the primary and secondary servers using two cross-over Ethernet cables. During normal operation, the Prime Optical and Oracle applications run on the primary server (normally the active server). If a fault occurs on the primary server, the Prime Optical and Oracle applications run on the secondary server; at this stage, the secondary server becomes the active server.

The primary server can be switched back to be the active server after the fault on the primary server is fixed.

Single-Node Cluster Geographical Redundancy ConfigurationThe geographical redundancy configuration allows you to locate two Prime Optical instances at geographically remote sites. One server instance is active; the other server instance is standby. The HA agent switches to the standby Element Management System (EMS) instance if an unrecoverable failure occurs on the active EMS instance.

In a single-node cluster geographical redundancy configuration, there are two clusters with different names (one on each node), each containing a server.

Both of the nodes status must be as:

Primary server: HA1-105.cisco.com

Cluster Status for HA1_CLUSTER @ Wed Feb 11 17:40:18 2015Member Status: Quorate

Member Name ID Status------ ---- ---- ------ha1-105.cisco.com 1 Online, Local, rgmanager

Service Name Owner (Last) State------- ---- ----- ------ -----service:ctm_heartbeat ha1-105.cisco.com startedservice:ctm_monitoring ha1-105.cisco.com started

Primary cluster (dual-node cluster)

SecondaryPrimary

Heartbeat link between servers

Fiber channellinks

Fiber channellinksDisk

array

1808

95



service:ctm_service ha1-105.cisco.com started

Standby Server: HA2-105.cisco.com

Cluster Status for HA2_CLUSTER @ Wed Feb 11 17:42:26 2015Member Status: Quorate


Service Name Owner (Last) State------- ---- ----- ------ -----service:ctm_heartbeat ha2-105.cisco.com startedservice:ctm_monitoring ha2-105.cisco.com startedservice:ctm_service (ha2-105.cisco.com) disabled

The following figures illustrate a geographical redundancy HA configuration on Prime Optical:

• Figure 1-3 is a network diagram of Prime Optical in a geographical redundancy HA configuration for Linux.

• Figure 1-4 illustrates the hardware configuration for Prime Optical in a single-node geographical redundancy HA configuration.



Figure 1-3 Prime Optical in a Geographical Redundancy HA Environment for Linux

Geographic redundancy sited lP i i i l

Prime Opticalclients

Serviceprovider's OSS

Secondarylocal server

Disk array

Geographic redundancy siteor secondary cluster

Primarylocal server

Disk array

Primary site or primary cluster

Replicationlink

Switch/routernetwork

713

ONS 15305

ONS 15327 ONS 15327

ONS 15600SONET

ONS 15454SONET

ONS 15454SONET

ONS 155xx


Chapter 1 Installation Overview Prime Optical Integration with Cisco Prime Central

Figure 1-4 Geographical Redundancy (Single-Node Cluster)

In a single-node cluster geographical redundancy configuration, the clusters do not share a disk array (that is, the servers are connected to different disk arrays and do not share data). The Prime Optical application, Oracle application, and Prime Optical database are installed on each cluster. The Prime Optical database contains identical data on both clusters due to the volume replication (Oracle Active Data Guard on Linux).

The heartbeat link connects the primary and secondary clusters to monitor the health of both clusters. The heartbeat link is realized between the primary and secondary clusters through the WAN. The protocol used for the heartbeat is TCP/IP. During normal operation, the Prime Optical and Oracle applications run on the primary cluster (normally the active cluster). If a fault occurs on the primary cluster, the Prime Optical and Oracle applications can run on the secondary cluster; at this stage, the secondary cluster becomes the active cluster.

The primary cluster can be switched back manually to be the active cluster after the fault on the primary cluster is fixed.

The Replication link manages the Oracle Data Guard (ODG) communication between the active database (read/write) and standby database (read only).

Prime Optical Integration with Cisco Prime CentralAfter Prime Optical has been installed in standalone mode, you still have the option to integrate it with Prime Central. When Prime Optical is installed as part of the Prime Central suite, you can launch Prime Optical from the Prime Central portal. For more information about the Prime Central portal, see http://www.cisco.com/c/en/us/support/cloud-systems-management/prime-central/tsd-products-support-series-home.html.

Note If you are installing Prime Optical with Prime Central, DNS must be enabled on the machine where Prime Optical is being installed. If the machine cannot resolve hostnames into IP addresses through DNS, Prime Optical registration will fail.

To integrate Prime Optical with Prime Central after Prime Optical has been installed in standalone mode, as the root user, run the ./DMIntegrator.sh script. Follow the procedure described in "Configuring Domain Managers as Suite Components” in the Cisco Prime Central 1.1 Quick Start Guide.

Fiberchannel

links

Primaryserver

Replication link

Heartbeatlink betweenclusters overWAN

Primary cluster(single-node cluster)

Disk array

Fiberchannel

links

Primaryserver

Secondary cluster(single-node cluster)

Disk array

3705

79


http://www.cisco.com/c/en/us/support/cloud-systems-management/prime-central/tsd-products-support-series-home.html

http://www.cisco.com/c/en/us/support/cloud-systems-management/prime-central/tsd-products-support-series-home.html

http://www.cisco.com/c/en/us/support/cloud-systems-management/prime-central/products-installation-guides-list.html

Chapter 1 Installation Overview What’s New in Prime Optical High Availability Guide 10.5

What’s New in Prime Optical High Availability Guide 10.5

Workflow for Installation and UpgradeThe following process describes the typical HA solution for Prime Optical installation workflow.

Note See Cisco Prime Optical Supported Devices for the NE software versions that are supported. Verify NE support before updating the software image on an NE.

1. Select which hardware configuration you will be installing. See Local Redundancy Configuration, page 1-1, Single-Node Cluster Geographical Redundancy Configuration, page 1-3.

2. Check your system requirements before you install Prime Optical or upgrade select data from an earlier release. See Chapter 2, “Installation Requirements.”

3. To ensure a successful installation or upgrade, print out the checklists provided for your Linux installation scenario. These checklists are provided in Chapter 3, “Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Configuration on a Linux Server.” The checklists are for your reference throughout the installation process. You might find it helpful to check off each task as you complete it. For upgrading selected data from a previous CTM or Prime Optical release, follow the steps described in Chapter 5, “Upgrading Cisco Prime Optical in an High Availability Environment Using the Data Migrator Wizard.”

Note Before performing a fresh installation, remove all previous Cisco Transport Manager or Prime Optical installations.

Table 1-1 Whats’s New in Prime Optical High Availability Guide 10.5

Feature Update Description For More information, See...

Support for RHEL-6.5 and 6.6 Prime Optical 10.5 supports RHEL 6.5 and 6.6 version.

Hardware Requirements, page 2-1

Support for the latest CPO-10.5 version

Ability to support HA Geographical redundancy and local redundancy with latest CPO-10.5 server.

• Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Configuration on a Linux Server, page 3-1

• Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Linux Server, page 4-1

HA package version HA package version is updated to HA-1.5-3.x86_64.rpm

Installing the HA Package, page 4-11


http://www.cisco.com/c/en/us/support/cloud-systems-management/prime-optical/products-device-support-tables-list.html

Chapter 1 Installation Overview Workflow for Installation and Upgrade


Cisco Pri

C H A P T E R 2
Installation Requirements
This chapter describes the hardware and software components required to install Cisco Prime Optical in an HA environment. It contains the following sections:

• Hardware Requirements, page 2-1

• Software and Support Requirements for a Local Redundancy Configuration, page 2-7

• Software and Support Requirements for a Geographical Redundancy Configuration, page 2-8

• Overview of Prime Optical Commands, page 2-9

• Overview of Sudo Commands, page 2-10

Note Although Cisco makes every attempt to ensure the availability of third-party hardware and software platforms specified for Prime Optical, Cisco reserves the right to change or modify system requirements due to third-party vendor product availability or changes that are beyond Cisco’s control. Part numbers are subject to change.

Hardware RequirementsTo install Prime Optical in an HA environment, you must configure a fault-tolerant storage array. This section describes an installation that uses a fault-tolerant storage array for small, medium, large, and high-end configurations.

Note Prime Optical 10.5 is compatible with Linux RHEL 6.5 and 6.6.

Server ConfigurationThe Prime Optical server runs on any of the platforms listed in Table 2-1.

Table 2-1 Supported Platforms and Operating Systems

Platform Operating System

Cisco UCS B series Red Hat Enterprise Linux (RHEL) 6.5 and 6.6, server 64-bit

Cisco UCS C series


Chapter 2 Installation Requirements Hardware Requirements

Note It is possible to use and configure disk arrays for database storage with any type of Redundant Array of Independent Disks (RAID) layout. You can do so only if the disk throughput is equal to or higher than the throughput calculated for internal disks. The minimum RAID configuration is 1+0.

Cisco tests the simulated network configurations listed in Table 2-3. Your setup and performance might vary depending on the size of your network and the usage pattern of management tasks.

The following list describes the server configuration parameters and the effect of changes in each parameter on the maximum number of NEs the server can manage:

• Network Size—If the network size increases, more resources are reserved for the higher number of NEs to be managed.

• RAM—The server can manage more NEs as the RAM increases. The server can manage fewer NEs as the RAM decreases.

Table 2-2 Minimum Processor Requirements

Network Size Cisco UCS B or UCS C Series

Small 2 x Intel Xeon 5620

Medium 2 x Intel Xeon 5620

Large 2 x Intel Xeon 5620

High end 2 x Intel Xeon 5640

Table 2-3 Minimum Memory Requirements for the Prime Optical Server Installation

Network Size Oracle Database Type RAM

No. of Network Partitions1

1. In Prime Optical 10.5, a single NE service (or a single network partition) can support up to 750 nodes.

A high-end network supports any combination of up to 5000 nodes, 8 network partitions, and 750 nodes per network partition. For example, 6 NE services with 750 nodes + 1 NE service with 500 nodes (6 x 750 + 1 x 500 = 5000).

A large network supports any combination of up to 2000 nodes, 4 network partitions, and 750 nodes per network partition. For example, 4 NE services with 500 nodes (4 x 500), 2 NE services with 750 nodes + 1 NE service with 500 nodes (2 x 750 + 1 x 500).

Maximum No. of NEs2

2. These numbers assume you are using fully equipped ONS 15454 MSPP nodes (for example, 12 OC-48, 2 TCC, and 1 XC10G with up to 2 links and 48 SONET cross-connections per node). Note that these numbers should be used as a guideline, and vary depending on the software and hardware configuration of your NEs.

Small Enterprise Edition 8 GB 1 200

Medium Enterprise Edition 16 GB 1 500

Large Enterprise Edition 32 GB 4 2000

High end Enterprise Edition 96 GB 8 5000



• Number of Network Partitions—For each network partition, the server reserves resources for the higher number of NEs to be managed. The server can manage more NEs as the number of network partitions increases. The server can manage fewer NEs as the number of network partitions decreases.

Generic Disk Storage Array ConfigurationTable 2-4 lists the disk partition requirements for a small-to-high-end generic storage array.

The optional database partitions /db01_rd and /db02_rd described in the Cisco Prime Optical 10.5 Installation Guide are not supported in HA geographical redundancy configurations.

Note the following disk array constraints:

• The disk array should support at least RAID 1+0.

Table 2-4 Configuration for a Small-to-High-End Single- or Dual-Node Generic Disk Storage Array—PM Collection

Enabled in a Linux Environment

Network Size Disk Space root swap /oracle /cisco1

1. 10 GB is required for cisco, but 20 GB is strongly recommended.

Table 2-5 show the disk array configuration when PM collection is enabled and disabled in a Linux environment with an embedded database installation.

/db01 /db02 /db03 /db04 /db05/ctm_backup

Small 101 GB (min)

251 GB (max)

12 GB 12 GB 10 GB 20 GB 12 GB 6 GB 40 GB 30 GB 8 GB 101 GB

Medium 268 GB (min)

476 GB (max)


Large 478 GB (min)

872 GB (max)


High end 890 GB (min)

1648 GB (max)



Disabled in a Linux Environment

Network Size Disk Space root swap /oracle /cisco1 /db01 /db02 /db03 /db04 /db05

/ctm_backup

Small 95 GB (min)

135 GB (max)


Medium 130 GB (min)

192 GB (max)


Large 186 GB (min)

276 GB (max)



http://www.cisco.com/c/en/us/support/cloud-systems-management/prime-optical/products-installation-guides-list.html

http://www.cisco.com/c/en/us/support/cloud-systems-management/prime-optical/products-installation-guides-list.html


Figure 2-1 and Figure 2-2 shows the hardware configuration of a small-to-high-end generic storage array at single-node and dual-node cluster sites.

Figure 2-1 Hardware Configuration of a Small-to-High-End Generic Disk Array for a Single-Node

Site

High end 282 GB (min)

418 GB (max)


1. 10 GB is required for cisco, but 20 GB is strongly recommended.


Disabled in a Linux Environment (continued)

3027

13

Primary server

FC 1

FC 2

Interconnect cables

Generic disk array

FC0

FC1

FC4

FC5



Figure 2-2 Hardware Configuration of a Small- and Medium-to-High-End Generic Disk Array for

a Dual-Node Site

Additional Hardware for a Single-Node SiteTo ensure redundancy for the network connection, verify that each server has at least two network interface cards (NICs).

To ensure redundancy for the disk array connections, each server requires either one fiber channel, dual-port host bus adapter (HBA) or two fiber channel, single-port HBAs (preferred).

You must have two fiber cables per server to connect the server to the disk array.

3445

48

Primary server

FC 1

FC 2

Secondary server

FC 1

FC 2

Generic disk array

FC0

FC1

FC4

FC5



Figure 2-3 shows the HA solution in a multiple NIC configuration for a single-node site.

Figure 2-3 Multiple NIC Configuration for a Single-Node Site

Note A fault-tolerant switched/routed network is recommended for communication with the HA servers.

Additional Hardware for a Dual-Node SiteTo ensure redundancy for the network connection, verify that each server has at least two NICs. The local heartbeat requires at least two additional NICs per server, for a total of at least four NICs per server.

To ensure redundancy for the disk array connections, each server requires either one fiber channel, dual-port HBA or two fiber channel, single-port HBAs (preferred).

You must have two fiber cables per server to connect the server to the disk array. The local heartbeat requires two cross-over Ethernet cables.

Note A fault-tolerant, switched/routed network is recommended for communication with the HA servers.

Figure 2-4 illustrates the HA solution in an NIC configuration for a dual-node site for Linux.

Switch/router network

3027

14

interface1

interface0

systemA

Primary server


Chapter 2 Installation Requirements Software and Support Requirements for a Local Redundancy Configuration

Figure 2-4 NIC Configuration for a Dual-Node Site for Linux

Note eth1 and eth5 are bonded in bond0. eth0 and eth4 are bonded in bond1.

Software and Support Requirements for a Local Redundancy Configuration

Table 2-6 lists the software that is required to install Prime Optical in a local redundancy HA configuration.

eth0

eth1

eth2

eth3

eth4

eth5

eth6

eth7

eth0

eth1

eth2

eth3

eth4

eth5

eth6

eth7

Heartbeat no. 1

Heartbeat no. 2

100BASE-T Ethernet hub/switch

systemA systemB

Switch/router network

Primary server Secondary server

100BASE-T Ethernet hub/switch

3445

62

Table 2-6 Software Requirements for Prime Optical in a Local Redundancy HA Configuration for Linux

Software Configuration Quantity

Cisco Prime Optical 10.5

Note If you install Prime Optical GateWay/CORBA, an optional CORBA-based interface that provides higher-layer management systems with information for all supported NEs, you must purchase a separate license.

1

Linux Red Hat Enterprise 6.5 and 6.6, 64-bit 1

Red Hat Cluster Suite (RHCS) 6.5 and 6.6 1


Chapter 2 Installation Requirements Software and Support Requirements for a Geographical Redundancy Configuration

Software and Support Requirements for a Geographical Redundancy Configuration

Table 2-7 lists the software and external support requirements for a local redundancy configuration.

The requirements for a geographical redundancy HA configuration for Linux are the same requirements as listed in the local redundancy HA configuration (Table 2-6).

VMware SupportPrime Optical can run on a virtual machine (VM) on VMware vSphere version 5.5 and 6.0. VM RAM. Disk and partitioning requirements must comply with the same requirements for a nonvirtualized setup based on system size. Do the following:

Step 1 Assign RAM as listed in Server Configuration, page 2-1.

Step 2 Follow disk space and partition requirements as listed in Generic Disk Storage Array Configuration, page 2-3.

Step 3 Configure VM vCPU settings. Refer to VMware VSphere performance guidelines for information on vCPU settings.

Cisco Prime Optical HA Agent Release 1.5-3 1

Oracle 12c software (Oracle Not Embedded database setup), plus the following patches:

• January 2011 Patch Set Update (PSU)

• 10404530

1

Table 2-6 Software Requirements for Prime Optical in a Local Redundancy HA Configuration for Linux (continued)


Table 2-7 Software Requirements for Prime Optical in a Geographical Redundancy HA Configuration for Linux


Cisco Prime Optical 10.5

Note If you install Prime Optical GateWay/CORBA, an optional CORBA-based interface that provides higher-layer management systems with information for all supported NEs, you must purchase a separate license.

1

Linux Red Hat Enterprise 6.5 and 6.6, 64-bit 1

RHCS 6.5 and 6.6 1

Cisco Prime Optical HA 1.5-3 Release (Oracle Not Embedded database setup) 1

Oracle 12c software (Oracle Not Embedded database setup), plus the following patches:

• January 2011 Patch Set Update (PSU)

• 10404530

1


Chapter 2 Installation Requirements Overview of Prime Optical Commands

As a general guideline, it is suggested to configure 2vCPU for every 500 nodes. This is based on a UCS C210 M2 machine, with a dual-socket, multicore 2.93 GHz processor in a medium network Prime Optical setup during Cisco internal validation.

Step 4 Confirm that the /etc/hosts file exists inside the guest OS of the VM and that the hostname (with and without a domain) is associated to the proper IP address and not to the loopback IP 127.0.0.0.

Example of a correct entry in the /etc/hosts file:

127.0.0.1 localhost localhost.localdomain::1 localhost localhost.localdomain10.58.235.252 ha1-105 ha1-105.cisco.com ha1-105-hb.cisco.com10.58.235.253 ha2-105 ha2-105.cisco.com ha2-105-hb.cisco.com

Example of an incorrect entry in the /etc/hosts file:

127.0.0.1 ha1-105 localhost.localdomain localhost::1 localhost6.localdomain6 localhost6

where the hostname ha1-105 is associated to the loopback address 127.0.0.1.

Note The file can be wrong if the VM OS was installed without network connectivity.

Overview of Prime Optical CommandsTable 2-8 describes the Prime Optical commands that can be used during installation and troubleshooting.

Table 2-8 Prime Optical Commands

Command Description

opticalctl start A complete set of administrative command scripts is added to the application during installation. One command automatically starts the Prime Optical server processes every time the server is started. The server processes can also be started or stopped manually as necessary; the scripts are located in the /opt/CiscoTransportManagerServer/bin directory.

The opticalctl start command sets the appropriate environment variables and starts the Prime Optical server. The amount of time it takes for the Prime Optical server to start varies based on the number of NEs in the configuration and the size of the database. Use opticalctl start only when the Prime Optical server has stopped.

It can take from 0 to 5 minutes for the server processes to start after the opticalctl start command has finished execution. This is because NE services and gateway services (if enabled) are still initializing for all of the NEs that are deployed.

Five minutes after entering the opticalctl start command, enter the opticalctl status command. By this time, the NE service corresponding to all the deployed NEs should have started.

opticalctl stop The opticalctl stop command stops the Prime Optical server gracefully. The stop procedure shuts down the server and cleans all memory and connections. The overall process takes approximately 5 minutes.


Chapter 2 Installation Requirements Overview of Sudo Commands

Overview of Sudo Commands

Note Only a Prime Optical root user (chosen during installation) can use Sudo commands.

Sudo software (freeware) version 1.6.9 is bundled with the Prime Optical software. The sudo software enables nonroot UNIX users to run UNIX commands throughout the installation process. The following commands are available:

• opticalctl start

• opticalctl stop

• opticalctl abort

• opticalctl status

• prune_auditlog.sh

• prune_errlog.sh

• prune_audittrail.sh

• prune_fm.sh

• prune_pm.sh

• prune_ne.sh

• prune_server_monitor.sh

• prune_admin_job_table.sh

Note You can still use the following legacy commands in Prime Optical 10.5:

– ctms-start

– ctms-abort

opticalctl abort The opticalctl abort command kills all of the running processes and stops the Prime Optical server. The overall process takes no longer than 2 to 3 minutes.

opticalctl status The opticalctl status command provides Prime Optical version and process information. The following options are available:

• opticalctl status -h—Shows help information about all options available.

• opticalctl status -c—Shows TCP/LISTEN and UDP open connections of application processes.

• opticalctl status -a—Shows all connections of application processes.

• opticalctl status -v—Shows all application processes.

For example:

opticalctl status -c -a

Note To enable the opticalctl status options for Solaris 10, you must install the lsof-4.80 package. You can download the Solaris 10 lsof-4.80 package at www.sunfreeware.com.

Table 2-8 Prime Optical Commands (continued)

Command Description


http://www.sunfreeware.com


– ctms-stop

– ctms-stop-service

– showctm

If you selected the Install Sudo Software option during the Prime Optical server installation, the setup program prompts you to specify the name of the UNIX group to which you want to assign administrator privileges. By default, this group is set to the root group. If you specify a group other than root, the setup program verifies that the UNIX group exists on the system and adds entries to the /etc/sudoers file. Entries in this file reflect the commands that the specified UNIX group can run by using the sudo command.

To issue each command, log in to the Prime Optical server workstation as the root user or optusr and enter the command on the command line.

The following entries in the /etc/sudoers file reflect the commands that can be run as nonroot:

%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/prune_admin_job_table.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/prune_admin_job_table.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/prune_server_monitor.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/prune_server_monitor.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/prune_ne.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/prune_ne.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/prune_pm.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/prune_pm.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/prune_fm.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/prune_fm.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/prune_audittrail.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/prune_audittrail.%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/prune_errlog.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/prune_errlog.%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/prune_auditlog.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/prune_auditlog.%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/getinfo.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/getinfo.sh%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/opticalctl status%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/opticalctl status%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/opticalctl stop%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/opticalctl stop%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/opticalctl abort%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/opticalctl abort%CTM_UNIX_group hostname=(root) NOPASSWD: /usr/bin/opticalctl start%CTM_UNIX_group hostname=(root) NOPASSWD: /opt/CiscoTransportManagerServer/bin/opticalctl start


Cisco Pri

C H A P T E R 3
Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Configuration on a Linux Server
This chapter contains the following topics:

• Geographical Redundancy Configuration Checklists

• Setting the Environment for Installation

• Installing Oracle 12c (for Not Embedded Database Setup)

• Installing the Prime Optical Server with Embedded Database Setup

• Verifying Packages

• Disabling the Prime Optical Service and Configuring SSH

• Installing the HA Package (HA RPM Files)

• Configuring Oracle Data Guard

• Editing the Agent Configuration File

• Editing the Logging Configuration File

• Editing the RHCS Configuration File

• Starting the Cluster

• Supporting Installation Information

• Maintaining and Monitoring Cluster Activities

For more information about geographical redundancy configuration on Linux, including hardware configuration and network diagrams, see Single-Node Cluster Geographical Redundancy Configuration, page 1-3.

Geographical Redundancy Configuration Checklists This section provides checklists to help with installation tasks for Prime Optical HA in a single-node geographical redundancy configuration on Linux. We recommend that you print the checklists from the PDF, which shows the section number where you will find each task, and either check off tasks as you complete them or enter information as needed.

This section contains the following information:


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Geographical Redundancy Configuration Checklists

• Geographical Redundancy Configuration Preinstallation Checklist

• Geographical Redundancy Configuration Installation Checklist

Geographical Redundancy Configuration Preinstallation ChecklistComplete the fields in Table 3-1. The information will assist you in completing the installation.

Table 3-1 is the preinstallation checklist for a geographical redundancy configuration.

Note ODG IP and heartbeat IP can be the same.

Geographical Redundancy Configuration Installation ChecklistTo help you keep track of the steps in the installation process, print the checklist from the PDF, which shows the section number where you will find each task, and check the appropriate cells in Table 3-2 as you complete each step in the HA installation configuration process. In the cells where N/A is displayed, the step is not applicable for the primary and secondary server.

Table 3-2 is the HA installation configuration checklist.

Table 3-1 Geographical Redundancy Configuration Preinstallation Checklist

Information Primary Server Secondary Server

Prime Optical Hostname

Prime Optical IP Address

ODG Hostname

ODG IP Address

Heartbeat Hostname

Heartbeat IP Address

Table 3-2 Geographical Redundancy Configuration Installation Checklist

Description SeePrimary Cluster

Secondary Cluster

New Installation

1 Install Red Hat Enterprise Linux.

(See Installation Requirements for version)

— x x

2 (Optional) Verify that LUNs from disk arrays are visible.

Only necessary when Prime Optical files are installed on a dedicated disk array.

x x

3 Create disk group, volumes, file system, and swap space.

Table 2-5 x x

4 Create Oracle user and group (only on an external database setup).

• Oracle Prerequisites, page A-1

• Setting the Environment for Installation, page 3-4

x x


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Geographical Redundancy Configuration Checklists

5 Install Oracle 12c group (only on an external database setup).

Installing Oracle 12c (for Not Embedded Database Setup), page 3-5

x x

7 Install the Prime Optical server and database.

Installing the Prime Optical Server with Embedded Database Setup

x x

8 Verify packages. Verifying Packages, page 3-9 x x

9 Disable auto startup of Prime Optical and Oracle 12c.

Disabling the Prime Optical Service and Configuring SSH, page 3-12

10 Verify Prime Optical. Setting the Environment for Installation, page 3-4

(Step 23 only)

x x

Existing Installation

11 Install RHCS.

(See Installation Requirements for version)

• Installing the HA Package (HA RPM Files)

• Installing the Data Migrator Wizard

x x

12 Install Prime Optical HA RPM. Installing the HA Package (HA RPM Files) x x

13 Edit the ODG configuration files. • Configuring RMAN (Including Flashback)

• Editing the ODG Configuration File

• Generating the ODG Configuration File

x x

14 Deploy the ODG on the primary cluster. Configuring the Primary Server x N/A

15 Synchronize Prime Optical database with the secondary cluster.

Configuring the Mount Point to Be Backed Up

x N/A

16 Deploy the ODG on the secondary clusters. Performing a Cold Backup N/A x

17 Perform a cold backup. Performing a Cold Backup

18 Configure a backup on both servers. Configuring a Backup on the Primary and Standby Servers

19 Edit Prime Optical cluster configuration files.




x x

20 Set up and start primary and secondary clusters.

Starting RHCS Services

Starting ctm_services on the Primary Server

x x

21 Verify primary and secondary cluster system setup.

Verifying Cluster Status on the Standby Node x x

Table 3-2 Geographical Redundancy Configuration Installation Checklist (continued)

Description SeePrimary Cluster

Secondary Cluster


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Setting the Environment for Installation

Setting the Environment for InstallationSection Geographical Redundancy Configuration Checklists summarizes the steps in this section that are required to perform a Prime Optical HA installation. We recommend that you print out these checklists and check off each step as you complete it.

If you have previously installed Prime Optical, you can skip this section and begin with Verifying Packages, page 3-9.

Note TFTP is required only for specific NEs. If Prime Optical will not be used to manage NEs requiring TFTP, you can skip the steps to configure TFTP.

Note Verify that /cisco, /oracle, /db01, /db02, /db03, /db04, /db05, and /ctm_backup are mounted on the primary and secondary servers before continuing with the installation.

Enter the df -h command to verify that the file systems are mounted on the primary server.

To set up the environment for installation on the primary and secondary servers, follow these steps:

Step 1 Log in as the root user, and then enter the following commands to create a new home directory:

mkdir /cporaclechown oracle_user:oinstall /cporaclechmod 700 /cporacle

Step 2 As the root user, enter the following command to change the /oracle home directory:

usermod -d /cporacle cporacle

Step 3 Verify that the /etc/security/limits.conf file contains the following entries for the created oracle user. If they are not present, add them before the "#End of file" line:

cporacle hard memlock unlimitedcporacle hard nofile 65536cporacle hard nproc 16384cporacle soft nofile 5120cporacle soft nproc 4096cporacle soft memlock unlimited

Please adhere to the following recommendations:

• Do not use uppercase letter for hostname.

• Make sure all IPs/Hostnames (CPO IP, ODG IP, heartbeat IP) are declared inside /etc/hosts file.

• Do not use loopbacks ODG or heartbeat interface but only use hardware Ethernet interfaces.

• Test reachability of ODG and Heartbeat interfaces of the other server using the following command:

# ping Hostname-of-remote-ODG-Interface –I Hostname-of-Local-ODG-Interface


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Installing Oracle 12c (for Not Embedded Database Setup)

Installing Oracle 12c (for Not Embedded Database Setup)If you are installing an Oracle Not Embedded database, follow these steps:

Step 1 Install Oracle 12c. Complete Oracle Prerequisites, page A-1 and Installing the Oracle 12c Software with the Response File (*.rsp) Provided by Cisco, page A-3.

Step 2 Complete Downloading and Installing the Required Oracle 12c Patch for 64-Bit Linux Platforms, page A-5.

Step 3 Edit the listener.ora file by replacing the newdbname parameter with the Oracle SID (the default is CTM) and the Prime Optical hostname parameter with the hostname or IP address of the workstation where the Prime Optical database will run.

Note There are multiple instances of newdbname. You must replace all instances with the Oracle SID.

Installing the Prime Optical Server with Embedded Database Setup

Note CPO running in HA configuration cannot be installed with optusr as runtime user. For HA GEO configuration, only root user is supported as runtime user.

To install the Prime Optical server and database on both the primary and secondary servers, follow these steps:

Step 1 Log in as the root user.

Step 2 Enter the following command to verify that the display is set correctly:

echo $DISPLAY

In the command output, you should see:

hostname-or-IP-address:0.0

Step 3 If you are using an xterm window or a remote host, enter the following command to enable the xterm connection from the clients:

xhost +

The default path is /usr/bin.

Step 4 Enter the following command for Linux if the device is not automatically mounted:

a. Check if the USB device is discovered by OS and enter the following command:

fdisk -l

The output is displayed as follows:

Disk /dev/sda: 1167.9 GB, 1167996223488 bytes


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Installing the Prime Optical Server with Embedded Database Setup

255 heads, 63 sectors/track, 142000 cylindersUnits = cylinders of 16065 * 512 = 8225280 bytes Device Boot Start End Blocks Id System/dev/sda1 * 1 13 104391 83 Linux/dev/sda2 14 142000 1140510577+ 8e Linux LVM Disk /dev/sdd: 32.0 GB, 32027705344 bytes64 heads, 32 sectors/track, 30544 cylindersUnits = cylinders of 2048 * 512 = 1048576 bytes Disk /dev/sdd doesn't contain a valid partition table

b. Mount the USB device and enter the following command:

mount -t ext3 <USB device> /<USB-mount-point>

For example:

mount -t ext3 /dev/sdd /media

Step 5 Enter the following commands to mount the ISO image:

cd /media/<USB-mount-point>/PrimeOptical_10_5

mount -o loop PRIME_OPTICAL_10_5_LNX.iso /mnt

Note You can also replace the /mnt directory with a different mount point.

Step 6 Enter the following command to start the installation:

cd /mnt/mnt/Disk1/setup.sh

The setup program searches for Sun Microsystems JDK Version 1.7.0_80 on your workstation.

The Cisco Prime Optical Server Installation wizard appears. Wait for up to 90 seconds while the following message appears:

Continuing Cisco Prime Optical Server Installation.

Step 7 At the Introduction screen, click Next.

Step 8 At the License Agreement screen, read the license agreement and click the I accept the terms of the license agreement radio button. Click Next.

Step 9 At the Configure the Server and Database screen, do the following:

a. From the list of Prime Optical installation types, select the Prime Optical server and database radio button to install the Prime Optical server along with the database.

The Database field is disabled when the server and database are installed on the same workstation. Prime Optical automatically uses the server hostname or IP address.

b. Verify that the correct hostname or IP address is shown for the Prime Optical server and then click Next.

Step 10 At the Configure the Database screen, select Embedded to install Prime Optical with an embedded database and click Yes to install the database in the ARCHIVELOG mode. Click Next and the following message appear.

This installation will remove any previous Oracle database installed on the server.If you do not want to continue, click Cancel.



Note ODG configuration fails if the database is not installed in the archive log mode.

Step 11 Click Continue to delete the previous Oracle version and continue the installation. (If you click Cancel, the installation quits.)

Step 12 At the OS Users screen, choose the root user from the drop-down list.

Step 13 At the Select Network Configuration screen, specify the size of your network and then click Next.

At the Oracle Pre-Installation Checks screen, the setup program checks the RAM, swap, and temp directory sizes. The screen indicates whether you have enough space to install Prime Optical and Oracle, and whether any required packages are missing.

Step 14 Review the preinstallation checks and then click Next.

• If errors are discovered during the check sizes phase of the preinstallation checks, the following message appears, and the installation quits:

An error occurred during the pre-installation check./temp_CTM/report_oracle_pre_install_checks.log.The installation has been canceled.

Check your system for errors. After you fix any errors, restart the installation.

• During the check packages phase of the preinstallation checks, the following message appears:

A warning occurred during the preinstallation check. See /temp_CTM/report_oracle_pre_install_checks.log. To continue the installation, click Continue. To cancel the installation and fix the problem, click Cancel.

You can choose to continue installation, or quit and fix the problem before proceeding.

Step 15 In the Optional Features screen, you can choose additional features from the following options:

• Install Prime Optical in a High Availability Setup.

• Select root as runtime user for CPO (optusr is not supported on HA configuration).

• Configure FTP (ONS15216 EDFA3).

• Install Sudo Software (available for root users only).

Note Specific installation screens are displayed, depending on the additional features you select in the Optional Features screen. If you do not wish to select any additional feature, click Next and go to Step 18.

Step 16 If you selected the Install Sudo Software option, at the Prime Optical Group Information & Sudo Installation screen, do the following:

a. Enter the name of the UNIX group to which you want to assign administrator privileges.

b. To install sudo, check the Install Prime Optical Sudo check box. If you do not want to install sudo, uncheck the check box.

Note If you already installed and configured sudo with rules in a file different from /etc/sudoers (for example, if you created a custom sudo configuration in the /user/local/etc/sudoers file), then you must copy the rules from /etc/sudoers to /user/local/etc/sudoers at the end of the installation.

c. Click Next.



Step 17 If you selected the Configure FTP (ONS15216 EDFA3) option, at the FTP Information screen, do the following to configure an FTP account for software download operations:

Note NE types that require FTP configuration are:

• Cisco ONS15216 EDFA3

• Cisco ONS15305 with release lower than 3.0 (CEC-based 15305 NEs)

a. Enter the following information:

• Username

• Password

• Confirm Password

• FTP directory

b. Check or uncheck the Create new FTP account check box.

If checked, the installation script automatically creates the FTP user on the Prime Optical server workstation. If unchecked, it is assumed that an FTP user already exists on the Prime Optical server workstation.

c. Click Next.

Step 18 At the Destination Folder screen, specify where you want to install the Prime Optical server. The default directory is /cisco/PrimeOpticalServer. You can click Change to choose a different destination. After you specify your destination, click Next.

Note If the destination directory that you specified is a new directory, you will receive the message, “The specified directory does not exist. Do you want to create it?” Click Yes.

Caution Do not specify a mount point as the target installation directory for the Prime Optical server installation, or the installation data will be lost when the workstation restarts. You must create a dedicated Prime Optical subdirectory. For example, if /cisco is the mount point for the disk partition /dev/dsk/c0t0d0s5, you cannot specify /cisco as the Prime Optical installation directory. Instead, specify a dedicated subdirectory such as/cisco/PrimeOpticalServer.

Caution Do not delete any instances of /opt/CiscoTransportManagerServer from your Prime Optical file structure. Prime Optical checks for the /opt/CiscoTransportManagerServer directory or a symbolic link to it. If Prime Optical cannot find the /opt/CiscoTransportManagerServer directory or a symbolic link, Prime Optical creates a symbolic link automatically.

The Pre-Installation Summary screen shows the items that will be installed.

Step 19 Click Install.

Depending on your system performance, it might take 35 to 50 minutes for Linux operating systems.

If SSH is not configured, a warning message appears.

Step 20 Click Continue; in the terminal window that appears, enter the root password and follow the prompts to configure the SSH connection.


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Verifying Packages

The installation continues.

Step 21 In the Web Server Installation Summary window, click Next.

The Install Complete window summarizes the results of the installation.

Step 22 Click Done.

Caution Do not close the terminal or reboot the host before the following message is displayed: Prime Optical installation is complete.

Step 23 As the root user, enter the following command to reboot the system. The Prime Optical server starts automatically after rebooting:

init 6

After the server reboot, it might take up to 20 minutes for the Prime Optical server to come up.

To verify that the Prime Optical server is running, enter the opticalctl status command after the server reboots.

The opticalctl status command displays the Prime Optical server version running as 10.5, followed by the build number. In the command output, you should see two instances of “CTM Server,” “SnmpTrapService,” “SMService,” “Apache Web Server,” “TOMCAT,” and “OrchestrationService.” This indicates that the server is running.

Verifying PackagesVerify that the following packages are installed:

The RHCS packages—Available on the Red Hat Enterprise Linux (RHEL) CD under the /cluster directory. (See Installation Requirements for RHEL version)

The list of packages includes:

– cman

– rgmanager

– openais

– modcluster

– ricci

– luci

– cluster-cim

– cluster-snmp

– system-config-cluster (only for RHEL 5.8)

• rgmanager-3.0.12.1-21.el6.x86_64

Step 1 Verify that the required packages are installed. Enter one of the following commands:

rpm -q pkg

or



yum info pkg

The output displays the package details. If a required package is not installed, an error will appear.

Step 2 Use the yum command to install the Linux package. The yum command checks all the dependencies from other packages. If you have an active connection to the Red Hat website, all dependent packages are automatically installed on the server. For more information, see the Red Hat website.

If you do not have an active connection to the Red Hat website, do the following:

a. Mount the RHEL DVD or ISO file. For example:

mount -o loop -t iso9660 /mnt/redhat/rhel-server-6.6-x86_64.iso /iso

where /mnt/redhat/rhel-server-6.6-x86_64.iso is the Red Hat ISO file.

b. Enter the following commands to get the media ID:

view /iso/.discinfocat /iso/.discinfo

Here is an example of the output:

1269263646.691048Red Hat Enterprise Linux Server release 6.6 (Santiago)

x86_641,2,3,4,5,6

Server/baseServer/RPMSServer/pixmaps

The media ID in the preceding example is 1269263646.691048.

Any updates in the

c. Copy the proper local.repo file template from DVD-mount-point/Disk1 to /etc/yum.repos.d directory.

For RHEL 5.x installation, use the following command:

cp DVD-mount-point/Disk1/local.repo.5x /etc/yum.repos.d/local.repo





Open and follow the instructions in the local.repo file.

Step 3 Enter the following command to verify that the local repository has been set up correctly:

yum repolist

Output similar to the following is displayed:

Loaded plugins: rhnplugin, securityLocal | 1.3 kB 00:00Local/primary | 868 kB 00:00Local 3116/3116localHA | 1.3 kB 00:00localHA/primary | 6.1 kB 00:00



localHA 32/32repo id repo name statusLocal Local Media Repo enabled: 3,116localHA Local HA Media Repo enabled: 32rhel-x86_64-server-5 Red Hat Enterprise Linux (v. 6 for 64-bit x86_64) enabled: 14,137repolist: 17,285

Step 4 Enter the following command for all the missing packages:

yum install <name_of_rpm_file>

For example, for rgmanager, enter the following command:

yum localinstall rgmanager-3.0.12.1-21.el6.x86_64

Step 5 The cman and rgmanager configurations are enabled after you boot the server. To verify that cman and rgmanager are enabled, enter the following commands:

chkconfig --list cmanchkconfig --list rgmanager

The following output appears when cman or rgmanager is enabled (levels 2, 3, 4, and 5 show the on state):

cman 0:off 1:off 2:on 3:on 4:on 5:on 6:offrgmanager 0:off 1:off 2:on 3:on 4:on 5:on 6:off

When cman or rgmanager is not enabled, all the levels show the off state.

Step 6 Enter the following commands to enable the package:

chkconfig --level 2345 cman onchkconfig --list cmanchkconfig --level 2345 rgmanager onchkconfig --list rgmanager

Step 7 Enter the following command to check the openais service status:

service openais status

If the service is running, enter the following command to stop the openais service.

service openais stop

Step 8 Enter the following command to confirm that the openais service is not automatically started at boot time:

chkconfig openais --list

The result should look like this:

openais 0:off 1:off 2:off 3:off 4:off 5:off 6:off

Step 9 If any of the levels are set to on, enter the command to disable the service. For example, if levels 2,3,4, and 5 are set to on, enter the following command:

chkconfig --level 2345 openais off


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Disabling the Prime Optical Service and Configuring SSH

Disabling the Prime Optical Service and Configuring SSHTo disable the Prime Optical service and configure SSH, follow these steps:

Step 1 Enter the following command to disable the Prime Optical service:

chkconfig --level 345 CTMService off

Step 2 Enter the following commands from the primary node to configure SSH:

ssh "secondary-replication-if" lsssh "secondary-heartbeat-if" ls

Where:

• secondary-replication-if—Indicates the hostname associated with the interface used for the replication link on the secondary node. (For example, ha2-105.cisco.com)

• secondary-heartbeat-if—Indicates the hostname associated with the interface used for the heartbeat link on the secondary node. (For example, ha2-105.cisco.com)

Note Both commands must display the contents of the secondary node's /root directory.

SSH with root access is configured between the primary and the secondary nodes. Configure SSH for the replication and heartbeat links (see Figure 1-4).

Step 3 Enter the following commands from the secondary node:

ssh "primary-replication-if" lsssh "primary-heartbeat-if" ls

Where:

• primary-replication-if—Indicates the hostname associated with the interface used for the replication link on the primary node. (For example, ha1-105.cisco.com).

• primary-heartbeat-if—Indicates the hostname associated with the interface used for the heartbeat link on the primary node. (For example, ha1-105.cisco.com)

Note Both commands must display the contents of the primary node's /root directory.

Step 4 If SSH verification fails, open a shell from the Prime Optical server workstation and do the following on the Primary and secondary cluster servers:

On the primary cluster:

a. As the root user, enter the following command to start the SSH service:

service sshd start

b. If the /.ssh/id_rsa and /.ssh/id_rsa.pub files do not exist, enter the following command to generate public and private keys:

ssh-keygen -t rsa

Press Return to accept the default values for the following prompts:

Enter file in which to save the key (//.ssh/id_rsa):Enter passphrase (empty for no passphrase):


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Installing the HA Package (HA RPM Files)

Enter same passphrase again:

c. Enter the following command to publish the public key to the secondary cluster server:

ssh-copy-id -i ~/.ssh/id_rsa.pub root@Secondary-Cluster-Server

where secondary-cluster-server is the name of the secondary cluster server.

d. Edit the /etc/ssh/sshd_config file on the primary cluster server and change the PermitRootLogin value to yes.

e. Enter the following command on the primary cluster server to save and apply the changes:

service sshd restart

f. Enter the following command to open an SSH connection from the primary cluster server to the secondary cluster server and register the SSH key:

ssh Secondary Cluster server

g. Repeat Step 2 and Step 3 to verify the SSH configuration.

On the secondary cluster:

a. As the root user, enter the following command to start the SSH service:

service sshd start

b. If the /.ssh/id_rsa and /.ssh/id_rsa.pub files do not exist, enter the following command to generate public and private keys:

ssh-keygen -t rsa

Press Return to accept the default values for the following prompts:

Enter file in which to save the key (//.ssh/id_rsa):Enter passphrase (empty for no passphrase): Enter same passphrase again:

c. Enter the following command to publish the public key to the primary cluster server:

ssh-copy-id -i ~/.ssh/id_rsa.pub root@Primary-Cluster-Server

d. Edit the /etc/ssh/sshd_config file on the secondary cluster server and change the PermitRootLogin value to yes.

e. Enter the following command on the secondary cluster server to make the changes take effect:

service sshd restart

f. Enter the following command to open an SSH connection from the secondary cluster server to the primary cluster server and register the SSH key:

ssh Primary Cluster server

g. Repeat Step 2 and Step 3 to verify the SSH configuration.

Installing the HA Package (HA RPM Files)The High Availability solution is delivered with the HA-1.5-3.x86_64.rpm file. The rpm file is on the Prime Optical 10.5 installation USB drive or ISO image.


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Installing the HA Package (HA RPM Files)

To install the Prime Optical HA package, follow these steps on the Primary and Secondary servers:

Step 1 Insert the Cisco Prime Optical Server installation USB drive—Linux in your workstation.

Note Wait for the automatic discovery of the device.

Enter the following command for Linux only if the device is not automatically discovered.

mount -t ext3 <USB device> /media/<USB-mount-point>

For example:

mount -t ext3 /dev/sdd1 /media/<USB-mount-point>



mount -p loop PRIME_OPTICAL_10_5.iso /mnt


Step 3 Enter the following command to change to the HA RPMS directory.

cd /mnt/HA/RPMS/x86_64

Step 4 Enter the following command:

yum install --nogpgcheck HA-1.5-3.x86_64.rpm


Loaded plugins: product-id, refresh-packagekit, security, subscription-managerUpdating certificate-based repositories.Setting up Install ProcessExamining HA-1.5-3.x86_64.rpm: HA-1.5-3.x86_64.rpmMarking HA-1.5-3.x86_64.rpm to be installedResolving Dependencies--> Running transaction check---> Package HA.x86_64 0:1.5-3 will be installed--> Finished Dependency Resolution

Dependencies Resolved

======================================================================================================================Package Arch Version Repository Size======================================================================================================================Installing:HA x86_64 1.1-2 /HA-1.5-3.x86_64.rpm 274 k

Transaction Summary======================================================================================================================Install 1 Package(s)

Total size: 274 kInstalled size: 274 k


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Configuring Oracle Data Guard

Is this ok [y/N]: yDownloading Packages:Running rpm_check_debugRunning Transaction TestTransaction Test SucceededRunning Transaction Installing : HA-1.5-3.x86_64.rpm1/1

Override dbora startup scriptOverride rman_util.py scriptInstallation DONE!Installed products updated.

Installed:HA.x86_64 0:1.5-3

Complete!

After installation, navigate to the /opt/CiscoTransportManagerServer/HA directory. You should see the directory structure on successful installation. See HA Directory Structure, page 3-32 for more information on the directory structure.

Step 5 Enter the following commands to overwrite the existing backup and restore scripts:

cd /opt/CiscoTransportManagerServer/HA/binmv backupdb.sh /opt/CiscoTransportManagerServer/bin/mv restore_db.sh /opt/CiscoTransportManagerServer/bin/

Step 6 Enter the following commands to copy the logger configuration file:

cd /opt/CiscoTransportManagerServer/HA/installcp logRMANconf.ini /opt/CiscoTransportManagerServer/HA/cfg/

Configuring Oracle Data Guard

Note • Before you configure Oracle Data Guard (ODG), verify that the Prime Optical server is not running. If it is running, enter the ctms-stop command.

• Before you configure ODG, make sure that CPO has been installed on the both primary and secondary servers. ODG configuration fails CPO has not been installed on secondary server.

Oracle Data Guard provides a data protection and disaster recovery solution. This product offers protection for Oracle databases, providing customers the ability to protect business data.

The ODG is included in Oracle Enterprise Edition. You must configure Oracle to enable the ODG configuration. The scripts and the configuration files are installed with the HA-1.5-3.x86_64.rpm file. To configure ODG, you must install the Prime Optical Server on your primary and secondary servers.

The following sections describe ODG configuration:

• Configuring RMAN (Including Flashback), page 3-16

• Editing the ODG Configuration File, page 3-17

• Generating the ODG Configuration File, page 3-20



• Configuring the Primary Server, page 3-21

• Performing a Cold Backup, page 3-22

• Configuring the Mount Point to Be Backed Up, page 3-21

• Performing a Cold Backup, page 3-22

• Configuring the Standby Server, page 3-22

• Configuring a Backup on the Primary and Standby Servers, page 3-23

Configuring RMAN (Including Flashback)Oracle Recovery Manager (RMAN) is a tool that is used to back up, restore, and recover the Oracle database.

Oracle Flashback is used to view and store backups and archive logs. It reduces database recovery time without using the Backup operation.

To configure RMAN and Flashback, follow these steps:

Step 1 Connect to the primary server and enter the following command:

cd /opt/CiscoTransportManagerServer/HA/install

Step 2 Edit the oracle_rman_config.cfg configuration file.

• To configure Flashback, substitute these variables in the Flashback section:

+DIRECTORY_AREA+ +SIZEGB++UPPER_LIMIT_MINUTES+

Where:

– +DIRECTORY_AREA+—Directory where the backup and flashback must reside. The area must be large enough to store at least seven backups.

– +SIZEGB+—Size of memory, in gigabytes, reserved for the flashback. See Table 3-3 for more information. This is an upper limit for disk space pressure. If this threshold is reached, a warning message is written in the Oracle Alert logs. If this happens, use the RMAN utility to clean Flashback.

For example:

db_recovery_file_dest=/ctm_backupdb_recovery_file_dest_size=20Gdb_flashback_retention_target=30

• To configure RMAN, substitute these variables in the RMAN section:

Table 3-3 Prime Optical Size Requirements

Network Size Prime Optical Size

Small 50 G

Medium 100 G

large 200 G

HA 300 G



+NAME_RMAN_LOGGER++MINIMUM_DAYS_STORED++DAYS_RETENTION_BACKUP+

Where:

– +NUMBER_OF_BACKUPS+—Number of archive log backups considered by RMAN. Old archive log backups can be deleted using the RMAN command.

– +NAME_RMAN_LOGGER+—Name of the logger file used by Prima Optical when using the RMAN command.

– +MINIMUM_DAYS_STORED+—Minimum number of days to store backup information in the control file that RMAN considers active.

– +DAYS_RETENTION_BACKUP+—Minimum number of days of backups that RMAN considers active. After this time, the backup is considered obsolete and can be deleted.

For example:

log_rman=/ctm_backup/report_RMAN.logcontrol_file_record_keep_time=7recovery_window=7archivelogs_retention=1

Step 3 Copy the modified oracle_rman_config.cfg file to the standby server.

Editing the ODG Configuration FileTo edit the ODG configuration, follow these steps:

Step 1 Connect to the primary and standby server, and enter the following command:

cd /opt/CiscoTransportManagerServer/HA/install

Step 2 Edit the odg_config.cfg configuration file.

In the odg_config.cfg file, find all parameters that are contained within plus signs (+...+) and enter the correct values according to the actual server configuration. Table 3-4 lists the parameters and their descriptions.

Note Only replace parameters with plus (+...+) signs. Do not replace parameters with < > sign. Those will be automatically replaced by the script.

Step 3 Copy the modified odg_config.cfg file to the standby server.



Table 3-4 ODG Configuration Parameters

Parameter Description Example/Notes

Custom Section

redo_group_size=+SIZE_REDO_LOG+

Size, in megabytes or gigabytes, of the single redo log configured.

The default values for the following configurations are:

• Small—200 MB

• Medium—300 MB

• Large—500 MB

• High-end—800 MB

redo_group_number=+ACTUAL_NUMBER_OR_REDO_GROUP_PLUS_1+

Number of groups for the standby redo log. This is the actual number of Online Redo Log +1.

The default value of the Online Redo Log for Prime Optical is 3. In this scenario, the correct value of this parameter is 4.

redo_stby_p=+LIST_OF_REDOLOG_FILES+

List of all the new redo logs for the standby functionality. The syntax is:

<disk_partition_1>/< DB_NAME_STANDBY>/<name of the redolog file>,<disk_partition_2>/>/< DB_NAME_STANDBY>/<name of the redolog file>/.....

The syntax describes the following:

• <disk_partition_n>—Mount point or directory where the redo log must be created.

• < DB_NAME_STANDBY>—Value set for the db_unique_name_standby parameter (mandatory).

For example, four added redo logs (redo04, redo05, redo06, and redo07) appear as follows:

redo_stby_p=/db03/<db_unique_name_standby>/redo04_1.dbf,/db05/<db_unique_name_standby>/redo04_2.dbf,/db01/<db_unique_name_standby>/redo05_1.dbf,/db02/<db_unique_name_standby>/redo05_2.dbf,/db03/<db_unique_name_standby>/redo06_1.dbf,/db05/<db_unique_name_standby>/redo06_2.dbf,/db01/<db_unique_name_standby>/redo07_1.dbf,/db02/<db_unique_name_standby>/redo07_2.dbf

mode_authentication=password_file

Indicates that the remote authentication of SYSDBA uses a password file.

—

authentication_passwd_file=+ORACLE_PASSWD_FILE_LOCATION+

Path completed with the filename of the Oracle password file.

The default location is in the /oracle/product/12.1.0/dbs/orapw directory.

oracle_user=+ORACLE_USER+

Oracle user created to install Oracle software.

cporacle

oracle_group=+ORACLE_GROUP+

Group database administrators. dba

oracle_oinstall=+ORACLE_OINSTALL_GROUP+

Group for Oracle software installation. oinstall

oracle_home=+ORACLE_HOME+

Value of the ORACLE_HOME environment variable for Oracle user.

/oracle/product/12.1.0

Primary Section

ip_host_p=+IP_HOST_PRIMARY+

Prime optical hostname or Prime Optical IP address on primary server.

ha1-105.cisco.com



The following is an example of a completed odg_config.cfg file:

Note On launching odg_configure -f command, the odg_config.cfg file is displayed on replacing ++ parameters.

## Oracle Data Guard configuration file##[ODG_server_common]type=maximum_performancemode=physicalredo_group_number=4redo_group_size=300Mredo_stby_p=/db01/CPO_STBY/redo04_1.dbf,/db04/CPO_STBY/redo04_2.dbf,/db02/CPO_STBY/redo05_1.dbf,/db05/CPO_STBY/redo05_2.dbf,/db03/CPO_STBY/redo06_1.dbf,/db01/CPO_STBY/redo06_2.dbf,/db05/CPO_STBY/redo07_1.dbf,/db02/CPO_STBY/redo07_2.dbf

ip_host_odg_p= +HOSTNAME_OR_IP_ODG+

ODG hostname or ODG IP address on primary server.

ha1-105.cisco.com or hostname of ODG primary server (If ODG or Oracle have different IP address/hostname)

db_unique_name_primary=+DB_NAME_PRIMARY+

String to identify the primary server. CPO_PRIM

The value must be distinct from the from db_name and the db_unique_name_standby parameters.

db_name=+SID+ System identifier that is used to create the Prime Optical database. Because the two servers are distinct, the system ID is the same on both the primary and standby servers.

optdb

Note On an embedded configuration, SID is optdb.

StandBy Section

num_hosts=+NUM_STANDBY_SERVERS+

Number of hosts on the standby server. The fixed value is 1. Only one standby server can be managed for now.

1

ip_host_s=+IP_HOST_STANDBY+

Prime Optical hostname or Prime Optical IP address on secondary server.

ha2-105.cisco.com

ip_host_odg_s=+HOSTNAME_OR_IP_ODG+

ODG hostname or ODG IP address on secondary server.

ha2-105.cisco.com or hostname of ODG secondary server (If ODG or Oracle have different IP address/hostname)

db_unique_name_standby=+DB_NAME_STANDBY+

String to identify the standby server. CPO_STBY

This value must be distinct from the db_name and the db_unique_name_primary parameters.

db_name=+SID_EQUAL_TO_PRIMARY+

System identifier that is used to create the Prime Optical database. Because the two servers are distinct, the system ID is the same on both the primary and standby servers.

optdb

On an embedded configuration, SID is optdb.

Table 3-4 ODG Configuration Parameters (continued)

Parameter Description Example/Notes



mode_authentication=password_fileauthentication_passwd_file=/oracle/product/12.1.0/dbs/orapworacle_user=cporacleoracle_group=dbaoracle_oinstall=oinstalloracle_home=/oracle/product/12.1.0

[ODG_server_primary]num_hosts=1ip_host_p=ha1-105.cisco.comip_host_odg_p=ha1-105.cisco.comdb_unique_name_primary=CPO_PRIMdb_name=optdblog_arch_dest_primary=/db05log_archlog_dest_primary=USE_DB_RECOVERY_FILE_DESTlog_file_name_convert='/db05/CPO_STBY/','/db05/optdb/'log_archive_dest_1='LOCATION=/db05 VALID_FOR=(ALL_LOGFILES, ALL_ROLES) DB_UNIQUE_NAME=CPO_PRIM'log_archive_dest_2='SERVICE=CPO_STBY ASYNC VALID_FOR=(ONLINE_LOGFILES, PRIMARY_ROLE) DB_UNIQUE_NAME=CPO_STBY'log_archive_dest_state_1=ENABLElog_archive_dest_state_2=ENABLElog_archive_max_processes=30log_archive_config='DG_CONFIG=(CPO_PRIM,CPO_STBY)'fal_server=CPO_STBYstandby_file_management=AUTOremote_login_passwordfile=EXCLUSIVE

[ODG_server_standby]num_hosts=1ip_host_s=ha2-105.cisco.comip_host_odg_s=ha2-105.cisco.comdb_unique_name_standby=CPO_STBYdb_name=optdblog_arch_dest_standby=/db05log_archlog_dest_standby=USE_DB_RECOVERY_FILE_DESTcontrol_files='/db01/CPO_STBY/ctm_control1.ctl'log_file_name_convert='/db05/optdb/','/db05/CPO_STBY/'log_archive_dest_1='LOCATION=/db05 VALID_FOR=(ALL_LOGFILES, ALL_ROLES) DB_UNIQUE_NAME=CPO_STBY'log_archive_dest_2='SERVICE=CPO_PRIM ASYNC VALID_FOR=(ONLINE_LOGFILES, PRIMARY_ROLE) DB_UNIQUE_NAME=CPO_PRIM'log_archive_config='DG_CONFIG=(CPO_PRIM,CPO_STBY)'fal_server=CPO_PRIMstandby_file_management=AUTOremote_login_passwordfile=EXCLUSIVE

Generating the ODG Configuration File To generate the ODG configuration file, follow these steps

Note Make sure the odg_config.cfg file has been copied from the primary to the secondary server.

Step 1 Connect to the primary and standby servers, and navigate to the /opt/CiscoTransportManagerServer/HA/install directory.

The configured mount point parameter value is substituted automatically to generate the definitive version of the configuration file.



Step 2 Enter the following command on both the primary and secondary servers:

./odg_configure.py -f

-f indicates that the script generates the final version of the odg_config.cfg file. The script substitutes internal variables and generates the final version of all configuration files.

The following output appears:

./odg_configure.py -fUsing the default configuration files: Configuration ODG : /opt/CiscoTransportManagerServer/HA/install/odg_config.cfg Configuration RMAN: /opt/CiscoTransportManagerServer/HA/install/oracle_rman_config.cfg Configuration TNS : /opt/CiscoTransportManagerServer/HA/install/odg_tns_config.cfg Configuration Logger: /opt/CiscoTransportManagerServer/HA/install/logODGconf.ini

Configuring the Primary ServerTo configure the primary server, follow these steps:

Step 1 Connect to the primary server and navigate to the /opt/CiscoTransportManagerServer/HA/install directory.

Step 2 Enter the following command to configure the common port on the primary server.

./odg_configure.py -c

Step 3 Enter the following command to run the primary server-specific configuration:

./odg_configure.py -p

Configuring the Mount Point to Be Backed UpTo configure mount point backup, follow these steps:

Step 1 Enter the following command on the primary server:

cd /opt/CiscoTransportManagerServer/HA/cfg

Step 2 In the clusterConfig.ini file, edit the following parameters:

• mpDatafileList—Sets all of the mount points that must be monitored and backed up in datafile, redolog, and standby redolog.

For example, in a standard configuration:

mpDatafileList=/db01,/db02,/db03,/db04,/db05

• mpAppList—Sets all of the mount points for the Oracle installation and the Prime Optical installation.

For example, in a standard configuration:



mpAppList =/oracle,/cisco

Note To align the primary database with the standby database, and for existing backup files, add /ctm_backup to the standard configuration mount point list. Edit the clusterConfig.ini file to add the /ctm_backup mount point. If the /ctm_backup directory has large amount of data, the backup will take a long time to process.

Performing a Cold BackupTo perform a cold backup of the primary server on the standby server, follow these steps:

Step 1 Connect to the primary server. Enter the following command:

/opt/Cisco*Server/HA/bin/odg_console.py

Choose Shutdown the database in the ODG console menu.

Step 2 Connect to the standby server. Enter the following command:

cd /opt/Cisco*Server/HA/bin./odg_console.py

Step 3 Choose Shutdown the database in the ODG console menu.

Step 4 Connect to the primary server and start the backup:

cd /opt/CiscoTransportManagerServer/HA/install./RsyncDb.py

Note Every mount point to be backed up requires a confirmation to proceed.

Step 5 After the backup is completed, do the following:

a. Connect to the standby server and mount the database; then enter the following command:

/opt/Cisco*Server/HA/bin/odg_console.py

Choose Open in mount the database in the ODG console menu.

b. Connect to the primary server:

cd /opt/Cisco*Server/HA/bin./odg_console.py

Choose Startup the database in the ODG console menu.

Configuring the Standby ServerYou must complete the steps in Generating the ODG Configuration File, page 3-20 and Configuring the Primary Server, page 3-21 sections before you configure the standby server.


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Editing the Agent Configuration File

To configure the standby server, follow these steps:

Step 1 Connect to the standby server and navigate to the /opt/CiscoTransportManagerServer/HA/install directory.

Step 2 Enter the following command to configure the common port on the standby server:

./odg_configure.py -c

Step 3 Enter the following command to run the standby server-specific configuration after completing the ODG phase:

./odg_configure.py -s

Configuring a Backup on the Primary and Standby ServersTo configure a backup on the primary and standby servers, follow these steps:


cd /opt/Cisco*Server/HA/bin./backupdb.py

Choose Configure RMAN on Primary in the backup console menu.

Step 2 Connect to the standby server. Enter the following command:

cd /opt/Cisco*Server/HA/bin./backupdb.py

Choose Configure RMAN on Standby in the backup console menu.

Editing the Agent Configuration FileYou must configure the agents that Red Hat Cluster Suite (RHCS) uses. All of the configuration parameters are available in /opt/CiscoTransportManagerServer/HA/cfg/clusterConfig.ini. This file contains a main section for each agent (for example, [ODGAgent] for the OracleReplicator agent). Each section contains a series of configured parameters.

Note To complete agent configuration you must replace all the TODO tags in the clusterConfig.ini file. Table 3-5 lists the descriptions of the parameters associated with the TODO tag.


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Editing the Agent Configuration File

The following in an example of a completed clusterConfig.ini file:

#Name of CTM service as reported in /etc/cluster/cluster.conf filectmservice=ctm_service#Name of heartbeat service as reported in /etc/cluster/cluster.conf fileheartbeatservice=ctm_heartbeat[NetworkAgent]#Network interface used to monitor the Network elements. This is a critical resourceinterface=eth0#State of the network interface. Agent uses ifconfig to get the statusmatchPath=UP BROADCAST RUNNING[ODGAgent]#Network interface used to by ODG to replicate redologs. This is nor a critical resourceinterface=eth1#State of the network interface. Agent uses ifconfig to get the statusmatchPath=UP BROADCAST RUNNING#window time to check if replication works properly. It is the double of the time used by RHCS to invoke agentspolling=60#It is the other node of the GEO cluster.otherNode=ha2-105.cisco.com#If True caused a Failover of ODGforced_failover=False[CtmAgent]#maximum number of attempts to restart CTM before to declare CTM faultedmaxAttempts=3#wait time between attemptswaitBetweenAttempts=10#temporary file used to store the previous attemptsattemptFile=/tmp/attemptFile[ApacheAgent]#maximum number of attempts to restart Apache before to declare Apache faultedmaxAttempts=3#wait time between attemptswaitBetweenAttempts=10#command used to start/stop apachewebserver=/opt/CiscoTransportManagerServer/webServer/bin/apachectl

Table 3-5 TODO Tags for Agent Configuration

Agent Parameter Description

NetworkAgent interface Network interface that Prime Optical uses to retrieve data from a monitored network. This interface is a critical resource; we strongly recommend that you protect it by bonding interfaces on different NIC.

ODGAgent interface Interface that ODG uses to transfer redo logs to the other node of the geographical redundancy cluster. This interface is a critical resource and must be protected.

ODGAgent othernode Name associated to the network interface of the other node (if you are on the primary node, the other node is the standby, or vice-versa) designed to receive redo logs.

HeartBeatAgent interface Interface that HeartBeatAgent uses to ping the other node (if you are on the primary node, the other node is the standby, or vice-versa) of the geographical redundancy cluster; for example, eth0, eth1, bond0.

HeartBeatAgent othernode Name associated to the network interface of the other node (the same as the ODGAgent name).

RSynchAgent othernode Name associated to the network interface of the other node (the same as the ODGAgent name).


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Editing the Logging Configuration File

[HeartBeatAgent]#Network interface used by heartbeat service to monitor connection between nodes of the GEO clusterinterface=eth1#State of the network interface. Agent uses ifconfig to get the statusmatchPath=UP BROADCAST RUNNING#It is the other node of the GEO cluster.otherNode=ha2-105.cisco.com#wait time in case of ping failurewaitInterval=20[RsyncAgent]#It is the replicated directory on local host. It contains the NEs database backupsrcdir=/opt/CiscoTransportManagerServer/admin#It is the destination directory on remote host.dstdir=/opt/CiscoTransportManagerServer/#It is the other node of the GEO cluster.otherNode=ha2-105.cisco.com[FsAgent]#Comma separate list of mount point to be monitoredmpDatafileList=/db01,/db02,/db03,/db04,/db05mpAppList=/oracle,/cisco[Flashback]threshold=70[Test]value=0

Editing the Logging Configuration FileAgent scripts uses a logging framework to trace their monitoring activities. The logHAconf.ini file is used to set the logging properties. It has different loggers and handlers. The different log levels are:

• NOTSET

• CRITICAL

• ERROR

• WARNING

• INFO

• DEBUG

Table 3-6 lists the logging parameters descriptions.

Table 3-6 Logging Parameters

Logger Description

Logger_logAgentCluster The logger that python agent scripts use.

Logger_logGeoManager The logger that GEO_manager.py scripts use.

Handler_HCiscoAgentCluster

The handler that logAgentCluster uses. You can specify the filename, the size, and the number of files that the logger manages using this handler.

Handler_HGeoManager The handler that logGeoManager uses. You can specify the filename, the size, and the number of files that the logger manages using this handler.


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Editing the RHCS Configuration File

Editing the RHCS Configuration FileTo configure the cluster on both the primary and secondary nodes, follow these steps:

Step 1 Edit the Primary server's cluster.conf template in the /opt/CiscoTransportManagerServer/HA/template directory, replacing the TODO tags with the correct values. See Table 3-7 for more information on the RHCS configuration parameters.

Here is an example of a Primary server's cluster.conf file:

<?xml version="1.1"?><cluster config_version="1" name="HA1_CLUSTER"><fence_daemon clean_start="0" post_fail_delay="0" post_join_delay="3"/><clusternodes><clusternode name="ha1-105.cisco.com" nodeid="1" votes="1"><fence/></clusternode></clusternodes><cman/><fencedevices/><rm log_facility="local4" log_level="7"><failoverdomains><failoverdomain name="Ha1_Domain" nofailback="0" ordered="0" restricted="0"/>

Handle_mail The handler that is used to send notification messages via e-mail. You can specify:

• The level of sent messages

• The sender (root@<hostname>)

• A list of recipients

Note All python scripts that monitor resources use this handler to send a message related to the faulted resource when critical errors occur. Be sure to include the administrator user in the list of receivers of this handler. If you do not specify a user, there will be no notification in the event of a critical fault.

Note Handle_mail handler uses the mailer configured on your server. Please verify that at least one mailer is configured on your system.

For example run one of the following commands:

service sendmail status

or

service postfix status

handler_syslog The handler that is used to log the python agent activity in the default syslog configured on the server. The default syslog, is /var/log/message, and can be configured through /etc/syslog.conf file.

The logger in the default configuration logs only CRITICAL messages and can be modified by specifying the desired level. By default, the logAgentCluster does not use the handler_syslog. To enable the handler to log message on syslog, add the syslog handler to the list of handlers used by logAgentCluster.

Table 3-6 Logging Parameters (continued)

Logger Description



</failoverdomains><resources><script file="/cisco/PrimeOpticalServer/HA/bin/ODG_IFAgent.py" name="REPL_IF"/><script file="/cisco/PrimeOpticalServer/HA/bin/ODG_ReplicatorAgent.py" name="ORACLE_REPLICATOR"/><script file="/cisco/PrimeOpticalServer/HA/bin/OracleAgent.py" name="CTM_SID"/><script file="/cisco/PrimeOpticalServer/HA/bin/NtwIFAgent.py" name="NTW_IF"/><script file="/cisco/PrimeOpticalServer/HA/bin/FsCheckAgent.py" name="FSCheck"/><script file="/cisco/PrimeOpticalServer/HA/bin/ApacheAgent.py" name="CTM_APACHE"/><script file="/cisco/PrimeOpticalServer/HA/bin/CtmAgent.py" name="CTM_SRV"/><script file="/cisco/PrimeOpticalServer/HA/bin/RsyncAgent.py" name="CTM_RSYNC"/><script file="/cisco/PrimeOpticalServer/HA/bin/HeartBeat.py" name="CTM_HEARTBEAT"/><script file="/cisco/PrimeOpticalServer/HA/bin/FlashBackMonitor.py" name="FLASHBACK"/></resources><service autostart="0" domain="Ha1_Domain" exclusive="0" name="ctm_service" recovery="disable"><script ref="FSCheck"/><script ref="NTW_IF"/><script __independent_subtree="2" __max_restarts="20" __restart_expire_time="900" ref="CTM_RSYNC"/><script __independent_subtree="2" __max_restarts="10" __restart_expire_time="900" ref="REPL_IF"/><script __independent_subtree="2" ref="ORACLE_REPLICATOR"/><script ref="CTM_SID"><script ref="CTM_SRV"><script ref="CTM_APACHE"/></script></script></service><service autostart="1" exclusive="0" max_restarts="3" name="ctm_heartbeat" recovery="restart" restart_expire_time="900"><script ref="CTM_HEARTBEAT"/></service><service autostart="1" exclusive="0" max_restarts="3" name="ctm_monitoring" recovery="restart" restart_expire_time="900"><script ref="FLASHBACK"/></service></rm></cluster>

Step 2 Edit the Secondary server's cluster.conf template file in the /opt/CiscoTransportManagerServer/HA/template directory, replacing the TODO tags with the correct values. See Table 3-7 for more information on the RHCS configuration parameters.

Here is an example of a Secondary server cluster.conf file:

<?xml version="1.4"?><cluster config_version="1" name="HA2_CLUSTER"><fence_daemon clean_start="0" post_fail_delay="0" post_join_delay="3"/><clusternodes><clusternode name="ha2-105.cisco.com" nodeid="1" votes="1"><fence/></clusternode></clusternodes><cman/><fencedevices/><rm log_facility="local4" log_level="7"><failoverdomains><failoverdomain name="Ha2_Domain" nofailback="0" ordered="0" restricted="0"/></failoverdomains><resources><script file="/cisco/PrimeOpticalServer/HA/bin/ODG_IFAgent.py" name="REPL_IF"/><script file="/cisco/PrimeOpticalServer/HA/bin/ODG_ReplicatorAgent.py" name="ORACLE_REPLICATOR"/>


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Starting the Cluster

Starting the ClusterThe following sections describe the process of starting the cluster:

• Starting RHCS Services, page 3-30

• Starting ctm_services on the Primary Server, page 3-30

• Verifying Cluster Status on the Standby Node, page 3-31

Table 3-7 RHCS Configuration Parameters

Tags Description Example

TODO_alias Alias name of the cluster. It can be any string. Primary Server - HA1_CLUSTER

Secondary Server - HA2_CLUSTER

TODO_name Name of the cluster. It can be any string that is the same as the alias name.

Primary Server - HA1_CLUSTER

Secondary Server - HA2_CLUSTER

TODO_nodename Name of the server. It is the output of the hostname command.

Primary Server - ha1-105.cisco.com

Secondary Server - ha2-105.cisco.com

TODO_domain_name Name of the failover domain. It can be any string. Primary Server - Ha1_Domain

Secondary Server - Ha2_Domain

TODO_loglevel1-7 Log level that the resource manager uses. It is an integer from 1 to 7 (7 means debug). Logs are visible in the /var/log/messages directory.

Note The rm tag contains the log_facility="local4" attribute. This allows you to configure a dedicated file for RHCS log. For example, you can add the following entries to /etc/syslog.conf order to have resource manager (RHCS) log to /va/log/cisco_cluster.log file.

#RHCS log

local4.* /var/log/cisco_cluster.log

Save the file and restart the syslog.

7



Starting RHCS ServicesTo start RHCS services, follow these steps:


service cman start

Step 2 Enter the following command on the standby server:

service cman start


service rgmanager start



ctm_heartbeat automatically starts when you start rgmanager. On the primary server, ctm_heartbeat pings the standby server at startup to find out if the RHCS services (cman and rgmanager) are running. This check is performed three times, with a 10-second wait between attempts. If the three attempts fail, RHCS moves ctm_heartbeat to the failed state.

Step 5 Enter the following command to check the cluster status:

clustat

Step 6 Enter the following commands to recover functionality if ctm_heartbeat fails:

clusvcadm -d ctm_heartbeatclusvcadm -e ctm_heartbeat

Starting ctm_services on the Primary ServerAfter you have successfully started RHCS services, you can start Prime Optical from the primary server. All resources used by Prime Optical are defined in the ctm_service in /etc/cluster/cluster.conf file.

To start ctm_services on the primary server:

Step 1 Check the state of the cluster by executing the following command:

clustat

The clustat command must show that ctm_heartbeat has started when ctm_service is disabled.

If ctm_service is not disabled, invoke it by executing the following command:

clusvcadm -d ctm_service

Step 2 Open a ODG console and enter the following command to verify that ODG is aligned.

/opt/CiscoTransportManagerServer/HA/bin/odg_console.py



Invoke the Check DatabaseRole and status option. If the Switch over status shows RECOVERY NEEDED, invoke the Enable Redo Apply option. Invoke the Check DatabaseRole and status option again till the Switch over status shows TO STANDBY.

Step 3 Start the ctm_service. Enter the following command:

clusvcadm -e ctm_service

The clusvcadm -e ctm_service command starts all Prime Optical resources. You can check the startup sequence detailed in the /var/log/message directory and the output of /opt/Cisco*Server/HA/log/Cisco_agents_cluster.log (or the filename that you specified in the logHAgent.ini file when you configured logging).


clustat

The clustat command output should show that both ctm_heartbeat and ctm_service are started.

Prime Optical is now running on the primary server.


Cluster Status for HA1_CLUSTER @ Thu Feb 12 19:40:45 2015Member Status: Quorate


Service Name Owner (Last) State------- ---- ----- ------ -----service:ctm_heartbeat ha1-105.cisco.com startedservice:ctm_monitoring ha1-105.cisco.com startedservice:ctm_service ha1-105.cisco.com started

Verifying Cluster Status on the Standby NodeEnter the following command to verify the status of the standby server:

clustat

The output must show that ctm_heartbeat process started when ctm_service was disabled.

Enabling Flashback on the Standby Node

To enable flashback on the standby node:

Step 1 As a root user on the primary node, open a dedicated shell and enter the following command:


Step 2 As a root user on standby node, open a dedicated shell and enter the following command:



Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Supporting Installation Information

Step 3 Enter the following command to verify that the ctm_service has started on the secondary cluster after the switch over operation is terminated:

clustat

Step 4 To enable flashback, start the backup console by executing the following command:

/opt/CiscoTransportManagerServer/HA/bin/backupdb.py

Step 5 To verify flashback, enter the number that corresponds to the Check status of Flashback command. If the flashback status is enabled, go to Step 15. If the flashback status is not enabled, follow Step 6 through Step 15.

Step 6 To exit the backup utility, enter the number that corresponds to the Exit command in the backup console.

Step 7 Open a ODG console and enter the following command:

/opt/CiscoTransportManagerServer/HA/bin/odg_console.py

Step 8 To enable flashback, enter the number that corresponds to the Enable the flashback command in the ODG console.

Step 9 To exit the backup utility, enter the number that corresponds to the Exit command in the ODG console.

Step 10 Restart the backup console by executing the following command:

/opt/CiscoTransportManagerServer/HA/bin/backupdb.py

Step 11 To check if flashback is enabled, enter the number that corresponds to the Check status of Flashback command.

Step 12 To exit the backup utility, enter the number that corresponds to the Exit command.

Step 13 As a root user on the standby node, open a dedicated shell and enter the following command:


Step 14 As a root user on primary node, open a dedicated shell and enter the following command:


Step 15 Enter the following command to verify that the ctm_service has started on the primary cluster after the switch over operation is terminated:

clustat

Supporting Installation InformationThis section lists supporting ODG file and HA directory structure information that might be helpful during installation.

HA Directory StructureThe HA directory is located in /opt/CiscoTransportManagerServer/HA. It contains the following directories:

• HA Bin Directory, page 3-33



• HA Configuration Directory, page 3-34

• HA Install Directory, page 3-34

• HA Log Directory, page 3-34

• HA Template Directory, page 3-34

HA Bin Directory

The bin directory contains python scripts: agent scripts that the cluster uses to monitor the critical resources that Prime Optical uses. Table 3-8 lists and describes the scripts.

Table 3-8 HA Bin Directory Scripts

Script Description

ApacheAgent.py Monitors the HTTP Daemon status that Prime Optical uses.

backupdb.py Utility module to perform backup/restore of the database using RMAN.

backupdb.sh This file is moved to /opt/CiscoTransportManagerServer/bin directory after you complete the procedure described in “Installing the HA Package (HA RPM Files)” section on page 3-13.

clusterModule.py Library that provides functionality that other scripts use.

CtmAgent.py Monitors the status of Prime Optical processes.

error_codes.py Utility script for error handling.

FlashBackMonitor.py Monitors the usage of the database FlashBack area. If the usage percentage crosses the threshold (70%; see the value configured in the clusterConfig.ini file) an error message is sent. When the usage is below the threshold, a clear message is sent.

FsCheckAget.py Monitors the file system that Prime Optical uses.

HeartBeat.py Provides the heartbeat functionality between the servers.

InterfaceAgent.py Utility module to check the network interface status.

managed_redoapply.py Activates or deactivates the redo–apply functionality of ODG.

NtwIFAgent.py Uses the InterfaceAgent utility to check the status of the network interface that Prime Optical uses to monitor the NEs. Returns an error code if the link or network interface is down.

odg_console.py Provides an interactive menu to manage the database and data replication functionality.

ODG_IFAgent.py Uses the InterfaceAgent utility to check the status of the network interface that ODG uses to replicate data to the secondary host. Returns an error code if the link or network interface is down.

odg_management.py The script that OracleReplicatorAgent.py agent uses to manage the ODG functionality.

OracleAgent.py Monitors the Oracle processes; in particular, checks the status of System Monitor Process (SMON).

OracleReplicatorAgent.py Monitors the status of Oracle Data Guard (ODG) replication. It returns an error code if ODG is unable to replicate redo logs to the secondary host.

restore_db.sh This file is moved to /opt/CiscoTransportManagerServer/bin directory after you complete the procedure described in “Installing the HA Package (HA RPM Files)” section on page 3-13.

RsyncAgent.py Checks the synchronization on NE backup.



HA Configuration Directory

The configuration directory contains the configuration files. Table 3-9 lists and describes the scripts.

HA Install Directory

The install directory contains the scripts that you must use during the installation and configuration phase.Table 3-10 lists and describes the scripts.

HA Log Directory

The log directory contains the log files created by agents during runtime activities.

HA Template Directory

The template directory contains template files used for high availability. These files must be configured during the configuration phase. The template directory contains the cluster.conf file that describes the resources monitored by RHCS. The cluster.conf file contains the TODO tags that are replaced with proper values during the configuration phase. Once configured, the file is moved under the /etc/cluster directory.

Table 3-9 Configuration Directory Scripts

Script Description

clusterConfig.ini Characterizes parameters that the agents use; created during the configuration phase.

logHAconf.ini Configuration file that agent scripts use for the log activity.

Table 3-10 HA Install Directory Scripts

Script Description

error_codes.py Utility script for error handling.

logODGconf.ini Configuration file that ODG configuration scripts use for log activity.

logRMANconfig.ini Configuration file that RMAN configuration scripts use for log activity.

odg_config.cfg Configures the ODG.

odg_configure.py Configures the ODG using the information provided in the odg_config.cfg file.

odg_tns_config.cfg Contains the configuration for Oracle Listener.

oracle_rman_config.cfg Contains information to configure the Oracle flashback area.

oracle_util.py Utility script that the oracle scripts use.

ora_user_env.py Manages the cporacle user for accessing the database.

README Readme file.

rman_config.py Script to configure RMAN and flashback.

rman_util.py Utility script that the RMAN configuration scripts use.

RsyncDb.py Python script based on rsync to synchronize the primary database with the secondary database.


Chapter 3 Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Maintaining and Monitoring Cluster Activities

GEO Directory StructureThe GEO directory is located in /opt/CiscoTransportManagerServer/HA/GEO. It contains the following directories:

• GEO Bin Directory, page 3-35

• GEO Configuration Directory, page 3-35

• GEO Install Directory, page 3-35

• GEO Log Directory, page 3-35

GEO Bin Directory

The bin directory contains python scripts to manage the GEO Manager console. The following table lists and describes the scripts.

GEO Configuration Directory

The configuration directory contains the configuration files. Table 3-12 lists and describes the scripts.

GEO Install Directory

The install directory contains the GEOService script that provides the GEO monitor service.

GEO Log Directory

The log directory contains the geoManager.log file that traces GEO Manager activities.

Maintaining and Monitoring Cluster ActivitiesMonitoring and maintenance operations on geographical cluster nodes are performed using the following tools:

• GEO Manager—Tool to monitor the single-node geographical redundancy configuration.

Table 3-11 GEO Bin Directory Scripts

Script Description

clusterModule.py (RHCS) Provides functionality that allows cluster command management.

geoClient.py Provides the GEO Server client interface. It is used to monitor the GEO HA solution.

GEO_Server.py Manages the GEOService service.

Table 3-12 GEO Configuration Directory Scripts

Script Description

geo_config.ini Configuration file that allows GEO Manager console access to clusters.

logHAconf.ini Configuration file that agent scripts use for logging GEO Manager activity.



• ODG Console—Tool to monitor and operate the Oracle Data Guard layer.

This section describes the following topics:

• GEO Manager, page 3-36

• ODG Console, page 3-42

• Maintaining the Cluster, page 3-44

GEO ManagerThe GEO Manager tool is based on a dedicated service (GEOServer) and a CLI command console (GEO Manager console). The GEOServer service and GEO Manager console can run on the cluster node or in a separate server (recommended). Figure 3-1, Figure 3-2, and Figure 3-3 show the different types of GEO Manager console configurations.

Figure 3-1 GEO Manager Configuration Example 1

Workstation 1(Linux)

GEO Managerconsole 1


GEO Managerconsole 2

CPOCluster 1

CPOCluster 2

Workstation 3GEOServerservice

3205

59





The following sections describe how to install, configure, and use GEO Manager.

• Installing GEO Manager, page 3-37

• Configuring GEO Manager, page 3-38

• Running the GEO Manager Console, page 3-40

• Starting, Stopping, and Checking the GEOService Service, page 3-40

Installing GEO Manager





For example:


CPOCluster 1

CPOCluster 2


GEO Manager consoleGEOServer service

3205

60

CPOCluster 1

CPOCluster 2

GEO Manager consoleGEOServer service 32

0561









Step 4 Enter the following command to install the GEO Manager:

yum install --nogpgcheck GEOManager-1.5-2.x86_64.rpm

The installation script displays installation progress messages on the screen. When the installation completes you should see the following messages:

Installation DONE!Installed products updated.

Installed: GEOManager.x86_64 0:1.0-1

Complete!

Step 5 Enter the following command to list the files in the /opt/CiscoTransportManagerServer/HA/GEO directory:

ls /opt/CiscoTransportManagerServer/HA/GEO/*

Confirm the following directories and files are displayed:

ls /opt/CiscoTransportManagerServer/HA/GEO/*/opt/CiscoTransportManagerServer/HA/GEO/bin:clusterModule.py clusterModule.pyc geoClient.py GEO_Server.py

/opt/CiscoTransportManagerServer/HA/GEO/cfg:geo_config.ini geo_config.ini.rpmsave logGEOconf.ini

/opt/CiscoTransportManagerServer/HA/GEO/install:GEOService

/opt/CiscoTransportManagerServer/HA/GEO/log:geoManager.log

Configuring GEO Manager

GEO Manager retrieves information from the /opt/Cisco*Server/HA/cfg/geo_config.ini file. Before you start GEO Manager, the geo_config.ini file must be compiled. Table 3-13lists the TODO tags that must be configured with the node configuration.

Note The resource inside brackets (< >) must be initialized with the proper value.



Note • The logging level of the GEO Manager is defined in the logHAconfig.ini file. There is a dedicated logger (logGeoManager) and a dedicate handler (handler_HGeoManager) to define the desired logging level.

• The GEO Manager default port is 8001. Verify the port availability in the server configuration. If it is not available, change it to an available port.

Configuration Parameters for Geo Manager

Perform the following steps for the Configuration of parameters in logGEOconf.in file:

Step 1 Enter into the GEO Manager folder:

cd /opt/CiscoTransportManagerServer/HA/GEO/cfg

Step 2 Open logGEOconf.ini and enter the following command:

vi logGEOconf.ini

Replace the template (mail_address@domain) of the email address in the [handler_mail] section, with a valid value.

Verify SSH Connection

The GEO manager uses the SSH connection to retrieve information from nodes. SSH with root access must be enabled.

To verify the connection, enter the following commands:

ssh "primary-hostname" lsssh "secondary-hostname" ls

Where:

primary-hostname and secondary-hostname are the values in the geo_config.ini file.

If SSH configuration failed, configure SSH where the GEO manager is running and do the following:

Step 1 As the root user, connect to the server in the ~/.ssh directory.

Table 3-13 TODO Tags to Configure

Section Resource Description

GEOServer <hostname> Server hostname running the GEOService service.

HeartBeat <interface> Network interface that GEO manager uses to ping the cluster nodes; for example, eth0 and bond0.

Primary <hostname> Primary node hostname or the name associated to the ODG interface on the primary node.

Secondary <hostname> Secondary node hostname or the name associated to the ODG interface on the secondary node.




cat id_rsa.pub >> authorized_keys

Step 3 Connect SSH on the server and type yes in the prompt (In this scenario, margot is the server).

[root@margot .ssh]# ssh margotThe authenticity of host 'margot (10.58.67.58)' can't be established.RSA key fingerprint is 28:d2:14:66:ca:b1:93:98:24:bc:44:fb:bf:d2:ac:e9.

You will receive the message, "Are you sure you want to continue connecting (yes/no)?" Click Yes.

The following message might appear:

Warning: Permanently added 'margot' (RSA) to the list of known hosts.Last login: Wed Feb 6 11:40:04 2013 from mnza1-dhcp-vl250-144-254-172-242.cisco.com[root@margot ~]#

Step 4 Close the SSH connection and execute a new SSH connection. The connection is established without a password being required.

[root@margot ~]# exitlogoutConnection to margot closed.[root@margot .ssh]#[root@margot .ssh]# ssh margotLast login: Wed Feb 6 11:46:27 2013 from margot.cisco.com[root@margot ~]#

Starting, Stopping, and Checking the GEOService Service

The GEOService service manages the GEO Manager tool. To start, stop, or check the service status, go to the /opt/CiscoTransportManagerServer/HA/GEO/bin directory and enter the following command:

./ GEO_Server.py start/stop/status

Running the GEO Manager Console

The GEO manager provides an interactive console where you can monitor and execute operations on the geographical cluster. To start the console, navigate to the /opt/CiscoTransportManagerServer/HA GEO/bin directory and enter the following command:

service GEOService start/stop/status

For example,

./geoClient.py margot 8001

The GEO Manager console appears with the following commands:

Choose an operation:0. Exit1. show status2. start monitor cluster3. stop monitor cluster4. switchover5. switch nodes role6. freeze service7. unfreeze service8. recovery service9. start service10. stop service



Enter your choice number (monitor is running - False):

Table 3-14 lists the GEO Manager console commands and descriptions.

Table 3-14 GEO Manager Console Commands

Command Description

show status Retrieves status of the node cluster. The GEO manager connects both primary and secondary (as defined in the geo_config.ini file) nodes with the show status command.

start/stop monitor cluster thread

Starts a dedicated thread to monitor the status on both nodes of a geographical redundancy cluster. The thread (as defined in the polling interval property in the geo_config.ini file) periodically checks the connection on a node and the state of ctm_heartbeat and ctm_service services on that node. This check is run on both the primary and secondary nodes.

When the monitor thread is running on the console, you should see the following:

Enter your choice number (monitor is running - True):

If ctm_service fails on the primary node, the monitor thread tries to start the failover procedure. If ctm_service succeeds, the thread is moved to the standby node.

The stop monitor thread command prevents monitor thread exits from the GEO manager script console.

switchover Manages a controlled switchover between the primary and standby nodes. Switchover is allowed only if the monitor thread is not running.

After selecting the switchover command, GEO Manager performs the following steps:

1. Invokes clusvcadm -d ctm_service on the primary node.

2. Invokes clusvcadm -e ctm_service on the standby node.

3. Changes the primary-secondary roles in the geo_config.ini file.

The switch nodes role command changes the primary and secondary roles in the geo_config.ini file.

switch nodes role Updates and changes the primary and secondary nodes in the geo_config.ini file.

The primary and secondary node roles are saved in the geo_config.ini file. If the geo_config.ini file is not in sync with the actual primary and secondary node assignment, the switch nodes role command allows the user to update and correct the configuration file.

Note To view node role information launch the GEO Manager tool and enter the number that corresponds to the show status command.

Freeze Sends an RHCS command to freeze the ctm_service or ctm_heartbeat services on the desired node.

When this option is selected, you are prompted to enter the node and service name.

Unfreeze Sends an RHCS command to unfreeze the service ctm_service or ctm_heartbeat services on the desired node.

When this option is selected, you are prompted to enter the node and service name.

Recovery service Sends an RHCS command to recover the ctm_service or ctm_heartbeat services.

Start service Sends an RHCS command to start the ctm_service or ctm_heartbeat services on the desired node.

An interactive console ask server name and related service name to be started.

Stop service Sends an RHCS command to stop the ctm_service or ctm_heartbeat services on the desired node.

An interactive console ask server name and related service name to be stopped.



ODG ConsoleAn ODG console is created to operate on the ODG subsystem. The ODG console implements the basic commands and must be used only for troubleshooting or to complete the failover operation.

To start the ODG console, enter the following command in the /opt/CiscoTransportManagerServer/HA/bin HA installation root directory:

/odg_console.py

The ODG Console appears with the following commands:

In the Oracle Data Guard for Optical Prime console menu, choose an operation:0—Exit1—Check DatabaseRole and Status2—Check Log Archived and apply report3—Check of general configuration4—Disable the flashback area5—Enable the flashback area6—Shutdown the database7—Startup the database8—Open in mount the database9—Open in Read Write the database10—Enable Redo Apply11—Disable Redo Apply12 Switchover to StandBy13 Switchover to Primary20—Reinstatement of Old Primary (using Flashback)Enter your choice:

Table 3-15 lists the ODG Console commands and descriptions.



Table 3-15 ODG Console Commands

Command Description

Check database role and status

Checks the status of the primary and standby nodes. It is a read-only command.

The following information is displayed for each server:

• Role—Whether the server is physical or primary.

• Open Mode—How the database is opened: read only, read/write, or mount.

• Switchover Status—The state that ODG uses to perform a switchover.

The most important values are:

• Sessions Active

• To Standby

• To Primary

For a complete list and description of possible status entries, see Oracle documentation.

When you select the Check Database Role and Status command, the following output appears:

####### Report Status on Primary ha1-105.cisco.com

Role = PHYSICAL Open Mode = MOUNT Switch over status = TO PRIMARY####### Report Status on Primary ha2-105.cisco.com

Role = PRIMARY Open Mode = READ WRITE Switch over status = TO STANDBY

In this output, the primary geomon server is switched to the standby server. The PHYSICAL role state is displayed as MOUNT mode. It is possible to switch back to the primary server.

Check log archived and apply report

Displays a report about the log archives transferred by the primary server to the standby server. Reports about the alignment of the two databases are also displayed if the command is executed on the standby server.

Check of general configuration

Reports on the basic configuration of the primary server and the standby server. Be sure to check the report to determine if ODG has been configured correctly.

Disable the flashback area

Disables the flashback management area where the backup and the archive logs are stored.

Note This command is for management purposes only.

If the flashback area is disabled, Oracle cannot use it for the flashback operation.

Enable the flashback area

Enables the flashback management area where the backup and the archive logs are stored.

Note This command is for management purposes only.

Shutdown the database

Performs a complete shutdown of the database. A prompt to confirm the operation is shown.

Caution You may experience a loss of service if you use this command improperly.



Maintaining the ClusterThe following sections describe the maintenance operations:

• Maintaining and Optimizing Flashback, page 3-44

• Freezing or Unfreezing the Cluster, page 3-46

• Forcing a Failover, page 3-46

• Realigning the Servers After a Forced Failover, page 3-47

• Backup/Restore, page 3-47

Maintaining and Optimizing Flashback

Flashback maintenance is an important activity that should be done periodically. As a default, an e-mail notification (if configured) is sent when flashback usage surpasses 70% of space availability. To check flashback usage and optimize flashback efficiency:

Step 1 From /opt/CiscoTransportManagerServer/HA/bin, run the following script:

./backupdb.py

Step 2 Enter the number that corresponds to the check flashback usage command.

The percentage of flashback usage is shown.

Startup the database Starts the database and opens it in read/write mode.

Open in mount the database

Starts the database and opens it in the Mount state.


Open in read-write the database

Opens the database in read/write mode.

Enable Redo Apply Automatically starts the Redo Apply real-time functionality to apply the archive log to the standby server.


Disable Redo Apply Automatically stops the Redo Apply real-time functionality to apply the archive log to the standby server.


Switchover to StandBy

Switches over the database from primary to standby server.

Switchover to Primary

Switches over the database from standby to primary server.

Table 3-15 ODG Console Commands

Command Description



Step 3 To check if a previous backup database has been performed, enter the number that corresponds to the Sintetic list of backup in RMAN command. The last database backup is displayed. In the following example, the 27-JUN-13 backup is the last registered database backup:

2013-07-10 17:24:18,799 INFO Show the list of backup ....2013-07-10 17:24:18,799 INFO List of summary Backup stored ....2013-07-10 17:24:23,271 INFORecovery Manager: Release 11.2.0.3.0 - Production on Wed Jul 10 15:24:21 2013

Copyright (c) 1982, 2011, Oracle and/or its affiliates. All rights reserved.

RMAN>connected to target database: CTM (DBID=3731458588, not open)

RMAN>using target database control file instead of recovery catalog

List of Backups===============Key TY LV S Device Type Completion Time #Pieces #Copies Compressed Tag------- -- -- - ----------- --------------- ------- ------- ---------- ---1 B 0 A DISK 27-JUN-13 1 1 YES TAG20130627T1452182 B 0 A DISK 27-JUN-13 1 1 YES TAG20130627T1452183 B 0 A DISK 27-JUN-13 1 1 YES TAG20130627T1452184 B 0 A DISK 27-JUN-13 1 1 YES TAG20130627T1452185 B 0 A DISK 27-JUN-13 1 1 YES TAG20130627T1452186 B A A DISK 27-JUN-13 1 1 YES TAG20130627T1453257 B A A DISK 27-JUN-13 1 1 YES TAG20130627T1453258 B A A DISK 27-JUN-13 1 1 YES TAG20130627T1453259 B A A DISK 27-JUN-13 1 1 YES TAG20130627T14532510 B A A DISK 27-JUN-13 1 1 YES TAG20130627T14532511 B A A DISK 27-JUN-13 1 1 YES TAG20130627T14532512 B A A DISK 27-JUN-13 1 1 YES TAG20130627T14532513 B A A DISK 27-JUN-13 1 1 YES TAG20130627T14532514 B A A DISK 27-JUN-13 1 1 YES TAG20130627T14532515 B A A DISK 27-JUN-13 1 1 YES TAG20130627T145325

RMAN>

Recovery Manager complete.

2013-07-10 17:24:23,275 INFO End the summary list of backup2013-07-10 17:24:23,275 INFO End show the list of backup ....

Step 4 Delete the expired database backup. Enter the number that corresponds to the Delete the expired backup command.

Step 5 Obsolete the database backup. Enter the number that corresponds to the Delete the expired backup command.

Step 6 Start a database backup. Enter the number that corresponds to the Backup command.

Step 7 Purge the obsolete archivelog. Enter the number that corresponds to the Purge obsolete archives command.

Step 8 Confirm flashback usage by comparing current usage with previous one. Enter the number that corresponds to the Check flashback usage command.



Freezing or Unfreezing the Cluster

For maintenance operations on a specific resource, you must freeze the cluster. Freezing the cluster allows you to stop a specific resource. Otherwise, the RHCS will detect a fault on that resource.

To freeze or unfreeze the cluster, follow these steps:

Step 1 To freeze a service in the cluster, enter the following command:

clusvcadm -Z service

The service may be ctm_heartbeat or ctm_service.

A frozen service shows a [Z] value in the state column when you invoke the clustat command.

Step 2 To unfreeze a service in the cluster, enter the following command:

clusvcadm -U service

Forcing a Failover

There are some instances where an automatic failover fails to occur even if the GEO Manager is running with an active monitor thread command.

The following list describes possible scenarios in which this might occur:

• Crash of a primary node.

• Crash of Oracle processes in the primary node.

• Ctm_service is down on the primary node and there is no connection between the secondary node and the primary node.

In all of these scenarios, the ctm_service is started on a standby node, but it is unable to contact Oracle on the other node (the old, primary in this case). Thus, it fails to inform ODG to prepare the database to receive the redo logs from the other node. When this happens, the ctm_service startup fails and the ctm_service is moved to a failed state by RHCS. (You can check the state of ctm_service by entering the clustat command.)

For these types of scenarios, you can manually restore Prime Optical on the secondary node. With this procedure, you can split the geographically redundant cluster; then, when you restore ctm_service on the secondary node, the data will not replicate to the other node of the cluster (the other node will not be reachable).

To manually restore Prime Optical on the secondary node, follow these steps:

Step 1 Disable the ctm_heartbeat and ctm_service by executing the following commands:

clusvcadm -d ctm_heartbeatclusvcadm -d ctm_service

Step 2 Edit the clusteConfig.ini file by setting the forced_failover property to True; then save the file.

Step 3 Enable ctm_service by executing the following command:

enable ctm_serviceclusvcadm -e ctm_service



The two nodes of the geographically redundant cluster are not aligned after these steps are completed. Once you recover the primary node, you must then realign it with the secondary node. See Realigning the Servers After a Forced Failover, page 3-47 for more information.

Realigning the Servers After a Forced Failover

You can realign cluster nodes that are not aligned after a forced failover caused by:

• Crash of the primary node. See Forcing a Failover, page 3-46 for more information.

• Recovery of a server on the faulty primary node.

• ctm_service running on the secondary node, with the database in both the primary and the secondary node not aligned.

To realign the primary and standby servers after a forced failover, follow these steps:

Step 1 From the primary node (which will become the standby node), enter the following command in the /opt/CiscoTransportManagerServer/HA/bin directory:

./odg_console.py

Step 2 From the ODG Console:

• Choose Reinstatement of Old Primary (using Flashback) to realign the primary server to the standby server.

• Choose Open in mount the database to mount the database.

Step 3 From the primary cluster, enable ctm_heartbeat by executing the following command:

clusvcadm -e ctm_heartbeat

Backup/Restore

With the introduction of Flashback and archived log file management for HA on Linux servers, normal Prime Optical backup and restore procedures cannot be used.

The backup and restore procedures for HA on Linux servers are described in the following sections:

• Backup/Restore Overview, page 3-48

• Backup/Restore Command Console, page 3-48

• Backing Up the Database in an Oracle Data Guard Setup, page 3-50

• Restoring the Database on a Single Server, page 3-51

• Restoring the Database to a Specific Backup on a Single Server, page 3-52

• Restoring the Standby Server on the Cluster, page 3-53

• Restoring the Standby Server to a Specific Backup, page 3-53



Backup/Restore Overview

The backup/restore operation for HA on Linux servers does not use the previous Prime Optical backupdb.sh and restore_db.sh scripts. The new script inhibits any operation and trace in the DB_HISTORY_TABLE and in the log file in which the backup with this modality is disabled. The backup/restore operation fails if done from the Prime Central portal.

Note To avoid errors, the original script is substituted in the installation procedure of the HA on Linux server package.

The new procedure for backup is based on RMAN, the official tool for Oracle Recovery Manager.

The available backup operations are:

• Cold backup

• Restore the database of the last previous backup

• Recover the database to the last transaction

• Recover the database to a specific time

• List of backups

• Synthetic list of backups

The basic configuration of RMAN for the backup is in the oracle_rman_config.cfg file located in the /opt/CiscoTransportManagerServer/HA/install directory.

You can see the following logs in the /ctm_backup directory:

• reportRMAN.log—RMAN utility report

• logRMANOptPrime.log—Log of backupdb.py utility

Note The primary and secondary servers must be restored with the same backup, and their database must be synchronized. To restore the standby server, complete the steps described in Restoring the Standby Server on the Cluster, page 3-53 and Restoring the Standby Server to a Specific Backup, page 3-53.

Backup Guidelines for ODG

To save disk space in your archiving location, you may want to back up archived logs to tape or to an alternative disk location. Oracle database recovery robustness is based on the archivelog files located in the Prime Optical configuration under the /ctm_backup directory.

The guidelines are as follows:

• You cannot restore/recover the database to the last transaction if the archivelog files are deleted.

• Low disk space due to the presence of a high number of archivelog file may cause Prime Optical to stop.

• Perform regular backups to remove unnecessary archivelog files.

• Before starting a backup, remove the old backup even if it has been compressed.

Backup/Restore Command Console

From the /cisco/PrimeOpticalServer/HA/bin directory, enter the following command to access the Backup/Restore console:

/backupdb.py



Table 3-16 lists the field descriptions for the backup/restore commands.

Table 3-16 Available Backup/Restore Commands

Command Name Description Restriction

Exit Exits from functionality. —

Backup Performs a cold backup. —

Restore Performs a restore to the last backup.

—

Complete Recover Performs a complete recovery of the database to the last transaction.

—

Preview the backup to restore

Displays a report about the last backup to restore. Indicates the System Change Number (SCN) that is possible to recover the database.

—

Extract the previous version for the rollback

Reports the list of the incarnations. —

Rollback the database to previous version

Sets the database incarnation.

The incarnation ID is extracted via the Extract the previous version for the rollback command.

Use this command with caution. The database is reverted to an older configuration and may cause ODG service loss.

Recover to a specific backup

Recovers the database to an old backup using the SCN extracted with the Preview the backup to restore command.

—

Open database Opens the read/write database. —

Open database after recover

Opens the read/write database with resetlogs.

—

Shutdown database Shuts down the database. This operation is critical.

Startup database Starts up the database. —

Configure RMAN on Primary

Configuring the Primary Server, page 3-21.

This command must be entered only once during the installation.

Configure RMAN on StandBy

Performing a Cold Backup, page 3-22.

This command must be entered only once during the installation.

Show Actual Configuration of RMAN

Shows the actual configuration of RMAN for the backup.

—



Backing Up the Database in an Oracle Data Guard Setup

To back up the database, follow this process:

• Performing the First Backup, page 3-50

• Performing Subsequent Backups, page 3-51

Performing the First Backup

Make sure to perform a backup on the standby server to avoid overloading the active, because the standby and active databases are exact copies of each other. For more information on ODG guidelines, see Backup Guidelines for ODG, page 3-48.

Note After the backup is completed, all archive log files stored in the /ctm_backup directory with a time stamp from before the backup are of no use once the backup is complete.

To perform the first backup:

Step 1 As a root user, connect to the standby server.

Step 2 Enter the following commands to overwrite the existing backup and restore scripts:

Purge obsolete archivelogs

Purges the obsolete archive logs. The disk space reserved for Flashback is not sufficient; hence, this operation should only be performed if absolutely necessary.

The obsolete archive logs are related to the retention window.

List backup in RMAN Lists backups present in the system. A retention window of 7 days is available for now.

Sintetic list of backup in RMAN

Lists backups present in the system. —

Check status of Flashback

Checks if the flashback is on or off. —

Delete the expired backup

Deletes the expired backup. —

Delete the obsolete backup

Deletes the obsolete backup. A backup is obsolete if it is past the retention period.

Backup archivelogs Perform the backup of all archive logs.

—

Crosscheck backup Checks the status and whether or not the backup is valid for a restore

At least one backup must be performed.

Check flashback usage Checks the usage percentage of the flashback area.

The flashback must be enabled.

Table 3-16 Available Backup/Restore Commands

Command Name Description Restriction



cd /opt/CiscoTransportManagerServer/HA/bin

Step 3 Run the following script:

./backupdb.py

Step 4 Enter the number that corresponds to the Backup command.

Wait for the backup to finish; this can take a long time. You can check the progress in the /ctm_backup/reportRMAN.log stored in the /ctm_backup directory

Step 5 After the backup is complete, purge the obsolete archive log files from the /ctm_backup area. Enter the number that corresponds to the Purge obsolete archivelogs command.


Performing Subsequent Backups

It is essential to remove any old backups to free up disk space for the new backup. See Backup Guidelines for ODG for more information.

To perform subsequent backups:

Step 1 As a root user, connect to the standby server.

Step 2 Enter the following command to overwrite the existing backup and restore scripts:

cd /opt/CiscoTransportManagerServer/HA/bin

Step 3 Run the following script:

./backupdb.py

Step 4 To check the previous backups, enter the number that corresponds to the Sintetic list of backup in RMAN command.

If the Sintetic list of backup in RMAN command, shows a list of information, you can ensure that the backup has taken place in /ctm_backup directory.

Step 5 To clean the old backup, enter the number that corresponds to the Delete the obsolete backup command.

Step 6 To perform the backup, enter the number that corresponds to the Backup command.

The process takes a lot of time. Wait till the backup is complete. You can check the progress in the /ctm_backup/reportRMAN.log stored in the /ctm_backup directory. After the backup is complete, purge the archive log files in the ctm_backup directory.

Step 7 To purge the archive logs, enter the number that corresponds to the Purge obsolete archivelogs command.


Restoring the Database on a Single Server

Note Before you start database recovery, you must first restore the data files stored in the backup directory.

To recover the database, follow these steps:



Step 1 Enter the following command to freeze the ctm_service:

clusvcadm -Z ctm_service

Step 2 Enter the following command to stop the Prime Optical server:

opticalctl stop

Step 3 From the /cisco/PrimeOpticalServer/HA/bin directory, enter the following command:

./backupdb.py

Step 4 Enter the number that corresponds to the Restore command:

You can view the report in the /ctm_backup directory.

Step 5 To perform a complete recovery of the database to the last transaction, enter the number that corresponds to the Complete Recovery command.

Step 6 Enter the number that corresponds to the Open database command.


Step 8 Enter the following command to start the Prime Optical server:

opticalctl start

Step 9 Enter the following command to unfreeze ctm_service:

clusvcadm -U ctm_service

Restoring the Database to a Specific Backup on a Single Server

Note Before you start database recovery, you must first restore the data files stored in the backup directory.

To recover the database to a specific backup, follow these steps:

Step 1 Enter the following command to freeze the ctm_service:


Step 2 Enter the following command to stop the Prime Optical server:

opticalctl stop

Step 3 From the /cisco/PrimeOpticalServer/HA/bin directory, enter the following command:

./backupdb.py

Step 4 Enter the number that corresponds to the Restore command:

You can view the report in the /ctm_backup directory.

Step 5 Enter the number that corresponds to the Preview the backup to restore command to extract the SCN:

You can extract the SCN by generating the report using the SCN in the following message:

Recovery must be done beyond SCN <> to clear datafile fuzziness.

Step 6 Launch the ODG console by executing the following command:

./odg_console.py



Step 7 Enter the number that corresponds to the Disable the flashback area command.

Step 8 To exit the ODG console, enter the number that corresponds to the Exit command.

Step 9 Launch the backup utility by entering the following command:

./backupdb.py

Step 10 To recover the database using the SCN extracted in Step 5, enter the number that corresponds to the Recover to a specific backup command.

Provide inputs to SCN when required.

Step 11 Check the report in the /ctm_backup directory. If all of the processes appear normal, proceed to Step 15. Otherwise, proceed to Step 12.

Step 12 Enter the number that corresponds to the Extract the previous version for the rollback to extract the right level of incarnation to reestablish. (The backup requires an old incarnation level if the extracted SCN is old.)

Step 13 Enter the number that corresponds to the Rollback the database to previous version command to revert to the correct database incarnation level.

Step 14 Perform the restore again by following Step 4 through Step 11.

Step 15 Enter the number that corresponds to the Open database after recover command to open the database with resetlogs.

Step 16 To start the Prime Optical server, enter the following command:

opticalctl start

Step 17 Enter the following command to unfreeze ctm_service:

clusvcadm -U ctm_service

Restoring the Standby Server on the Cluster

Note Before you start the standby server recovery, you must first restore the data files stored in the backup directory.

To perform a complete recovery of the standby server, do the following:

1. Restore the database to the standby server. Follow the steps in Restoring the Database on a Single Server, page 3-51.

2. Share the backup area of the standby server with the primary server by performing a switchover using the GEO console, so that the standby server becomes the primary server. See Table 3-14 on page 3-41 for more information on the switchover command.

3. Proceed to the standby server (previously the primary server) and perform the steps in Restoring the Database on a Single Server, page 3-51.

Restoring the Standby Server to a Specific Backup

Note Before you start the standby server recovery, you must first restore the data files stored in the backup directory.



To perform a complete recovery of the standby server to a specific backup, do the following:

1. Recover the standby database to a specific backup. See Restoring the Database to a Specific Backup on a Single Server, page 3-52.

2. Share the backup area of the standby server with the primary server by performing a switchover using the GEO console, so that the standby server becomes the primary server. See Table 3-14 on page 3-41 for more information on the switchover command.

3. Proceed to the standby server (previously the primary server) and perform the steps in Restoring the Database to a Specific Backup on a Single Server, page 3-52.


Cisco Pri

C H A P T E R 4
Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Linux Server
This section contains the following topics:

• Local Redundancy Configuration Checklists

• Setting the Environment for Installation

• Verifying Packages and yum setup

• Installing Oracle 12c (for Oracle Not Embedded Database Setup)

• Installing Prime Optical

• Installing the HA Package




• Starting the Cluster

• Installing the Multicast Heartbeat Services Package

For more information about local redundancy configuration on Linux, including hardware configuration and network diagrams; see Local Redundancy Configuration, page 1-1.

Local Redundancy Configuration Checklists This section provides checklists to help with Prime Optical HA on Linux installation tasks. We recommend that you print the checklists from the PDF, which shows the section number where you will find each task, and either check off tasks as you complete them or enter information as needed.

This section contains the following information:

• Preinstallation Checklist

• Local Redundancy Configuration Installation Checklist


Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Local Redundancy Configuration Checklists

Preinstallation ChecklistBefore you install the Prime Optical HA solution in a local redundancy configuration, complete the field in Table 4-1. The information will assist you in completing the installation.

Table 4-1 is the preinstallation checklist for a local redundancy configuration.

Note Private interface devices are dedicated to the heartbeat connection.

Local Redundancy Configuration Installation ChecklistTo help you keep track of the steps in the installation process, print the checklist from the PDF, which shows the section number where you will find each task, and check the appropriate cells in Table 4-2 as you complete each step in the HA installation configuration process. In the cells where N/A is displayed, the step is not applicable for the primary and secondary server.

Table 4-2 is the local redundancy configuration checklist.

Table 4-1 Preinstallation Checklist for a Local Redundancy Configuration

Information Primary Cluster

Primary site name:

Local cluster name:

Local cluster number:

Prime Optical Virtual IP address:

Primary Server Secondary Server

Hostname:

IP address:

Public interface device 1:

Public interface device 2:

Private interface device 1:

Private interface device 2:

Table 4-2 Local Redundancy Configuration Checklist

Description See...Primary Server Secondary Server

New Installation

1 Mount shared storage on primary server and configure virtual IP address

Setting the Environment for Installation, page 4-3

x

2 Run server and database installation wizard.

Installing Prime Optical, page 4-7 x

3 Install HA RPM files. Installing the HA Package, page 4-11 x


Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Setting the Environment for Installation

Setting the Environment for InstallationIn this procedure you will mount the shared disks between the primary and secondary servers and configure the shared virtual IP address that will be used during installation and runtime.

Step 1 On the primary server, mount the external disk that is shared between the primary and secondary servers to the appropriate mountpoints. See Table 2-4 and Table 2-5 for disk space and partition requirements.

For example:

mount /dev/sde /db01mount /dev/sdf /db02mount /dev/sdg /db03mount /dev/sdh /db04mount /dev/sdi /db05mount /dev/sdj /ctm_backupmount /dev/sdk /ciscomount /dev/sdt /oracle

4 Configure HA on Linux • Editing the Agent Configuration File, page 4-12

• Editing the Logging Configuration File, page 4-13

• Editing the RHCS Configuration File, page 4-14

x x

Note Only to copy RHCS configuration file.

5 Start the cluster • Starting the Cluster, page 4-16




x

6 Install Multicast Heartbeat Services • Installing the Multicast Heartbeat Services Package, page 4-18

• Editing the multicastConfig.ini File, page 4-19

• Editing the logMulticast.ini File, page 4-19

• Starting the Multicast HeartBeat Services, page 4-20

• Configuring the MultiCast Heartbeat Services on the Secondary Server, page 4-20

x x

Table 4-2 Local Redundancy Configuration Checklist (continued)

Description See...Primary Server Secondary Server


Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Verifying Packages and yum setup

Note • You can enter the fdisk command to view and confirm which disks are available and connected to the disk array: fdisk -l.

• Verify that /cisco, /oracle, /db01, /db02, /db03, /db04, /db05, and /ctm_backup are mounted correctly. Enter the df -h command to verify that the file systems are mounted on the primary server.

Step 2 Enter the following command to enable the virtual IP address that will be used during the HA solution runtime. This is the shared IP address mounted to an interface on the primary server.

ifconfig interface virtual-IP-address netmask netmask

For example:

ifconfig eth0:1 10.58.65.22 netmask 255.255.255.0

You can verify if the IP address was configured correctly by issuing the ifconfig command. The IP address for the interface should be displayed in the inet addr field.

Verifying Packages and yum setupVerify that the following packages are installed on the primary and secondary servers:

The RHCS packages—Available on the Red Hat Enterprise Linux (RHEL) CD under the /cluster directory. See Installation Requirements for supported RHEL version.

• The list of packages includes:

– cman

– rgmanager

– openais

– modcluster

– ricci

– luci

– cluster-cim

– cluster-snmp

– system-config-cluster (only for RHEL 5.8)

• rgmanager-3.0.12.1-21.el6.x86_64

Step 1 Verify that the required packages are installed. Enter one of the following commands:

rpm -q pkg

or

yum info pkg

The output displays the package details. If a required package is not installed, an error will appear.


Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Verifying Packages and yum setup

Step 2 Use the yum command to install the Linux package. The yum command checks all the dependencies from other packages. If you have an active connection to the Red Hat website, all dependent packages are automatically installed on the server. For more information, see the Red Hat website.

If you do not have an active connection to the Red Hat website, do the following:

a. Mount the RHEL DVD or ISO file. See Installation Requirements for supported RHEL version. For example:

mount -o loop -t iso9660 /mnt/redhat/rhel-server-6.6-x86_64.iso /iso

where /mnt/redhat/rhel-server-6.6-x86_64.iso is the Red Hat ISO file.

b. Enter the following commands to get the media ID:

view /iso/.discinfocat /iso/.discinfo


1269263646.691048Red Hat Enterprise Linux Server release 6.6 (Santiago)x86_641,2,3,4,5,6

Server/baseServer/RPMSServer/pixmaps

The media ID in the preceding example is 1269263646.691048.

c. Copy the proper local.repo file template from DVD-mount-point/Disk1 to /etc/yum.repos.d directory.







Open and follow the instructions in the local.repo file.

Step 3 Enter the following command to verify that the local repository has been set up correctly:

yum repolist


Loaded plugins: rhnplugin, securityLocal | 1.3 kB 00:00Local/primary | 868 kB 00:00Local 3116/3116localHA | 1.3 kB 00:00localHA/primary | 6.1 kB 00:00localHA 32/32repo id repo name statusLocal Local Media Repo enabled: 3,116localHA Local HA Media Repo enabled: 32rhel-x86_64-server-5 Red Hat Enterprise Linux (v. 5 for 64-bit x86_64) enabled: 14,137repolist: 17,285


Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Installing Oracle 12c (for Oracle Not Embedded Database Setup)

Step 4 Enter the following command for all the missing packages:

yum install <name_of_rpm_file>

For example, for rgmanager, enter the following command:

yum localinstall rgmanager-3.0.12.1-21.el6.x86_64

Step 5 The cman and rgmanager configurations are enabled after you boot the server. To verify that cman and rgmanager are enabled, enter the following commands:

chkconfig --list cmanchkconfig --list rgmanager

The following output appears when cman or rgmanager is enabled (levels 2, 3, 4, and 5 show the on state):

cman 0:off 1:off 2:on 3:on 4:on 5:on 6:offrgmanager 0:off 1:off 2:on 3:on 4:on 5:on 6:off

When cman or rgmanager is not enabled, all the levels show the off state.

Step 6 Enter the following commands to enable the package:

chkconfig --level 2345 cman onchkconfig --list cmanchkconfig --level 2345 rgmanager onchkconfig --list rgmanager

Step 7 Enter the following command to check the openais service status:

service openais status

If the service is running, enter the following command to stop the openais service.

service openais stop

Step 8 Enter the following command to confirm that the openais service is not automatically started at boot time:

chkconfig openais --list

The result should look like this:

openais 0:off 1:off 2:off 3:off 4:off 5:off 6:off

Step 9 If any of the levels are set to on, enter the command to disable the service. For example, if levels 2,3,4, and 5 are set to on, enter the following command:

chkconfig --level 2345 openais off

Installing Oracle 12c (for Oracle Not Embedded Database Setup)To install Oracle 12c, follow these steps:

Step 1 Install Oracle 12c. Complete Oracle Prerequisites, page A-1 and Installing the Oracle 12c Software with the Response File (*.rsp) Provided by Cisco, page A-3.


Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Installing Prime Optical

Step 2 Complete Downloading and Installing the Required Oracle 12c Patch for 64-Bit Linux Platforms, page A-5.

Step 3 Edit the listener.ora file by replacing the newdbname parameter with the Oracle SID (the default is CTM) and the Prime Optical hostname parameter with the hostname or IP address of the workstation where the Prime Optical database will run.

Note There are multiple instances of newdbname. You must replace all instances with the Oracle SID.

Installing Prime OpticalTo install Prime Optical:

Note This installation is performed only on the primary server because the primary and secondary servers share the disk array where the Prime Optical server and database are installed.

Step 1 Log in as the root user.


echo $DISPLAY

In the command output, you should see:


Step 3 If you are using an xterm window or a remote host, enter the following command to enable the xterm connection from the clients:

xhost +

Step 4 Enter the following command for Linux only if the device is not automatically mounted:

a. Check if the USB device is discovered by OS and enter the following command:

fdisk -l

The fdisk output is as follows:

Disk /dev/sda: 1167.9 GB, 1167996223488 bytes255 heads, 63 sectors/track, 142000 cylindersUnits = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System/dev/sda1 * 1 13 104391 83 Linux/dev/sda2 14 142000 1140510577+ 8e Linux LVM Disk /dev/sdd: 32.0 GB, 32027705344 bytes64 heads, 32 sectors/track, 30544 cylindersUnits = cylinders of 2048 * 512 = 1048576 bytesDisk /dev/sdd doesn't contain a valid partition table

b. Mount the USB device and enter the following command:

mount -t ext3 <USB device> /<USB-mount-point>



For example:

mount -t ext3 /dev/sdd /media





Step 6 Enter the following command to start the installation:

cd /mnt/mnt/Disk1/setup.sh

The following happens:

1. The setup program searches for Sun Microsystems JDK Version 1.7.0_80 on your workstation.

2. For Linux, if the installed release is not Red Hat 5.x, Red Hat 6.x or Red Hat 7.x, the following message appears:

WARNING: Installed Linux version (`cat /etc/redhat-release`) is different from the required one.Do you want to Continue? [y] :

If you enter y, the following message appears:

Continuing Cisco Prime Optical server installation.

If you enter n, the installation quits.

The Cisco Prime Optical Server Installation wizard appears. Wait for up to 90 seconds while the following message appears:

Continuing Cisco Prime Optical Server Installation.

Step 7 At the Introduction screen, click Next.

Step 8 At the License Agreement screen, read the license agreement and click the I accept the terms of the license agreement radio button. Click Next.

Step 9 At the Configure the Server and Database screen, do the following:

a. From the list of Prime Optical installation types, select the Prime Optical server and database radio button to install the Prime Optical server along with the database.

The Database field is disabled when the server and database are installed on the same workstation. Prime Optical automatically uses the server hostname or IP address.

b. Type the Virtual IP or related hostname for the Prime Optical Server and then click Next.

Step 10 At the Configure the Database screen, select Embedded to install Prime Optical with an embedded database.

When you select Embedded as the Oracle database installation type, the rest of the fields in the screen become disabled, and you can move to the next screen.

The following message appears:

This installation will remove any previous Oracle database installed on the server.If you do not want to continue, click Cancel.



Click Continue to delete the previous Oracle version and continue the installation. (If you click Cancel, the installation quits.)

Step 11 At the OS Users screen, choose the root user from the drop-down list. (The password for optusr is Ctm123!. You can change the password later using the passwd -u optusr command.)

Step 12 At the Select Network Configuration screen, specify the size of your network and then click Next.

At the Oracle Pre-Installation Checks screen, the setup program checks the RAM, swap, and temp directory sizes. The screen indicates whether you have enough space to install Prime Optical and Oracle, and whether any required packages are missing.

Step 13 Review the preinstallation checks and then click Next.

• If errors are discovered during the check sizes phase of the preinstallation checks, the following message appears, and the installation quits:

An error occurred during the pre-installation check./temp_CTM/report_oracle_pre_install_checks.log.The installation has been canceled.

Check the log file, correct any errors noted in the file, and restart the installation.

• During the check packages phase of the preinstallation checks, the following message appears:

A warning occurred during the preinstallation check. See /temp_CTM/report_oracle_pre_install_checks.log. To continue the installation, click Continue. To cancel the installation and fix the problem, click Cancel.

You can choose to continue installation, or quit and fix the problem before proceeding.

Step 14 At the Optional Features screen, you can choose additional features from the following options:

• Install Prime Optical in a High Availability Setup

If you choose this option, provide details in the Second High Availability Server field. If you chose optusr as the Prime Optical user, the following message appears:

The Prime Optical OS user must be the root user for High Availability setup. Click Continue to proceed, or click Cancel to change the selection.

If you choose to continue, the Prime Optical user is automatically changed to root.

• Configure FTP (ONS15216 EDFA3)

• Install Sudo Software (available for root users only)

Step 15 If you selected the Install Sudo Software option, at the Prime Optical Group Information & Sudo Installation screen, do the following:

a. Enter the name of the UNIX group to which you want to assign administrator privileges.

b. To install sudo, check the Install Prime Optical Sudo check box. If you do not want to install sudo, uncheck the check box.

Note If you already installed and configured sudo with rules in a file different from /etc/sudoers (for example, if you created a custom sudo configuration in the /user/local/etc/sudoers file), then you must copy the rules from /etc/sudoers to /user/local/etc/sudoers at the end of the installation.

c. Click Next.

Step 16 If you selected the Configure FTP (ONS15216 EDFA3) option, at the FTP Information screen, do the following to configure an FTP account for software download operations:



Note NE types that require FTP configuration are:

• Cisco ONS15216 EDFA3

• Cisco ONS15305 with release lower than 3.0 (CEC-based 15305 NEs)

a. Enter the following information:

• Username

• Password

• Confirm Password

• FTP directory

b. Check or uncheck the Create new FTP account check box.

If checked, the installation script automatically creates the FTP user on the Prime Optical server workstation. If unchecked, it is assumed that an FTP user already exists on the Prime Optical server workstation.

c. Click Next.

Step 17 At the Destination Folder screen, install the Prime Optical server in the default directory. The default directory is /cisco/PrimeOpticalServer. Click Next.

Note If the destination directory that you specified is a new directory, you will receive the message, “The specified directory does not exist. Do you want to create it?” Click Yes.

Caution Do not specify a mount point as the target installation directory for the Prime Optical server installation, or the installation data will be lost when the workstation restarts. You must create a dedicated Prime Optical subdirectory. For example, if /cisco is the mount point for the disk partition /dev/dsk/c0t0d0s5, you cannot specify /cisco as the Prime Optical installation directory. Instead, specify a dedicated subdirectory such as/cisco/PrimeOpticalServer.

Caution Do not delete any instances of /opt/CiscoTransportManagerServer from your Prime Optical file structure. Prime Optical checks for the /opt/CiscoTransportManagerServer directory or a symbolic link to it. If Prime Optical cannot find the /opt/CiscoTransportManagerServer directory or a symbolic link, Prime Optical creates a symbolic link automatically.

The Pre-Installation Summary screen shows the items that will be installed.

Step 18 Click Install.

Depending on your system performance, it might take 35 to 50 minutes for Linux operating systems.

If SSH is not configured, a warning message appears.

Step 19 Click Continue; in the terminal window that appears, enter the root password and follow the prompts to configure the SSH connection.

The installation continues.


Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Installing the HA Package

Step 20 In the Web Server Installation Summary window, click Next.

The Install Complete window summarizes the results of the installation.

Step 21 Click Done.

Caution Do not close the terminal or reboot the host before the following message is displayed: Prime Optical installation is complete.

Step 22 Enter the following command to safely remove the USB drive from the workstation:

umount /mnt

Step 23 Enter the following command to shut down the interface:

ifconfig interface down

For example:

ifconfig eth0:1 down

Step 24 On the secondary server, log in as the root user.

Step 25 Enter the following command to configure the virtual IP address used in Setting the Environment for Installation:

ifconfig interface virtual-IP-address netmask netmask

For example:

ifconfig eth0:1 10.58.65.22 netmask 255.255.255.0

You can verify if the IP address was configured correctly by issuing the ifconfig command. The IP address for the interface should be displayed in the inet addr field.

Step 26 Verify that the yum repolist is available on the server, and then enter the following script from the /temp_CTM directory:

installHAServer.sh

A terminal window appears. Before continuing, perform a check on /root/.ssh/known_hosts.

Step 27 Enter the root password and follow the prompts to configure the SSH connection on the secondary server.

Step 28 Enter the following command to shut down the interface:

ifconfig interface down

For example:

ifconfig eth0:1 down

Installing the HA Package To install the Prime Optical HA package, follow these steps:



Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Editing the Agent Configuration File




For Example:









yum install --nogpgcheck HA-1.5-3.x86_64.rpm


Preparing...######################################## [100%]1: HA######################################## [100%]Installation DONE!

If the installation was successful, you should see the directory structure when installation is complete. Navigate to /opt/CiscoTransportManagerServer/HA directory. See HA Directory Structure, page 3-32 for more information on the directory structure.

Editing the Agent Configuration FileThe agent configuration file allows you to configure the agents that RHCS uses. To edit the agent configuration file:


cd /opt/Cisco*Server/HA/cfg

Step 2 Edit the clusterConfig.ini file. Enter the following command:

vim clusterConfig.ini

Step 3 For local redundancy configuration, do the following:


Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Editing the Logging Configuration File

Note Only two fields need to be edited.

a. Specify the cluster type. Replace type=TODO_ClusterType with type=LOCAL.

b. Specify the network interface that Prime Optical uses to retrieve data from a monitored network element. Replace interface=TODO_interface with interface=interface.

For example:

interface=eth0.

Note This interface is a critical resource; we strongly recommend that you protect it by bonding interfaces on different NICs; for example, eth0, eth1, and bond0.

Step 4 Save the file.

Note For more information on the clusterConfig.ini file, see Table 3-5 on page 3-24.

Editing the Logging Configuration FileWhen the cluster agent detects a failure in a resource, any critical error is mailed to a receiver that is configured in the logHAconf.ini file.

To edit the file:


cd /opt/Cisco*Server/HA/cfg

Step 2 Edit the logHAconf.ini file. Enter the following command:

vim logHAconf.ini

Step 3 Specify the mail receiver. Under the [handler_mail] header, replace <hostname> and <mail_address@<domain> in the following line:

args=(‘localhost’, ‘root@cpo<hostname>, [‘<mail_address@<domain>’], ‘CTM High Availability notification’

for example,

args=(‘localhost’, ‘[email protected]’, [‘[email protected]’], ‘CTM High Availability notification’

Step 4 Save the file.

Note For more information on the clusterConfig.ini file, see Table 3-6 on page 3-25.


Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Editing the RHCS Configuration File

Editing the RHCS Configuration FileIn /opt/Cisco*ver/HA/template there are three different cluster configuration template for local configuration:

• Cluster.conf.local

This is the base configuration without any fencing device. It is not recommended for the production environment.

• Cluster.conf.local.fenceipmi

This template contains a customizable configuration for fence device base in IPMI (Intelligent Platform Management Interface).

• Cluster.conf.local.fencevmware

This template contained a customizable configuration for fence based on VMWARE.

For details about fence device and configuration see the RHCS documentation.

Choose the template that match you environment and edit it substituting all the "TODO" tags.

To edit the RHCS file:


cd /opt/Cisco*Server/HA/template

Step 2 Edit the cluster.conf.local file. Enter the following command:

vim cluster.conf.local

Step 3 Replace all tags listed in Table 4-3.

Table 4-3 RHCS Configuration Parameters


TODO_alias Alias name of the cluster. It can be any string. building1_cluster

TODO_name Name of the cluster. building1_cluster

TODO_Node1 Local cluster name on the primary server. clusterA-abc.cisco.com

TODO_Node2 Local cluster name on the secondary server. clusterB-abc.cisco.com

TODO_Failover_name Failover domain name of the cluster. building1_domain

TODO_VirtualIP Enter the virtual IP address that was configured in Setting the Environment for Installation, page 4-3.

10.58.65.22

TODO_NetMask Enter the netmask used. /24

TODO_Dev_cisco1 Enter the disk that you associated with /cisco. See example in Setting the Environment for Installation, page 4-3.

/dev/sdk

TODO_Dev_oracle1 Enter the disk that you associated with /oracle. See the example in Setting the Environment for Installation, page 4-3.

/dev/sdt


Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Editing the RHCS Configuration File

The following is an example of a cluster.conf.local file:

<?xml version="1.0"?><cluster alias="local" config_version="3" name="local">

<fence_daemon clean_start="1" post_fail_delay="20" post_join_delay="6"/><clusternodes>

<clusternode name="vesuvio.cisco.com" nodeid="1" votes="1"><fence/>

</clusternode><clusternode name="bernina.cisco.com" nodeid="2" votes="1">

<fence/></clusternode>

</clusternodes><cman expected_votes="1" two_node="1"/><fencedevices/><rm log_level="7">

<failoverdomains><failoverdomain name="local_domain" nofailback="0" ordered="0"

restricted="1"/></failoverdomains><resources/><service autostart="0" name="ctm_service" recovery="relocate">

<ip address="10.58.65.22/24" monitor_link="1"/><fs device="/dev/sdk" force_fsck="1" force_unmount="1" fstype="ext3"

mountpoint="/cisco" name="cisco" options="" self_fence="0"/><fs device="/dev/sdt" force_fsck="1" force_unmount="1" fstype="ext3"

mountpoint="/oracle" name="oracle" options="" self_fence="0"/><fs device="/dev/sde" force_fsck="1" force_unmount="1" fstype="ext3"

mountpoint="/db01" name="ora_db01" options="" self_fence="0"/><fs device="/dev/sdf" force_fsck="1" force_unmount="1" fstype="ext3"

mountpoint="/db02" name="ora_db02" options="" self_fence="0"/><fs device="/dev/sdg" force_fsck="1" force_unmount="1" fstype="ext3"

mountpoint="/db03" name="ora_db03" options="" self_fence="0"/>

TODO_Dev_db011 Enter the disk that you associated with /db01. See the example in Setting the Environment for Installation, page 4-3.

/dev/sde


/dev/sdf


/dev/sdg


/dev/sdh


/dev/sdi

TODO_Dev_ctm_backup1 Enter the disk that you associated with /ctm_backup. See the example in Setting the Environment for Installation, page 4-3.

/dev/sdj

1.Be sure that the disk space meets the requirements listed in Table 2-4 and Table 2-5.

Table 4-3 RHCS Configuration Parameters (continued)



Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Starting the Cluster

<fs device="/dev/sdh" force_fsck="1" force_unmount="1" fstype="ext3" mountpoint="/db04" name="ora_db04" options="" self_fence="0"/>

<fs device="/dev/sdi" force_fsck="1" force_unmount="1" fstype="ext3" mountpoint="/db05" name="ora_db05" options="" self_fence="0"/>

<fs device="/dev/sdj" force_fsck="1" force_unmount="1" fstype="ext3" mountpoint="/ctm_backup" name="ctm_backup" options="" self_fence="0"/>

<script file="/opt/CiscoTransportManagerServer/HA/bin/NtwIFAgent.py" name="NTW_IF"/>

<script file="/opt/CiscoTransportManagerServer/HA/bin/OracleAgent.py" name="CTM_SID">

<script file="/opt/CiscoTransportManagerServer/HA/bin/CtmAgent.py" name="CTM_SRV">

<script file="/opt/CiscoTransportManagerServer/HA/bin/ApacheAgent.py" name="CTM_APACHE"/>

</script></script>

</service></rm>

</cluster>

Step 4 Locate the line that begins with <service autostart. In the recovery parameter, enter relocate.

For example,

<service autostart="0" name="ctm_service" recovery="relocate">

Step 5 Save the file as cluster.conf file.

Step 6 Copy the modified cluster.conf file to the /etc/cluster directory on both the primary and secondary servers.

Starting the ClusterThe following sections describe the process of starting the cluster:




Starting RHCS ServicesTo start RHCS services, follow these steps:

Step 1 Shut down the interface associated to virtual IP on the primary server that is managed by the cluster:

ifconfig vip-interface down

Step 2 Enter the following command to disable dbora services (which are on during installation):

chkconfig --del dbora


service cman start


Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Starting the Cluster


service cman start





ctm_heartbeat automatically starts when you start rgmanager. On the primary server, ctm_heartbeat pings the standby server at startup to find out if the RHCS services (cman and rgmanager) are running. This check is performed three times, with a 10-second wait between attempts. If the three attempts fail, RHCS moves ctm_heartbeat to the failed state.

Starting ctm_services on the Primary ServerAfter you have successfully started RHCS services, you can start Prime Optical from the primary server. All resources used by Prime Optical are defined in the ctm_service in /etc/cluster/cluster.conf file.

To start ctm_services on the primary server:


clustat

If ctm_service is not disabled, invoke it by executing the following command:


Step 2 Start the ctm_service. Enter the following command:


The clusvcadm -e ctm_service command starts all Prime Optical resources. You can check the startup sequence detailed in the /var/log/message directory and the output of /opt/Cisco*Server/HA/log/cisco_agents_cluster.log (or the filename that you specified in the logHAgent.ini file when you configured logging).


clustat

The clustat command output should show that the ctm_service has started.

Verifying Cluster Status on the Standby NodeEnter the following command to verify the status of the standby server:

clustat

The output must show that ctm_service has started on the primary server.


Chapter 4 Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Installing the Multicast Heartbeat Services Package

Installing the Multicast Heartbeat Services PackageMulticast Heartbeat services checks the healthy status between the cluster nodes and sends e-mail notifications when there are connection problems between nodes.

To install the Multicast Heartbeat Services package, follow these steps:





For example:









rpm -ivh --relocate /=/opt MulticastHBService-1.5-2.x86_64.rpm


Preparing...######################################## [100%]1: MulticastHBSrv#################################### [100%]Installation DONE!

Step 5 Copy the MulticastHBService-1.5-2.x86_64.rpm file to the /root directory of the secondary server.


yum install-nogpgcheck MulticastHBService-1.5-2.x86_64.rpm


Preparing...######################################## [100%]1: MulticastHBSrv#################################### [100%]Installation DONE!

After the rpm file is installed, a MulticastHeartBeat directory is created, containing the following files:



Editing the multicastConfig.ini File

Step 1 Connect to the primary server and navigate to the /opt/MulticastHeartBeat directory.

Step 2 Edit the multicastConfig.ini file:

vim multicastConfig.ini

Step 3 Replace the tags listed in Table 4-5:

Step 4 Save the multicastConfig.ini file.

Editing the logMulticast.ini File

Step 1 Connect to the primary server and navigate to the /opt/MulticastHeartBeat directory.

Step 2 Edit the logMulticast.ini file:

vim logMulticast.ini

Step 3 Specify the mail receiver. Under the [handler_mail] header, replace <hostname> and <mail_address@<domain> in the following line:

args=(‘localhost’, ‘root@cpo<hostname>, [‘<mail_address@<domain>’], ‘Multicast interface heartbeat notification’

For example:

args=(‘localhost’, '[email protected]', [‘[email protected]’], ‘Multicast interface heartbeat notification’

Table 4-4 MulticastHeartBeat Directory Files

Files Description

MulticastHeartBeat.ph Python script to probe the communication status of cluster servers.

MulticastHBService Start, stop, and status operations.

multicastConfig.ini Configuration file that sets the network interface and polling period to monitor the connection between nodes.

logMulticast.ini Log activity configuration file for the MulticastHBService service.

clusterModule.PY Module that provides functions for cluster (RHCS) management; for example, starting, stopping, and retrieving status on the cluster.

Table 4-5 MulticastConfig.ini File Parameters

Tags Description

TODO_Ifname Network interface used by the heartbeat service to monitor the connection between nodes; for example, eth1.

TODO_othernode The name associated to the multicast interface of the other node on the cluster; for example, node2-abc.cisco.com.



Step 4 Save the logMulticast.ini file.

Starting the Multicast HeartBeat Services

Step 1 Connect to the primary server and navigate to the /etc/init.d directory.

Step 2 Enter the following command to create a symbolic link:

ln -s /opt/MulticastHeartBeat/MulticastHBService

Step 3 Enter the following command to start the MulticastHBService on the primary server:

service MulticastHBService start

Configuring the MultiCast Heartbeat Services on the Secondary Server

Step 1 Connect to the secondary server.

Step 2 Follow the steps described in Editing the multicastConfig.ini File, page 4-19 on the secondary server.

Note Be sure to enter values that are applicable to the secondary server.

Step 3 Follow the steps described in Editing the logMulticast.ini File, page 4-19 on the secondary server.

Note Be sure to enter values that are applicable to the secondary server.


Cisco Pri

C H A P T E R 5
Upgrading Cisco Prime Optical in an High Availability Environment Using the Data Migrator Wizard
This chapter describes how to use the Cisco Prime Optical 10.5 Data Migrator wizard to migrate select data from CTM R8.5 or later to Prime Optical 10.5. With the wizard, you can migrate a subset of an existing database to a newly installed Prime Optical 10.5 server.

The following are the benefits and limitations of the Data Migrator wizard:

• Reduces the migration time and provides a safer, easier migration.

• Lets you migrate the following data:

– Core tables and database pruning policy information.

– Service states and configurations, including CORBA and SNMP gateway states and configurations.

– Users, user profile information, and CORBA and SNMP gateway users.

– (Optional) Managed NEs, NE logical grouping information, network partition information, security information to authenticate to nodes, Domain Explorer and Subnetwork Explorer map coordinates, basic NE service configuration, and PM service states.

– (Optional) Manual links and Layer 1 circuits with user data.

– (Optional) VLAN customer-related information, Layer 2 topologies, and QoS templates.

– (Optional) Alarm Log Data

• Does not migrate PM data.

As the root user, install and run the tool on the same server where the Prime Optical database is installed.

The tool prepares a newly installed Prime Optical 10.5 server to use the data as used in the earlier server.

Caution You cannot migrate and preserve your entire database. You can only migrate a subset of your database by using the Data Migrator wizard.

Note In the bottom right corner of the wizard screens, the memory footprint progress bar shows the rate of available memory compared to the allocated memory; for example: “Available: x MB of y MB.” The “Request garbage collection” tool lets you free up memory when the amount of data is greater than what is expected, and the running task is taking too long to complete.


Chapter 5 Upgrading Cisco Prime Optical in an High Availability Environment Using the Data Migrator Wizard Overview of the Upgrade Process Using the Data Migrator Wizard

This chapter contains the following sections:

• Overview of the Upgrade Process Using the Data Migrator Wizard, page 5-2

• Installing the Data Migrator Wizard, page 5-2

• Using the Data Migrator Wizard in a Geographic Redundancy, HA Environment on Linux, page 5-4

• Using the Data Migrator Wizard in a Geographic Redundancy, HA Environment on Linux, page 5-4

• Using the Data Migrator Wizard in a Dual-Node Cluster, Local Redundancy, HA Environment on Linux

• Using the Data Migrator Wizard to Export the Database, page 5-5

• Using the Data Migrator Wizard to Import Your Data, page 5-6

• Uninstalling the Data Migrator Wizard, page 5-9

Overview of the Upgrade Process Using the Data Migrator Wizard

Using the Data Migrator wizard to upgrade from CTM 8.5, 9.0, 9.1, 9.2, 9.3, 9.3.1, 9.5, 9.5.1, 9.6, 9.8 or 10.0 to Prime Optical 10.5 involves the following process:

1. Perform a full backup or export of your earlier database.

2. Install the Data Migrator wizard on any server in the same network as the Prime Optical server. (See Installing the Data Migrator Wizard, page 5-2.)

3. Export the database from the previous CTM or Prime Optical release server that you want to upgrade. (See Using the Data Migrator Wizard to Export the Database, page 5-5.)

4. Install Prime Optical HA 10.5 from scratch. (See Chapter 3, “Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Configuration on a Linux Server.”)

Caution Before you install the operating system from scratch, save the .zip export archive on another server.

5. Import the database to the new Prime Optical 10.5 database. (See Using the Data Migrator Wizard to Import Your Data, page 5-6.)

Installing the Data Migrator WizardAs the root user, do the following:

Step 1 Insert one of the following USB drive based on your configuration, in your workstation:

Cisco Prime Optical Server installation USB drive—Linux




Chapter 5 Upgrading Cisco Prime Optical in an High Availability Environment Using the Data Migrator Wizard Installing the Data Migrator Wizard


For example:



For Linux, enter:




/dev/lofi/N

To mount the iso image, enter the command:

mount -F hsfs </dev/lofi/N> /mnt

Note You can also replace the /mnt directory with a different mount point. Sometimes Cisco will create a software update (patch) for a Prime Optical release. If you install a patch at a later date, be sure to install the latest Data Migrator tool. To install the latest tool, go to the Download Software page (http://www.cisco.com/cisco/software/navigator.html?mdfid=283812777&i=rm) and navigate to the latest Prime Optical 10.5 release. Select and download the appropriate Data Migrator configuration (LINUX), extract the DataMigrator.tar file, then proceed to Step 5.

Step 3 Copy the /mnt/DataMigrator/DataMigrator.tar file to a local folder.

Step 4 Enter the following commands to extract the .tar archive contents:

cd local-folder/DataMigratortar xvf DataMigrator.tar

Step 5 Enter the following commands to run the installer:

cd Disk1./install.sh

Step 6 In the Introduction screen, click Next.

Step 7 At the Choose Installation Folder screen, specify where you want to install the Data Migrator wizard. The default folder is $HOME/DataMigrator. You can click Choose to choose a different destination. After you specify your destination, click Next.

Step 8 The Preinstallation Summary screen shows the items that will be installed and the disk space required. Click Install.

Step 9 The Installing Data Migrator screen tracks the progress of the installation.

Step 10 The Installation Complete screen summarizes the results of the installation. Click Done.

After installation, the directory structure contains a folder with the DataMigrator executable script.


http://www.cisco.com/cisco/software/navigator.html?mdfid=283812777&i=rm

Chapter 5 Upgrading Cisco Prime Optical in an High Availability Environment Using the Data Migrator Wizard Using the Data Migrator Wizard in a Geographic Redundancy, HA Environment on Linux

Using the Data Migrator Wizard in a Geographic Redundancy, HA Environment on Linux

Step 1 Stop the server that is running on the active node.


opticalctl stop

Step 3 Use the Data Migrator wizard to export the database. See Using the Data Migrator Wizard to Export the Database.

Step 4 Save the.zip file generated by the Data Migrator wizard, and any other important files, to a different server.

Step 5 Reinstall Prime Optical HA 10.5 as a completely new installation. Follow the steps in Chapter 4, “Installing the Cisco Prime Optical High Availability Solution in a Local Redundancy Configuration on a Linux Server.”

Step 6 Freeze the Clusvcadm -Z ctm_service.

Step 7 Enter the following command to stop Prime Optical:

opticalctl stop

Step 8 Enter the following commands to perform a full export of the database. (At this point, the database is empty, but retaining an export can save time if the import fails during the next step.)

cd /cisco/CiscoTransportManagerServer/bin./export_full_db.sh

Step 9 Use the Data Migrator wizard to import the export.zip file to the newly installed server. See Using the Data Migrator Wizard to Import Your Data.

Step 10 Enter the following command to start Prime Optical:

opticalctl start

Step 11 Connect to a Prime Optical client and verify that Prime Optical is running correctly.

Step 12 Unfreeze the primary cluster by entering the following command:


Using the Data Migrator Wizard in a Dual-Node Cluster, Local Redundancy, HA Environment on Linux

Step 1 Freeze the primary cluster by executing the following command:


Step 2 Stop the server on the primary node by executing the following command:

opticalctl stop


Chapter 5 Upgrading Cisco Prime Optical in an High Availability Environment Using the Data Migrator Wizard Using the Data Migrator Wizard to Export the Database

Step 3 Use the Data Migrator wizard to export the database. See Using the Data Migrator Wizard to Export the Database.

Step 4 Save the .zip file generated by the Data Migrator wizard and any other important files to a different server.

Step 5 Reinstall Prime Optical HA 10.5 as a completely new installation. Follow the steps in Chapter 3, “Installing the Cisco Prime Optical High Availability Solution in a Single-Node Geographical Redundancy Configuration on a Linux Server.”

Step 6 Freeze the Clusvcadm -Z ctm_service.

Step 7 Stop the Prime Optical server by executing the following command:

opticalctl stop

Step 8 Enter the following commands to perform a full export of the database. (At this point, the database is empty, but retaining an export can save time if the import fails during the next step.)

cd /cisco/CiscoTransportManagerServer/bin./export_full_db.sh

Step 9 Use the Data Migrator wizard to import the export.zip file to the newly installed server. See Using the Data Migrator Wizard to Import Your Data.

Step 10 Enter the following command to start the Prime Optical server:

opticalctl start

Step 11 Connect to a Prime Optical client and verify that Prime Optical is running correctly.

Step 12 Unfreeze the primary cluster by executing the following command:

opticalctl stop

Step 13 Freeze the cluster executing the following command:


Using the Data Migrator Wizard to Export the Database

Step 1 Enter the following command to verify whether the Prime Optical server is running:

opticalctl status

If the Prime Optical server is running, enter the following command to stop it:

opticalctl stop

Step 2 Navigate to the installation folder (for example, /tmp/DataMigrator/bin) and run the DataMigrator executable.

Step 3 The Data Migrator wizard opens. Do the following:

a. (Optional) To view the supported versions that are available for export, click Show Migration Matrix. The Supported Migration Paths dialog box opens, listing the valid migration versions. After reviewing the supported migration paths, click Close.

b. From the Migration Mode drop-down list, choose Export.

c. Click OK.


Chapter 5 Upgrading Cisco Prime Optical in an High Availability Environment Using the Data Migrator Wizard Using the Data Migrator Wizard to Import Your Data

Note For Embedded database, the Oracle SID is CTM for 8.5 through 9.6.3 and optdb for 10.5.

Step 4 In the Database Connection screen, specify details of the database from which you want to export data.

Step 5 Click Connect. The progress bar at the bottom of the screen tracks the progress of the operation. (To cancel the operation, click the X icon to the right of the progress bar.)

After the connection succeeds, the Connection State area shows the state as “Connected.” The version, size, and installed modules are retrieved from the database and displayed.

Step 6 Click Next.

Step 7 If you receive the following prompt, shut down the server and reconnect to the database and then click OK:

The server is still running on the selected server. The connection to the database will be closed. Before continuing, you must shut down the server and then reconnect to the database.

Caution It is important that you shut down the server and reconnect to the database before proceeding to the next step.

Step 8 In the Categories screen, review the data categories that will be exported. Click Next.

Step 9 In the Open dialog box, select the directory and the .zip archive where you want to store the exported data and then click Open.

Step 10 The File Selection screen displays the selected archive file and the estimated time to complete the export process. Click Next.

Step 11 In the Export screen, click Start. All data categories are exported and compressed.

A progress bar tracks the progress of the export. You can quit exporting at any time by clicking the X button next to the progress bar.

The table summarizes the results of the export. For each data category, the table lists the exported database tables, the number of rows exported, and the length of time the export took.

Step 12 Click Exit.

Step 13 At the prompt “The operation is complete,” click OK.

Using the Data Migrator Wizard to Import Your Data

Step 1 Enter the following command to verify whether the server is running:

opticalctl status

Step 2 If the server is running, enter the following command to stop it:

opticalctl stop

Step 3 Navigate to the installation folder (for example, /tmp/DataMigrator/bin) and run the DataMigrator executable.



Step 4 The Data Migrator wizard opens. Do the following:

a. (Optional) To view the supported versions that are available for import, click Show Migration Matrix. The Supported Migration Paths dialog box opens, listing the valid migration versions. After reviewing the supported migration paths, click Close.

b. From the Migration Mode drop-down list, choose Import.

c. Click OK.

Step 5 In the Open dialog box, select the .zip archive file to import and then click Open.

Step 6 In the Archive Summary screen, review the details of the archive file to import. To select a different archive file to import, click the browse (...) button. The table summary shows the database tables that will be imported, including the data category and the number of rows in each table. Click Next.

Step 7 In the Categories screen, do the following:

a. Select the categories to import. General, Services, and Users are mandatory categories; the NEs, L1, and L2 topology, and Alarm Log categories are optional.

b. (Optional) In the General category, click the Setup link to open the General Import Options dialog box. If desired, check the Import pruning options check box and then click Apply. The dialog box closes.

c. (Optional) In the NEs category, click the Setup link to open the NE Import Options dialog box and specify additional setup options for the import. The dialog box has the following tabs:

– NE Services—Lets you tune the NE services of the target server after the import.

– NEs—Lets you change the NE state and disable automatic subnetwork grouping. (Automatic subnetwork grouping slows down the initial discovery of the new server.)

– Miscellaneous—Lets you customize the import of user maps and equipment inventory notes.

After making your selections in the NE Import Options dialog box, click Apply. The dialog box closes.

d. Click Next.

Step 8 In the Database Connection screen, specify details of the database to which you want to import data. Do the following:

a. If the release to which you want to import data has an embedded database, check the Server uses an embedded database check box.

When the Server uses an embedded database check box is checked:

– You cannot edit any fields in the Connection Properties area.

– The hostname is set to the current machine’s name.

– The instance name is set to CTM, which is the default for an embedded database and can be changed.

– The port number is set to 1521, which is the default for an embedded database and cannot be changed.

– The Username and Password fields are cleared. After you click the Connect button, the Username and Password fields are filled in if the Data Migrator wizard can find those properties in the CTMServer.cfg configuration file.

b. If the release to which you want to import data does not use an embedded database, verify that the Server uses an embedded database check box is unchecked.

When the Server uses an embedded database check box is unchecked:

– You can edit all fields in the Connection Properties area.



– The Username and Password fields are restored to their default values.

– You must enter a hostname in the Hostname field.

– You must enter a port number in the Port Number field. The default port number is 1521.

– You can either enter values in the remaining fields, or accept the defaults.

c. Click Connect.

The progress bar at the bottom of the screen tracks the progress of the operation. (To cancel the operation, click the X icon to the right of the progress bar.)

After the connection succeeds, the Connection State area shows the state as “Connected.” The version, size, and installed modules are retrieved from the database and displayed.

d. Click Next.

Step 9 If you receive the following prompt, shut down the server and reconnect to the database and then click OK:

The server is still running on the selected server. The connection to the database will be closed. Before continuing, you must shut down the server and then reconnect to the database.

Caution It is important that you shut down the server and reconnect to the database before proceeding to the next step.

Step 10 In the Validation screen, do the following:

a. Click the Start button to validate the size, installed modules, and data on the selected destination server. The Result field shows whether the validation succeeded (Allowed) or failed (Impossible).

The following server sizes produce the following results:

Table 5-1 Server Size Results

Original Server Size Target Server Size Result

Small Small Allowed

Small Medium Allowed

Small Large Allowed

Small High end Allowed

Medium Small Impossible

Medium Medium Allowed

Medium Large Allowed

Medium High end Allowed

Large Small Impossible

Large Medium Impossible

Large Large Allowed

Large High end Allowed

High end Small Impossible

High end Medium Impossible

High end Large Impossible


Chapter 5 Upgrading Cisco Prime Optical in an High Availability Environment Using the Data Migrator Wizard Uninstalling the Data Migrator Wizard

b. In the Data Validation area > Details column, click the Details link to see why a particular table could not be imported. At the following prompt, click OK:

Table-name must be empty; it contains x row(s) instead.

Note To avoid this error, click Back, return to the Validation screen, and select a different server that has a new Prime Optical 10.5 installation.

c. Click Next.

Step 11 In the Import screen, click the Start button to import the data to the target database.

Step 12 The confirmation popup reminds you that if you have not done so already, you should run the export_full_db.sh script to make a backup copy of your current data before proceeding. (Without a backup copy of your data, you would need to completely reinstall the Prime Optical server if problems were to occur.) Click Yes at the following prompt:

This action will write the selected data to the target database.Make a copy of the current data using the export_full_db.sh tool.

Select Yes to proceed or No to cancel the operation.

Step 13 A progress bar tracks the progress of the import. (To cancel the operation, click the X icon to the right of the progress bar.)

The table summarizes the results of the import. For each data category, the table lists the imported database tables, the result of the housekeeping phase (purging old data to make room for the new data), the result of the data migration phase, and the length of time the import took.

Caution If a database table was not imported, click the Details link to show the reason for the failure. If errors occurred during the import, use the import_full_db.sh script to reimport the Prime Optical database. See the Cisco Prime Optical 10.5 User Guide section “Restoring the Prime Optical Database from the Previous Backup.”

Step 14 Click Exit.

Uninstalling the Data Migrator WizardTo uninstall the Data Migrator wizard, delete the folder where you installed the wizard. The default folder is $HOME/DataMigrator.

High end High end Allowed

Small Small Allowed

Small Medium Allowed

Table 5-1 Server Size Results (continued)

Original Server Size Target Server Size Result


http://www.cisco.com/c/en/us/support/cloud-systems-management/prime-optical/products-user-guide-list.html

Chapter 5 Upgrading Cisco Prime Optical in an High Availability Environment Using the Data Migrator Wizard Uninstalling the Data Migrator Wizard


Cisco Prime Op

A

P P E N D I X A
Installing and Patching Oracle
This section provides supporting information to assist you with the Oracle 12c installation.

Note • Oracle 12c is available only for a 64-bit architecture.

• This section is applicable only to Oracle Not Embedded database setups.

• Use the information in this section in conjunction with the Oracle documentation available on the Oracle website.

Oracle PrerequisitesThis section describes how to create or update an existing Oracle user before you install Oracle 12c. Before you install Oracle 12c, verify that:

• Oracle is not running

• The oracle UNIX user exists and is configured correctly

• The following USB drive or ISO image:

– Cisco Prime Optical Server installation USB drive or ISO image—Linux

Creating a New Oracle UserFollow these steps to create a new Oracle user:

Step 1 Enter the following command to create a UNIX group for database administrators:

groupadd -g 3303 dba

Step 2 Enter the following command to create a UNIX group for installation of the Oracle software:

groupadd oinstall

Step 3 Enter the following command to create a UNIX account to own the Oracle software:

useradd -g oinstall -G dba -m -s /bin/csh -d /cporacle cporacle

A-1tical 10.5 High Availability Installation Guide

Appendix A Installing and Patching Oracle Oracle Prerequisites

Caution Do not enter /oracle for the home directory.

Step 4 Enter the following command to change the Oracle user password:

passwd cporacle

Step 5 Enter the new password and then re-enter it to confirm it.

Updating an Existing Oracle UserFollow these steps if an Oracle user already exists:

Step 1 Log in as root user, and then enter the following commands to create a new home directory:

mkdir /cporaclechown oracle_user:oinstall /cporaclechmod 700 /cporacle

Step 2 As root user, enter the following command to change the home directory if it is /oracle:

usermod -d /cporacle cporacle

Setting Shell Limits for an Oracle UserFollow these steps to set the shell limits for an Oracle user:

Edit the file /etc/security/limits.conf file by adding the following parameters:

oracle hard memlock unlimitedoracle hard nofile 65536oracle hard nproc 16384oracle soft nofile 65536oracle soft nproc 16384oracle soft memlock unlimited

Add the following parameters only if Prime Optical is installed with optusr at runtime:

optusr soft nofile 65536optusr hard nofile 65536optusr soft nproc 16384optusr hard nproc 16384

Note Do not add optusr settings if Prime Optical user at runtime is root.

A-2Cisco Prime Optical 10.5 High Availability Installation Guide

Appendix A Installing and Patching Oracle Installing the Oracle 12c Software with the Response File (*.rsp) Provided by Cisco

Installing the Oracle 12c Software with the Response File (*.rsp) Provided by Cisco

Step 1 Insert the following USB drive, based on your configuration, in your workstation:

• Cisco Prime Optical Server installation USB drive—Linux




For example:



For Linux, enter:




/dev/lofi/N

To mount the ISO image, enter the command:

mount -F hsfs </dev/lofi/N> /mnt


Step 3 Enter the following command to copy the default profile to the Oracle home directory:

cp /mnt/Disk1/oracle.cshrc /oracle-home-directory/.cshrc

where /oracle-home-directory is /cporacle by default.

Step 4 Enter the following command for 64-bit Oracle Enterprise Edition:

cp /mnt/Disk1/oracle12_enterp_solaris.rsp /oracle

Step 5 Enter the following command to safely remove the USB drive from the workstation:

umount /mnt

Step 6 Enter the following commands to change ownership of the Oracle software directories:

chown -R oracle-user:dba /oraclechown -R oracle-user:dba /oracle-home-directorychown -R oracle-user:dba /db01chown -R oracle-user:dba /db02chown -R oracle-user:dba /db03chown -R oracle-user:dba /db04chown -R oracle-user:dba /db05chown -R oracle-user:dba /ctm_backup


Appendix A Installing and Patching Oracle Installing the Oracle 12c Software with the Response File (*.rsp) Provided by Cisco

Step 7 Enter the following command to add read-write permissions:

/usr/bin/chmod +rw /oracle-home-directory/.cshrc

Step 8 Follow the preinstallation steps in the Oracle 12c Installation Guide for Solaris, available on the Oracle website. Note the following constraints:

• No RAC configuration

• No ASM configuration

• No Oracle application server

Step 9 Enter the following command to log in as the oracle user:

su - cporacle

Step 10 Enter the following command to set the display on your terminal:

setenv DISPLAY hostname-or-IP-address:0.0


echo $DISPLAY

In the command output, you should see the following:


Step 12 If you are using an xterm window or a remote host, as a root user enter the following command to enable the xterm connection from clients:

xhost +

Step 13 Depending on the distribution media of the Oracle software, change to the directory where the runInstaller application is located. For example, if you are using a DVD, enter:

cd /dvd/dvd0

Step 14 Enter the following command to start the Oracle installer:

./runInstaller -silent -responseFile /oracle/oracle12_0_1_solaris.rsp &

When this step is complete, you will be prompted to run several shell scripts as root. Instead, press the Enter key to return to the prompt.

Note • If you do not have a DVD, see your Oracle documentation for instructions on how to download and extract the correct software for the installation.

• When carrying out the steps in this section, see the Oracle Content Database Installation Guide on the Oracle website.

Step 15 If you were prompted to run the /oracle/oraInventory/orainstRoot.sh script with root privileges in Step 14, log in to another terminal window as the root user and enter the following command:

/oracle/oraInventory/orainstRoot.sh

Step 16 If you were prompted to run the /oracle/product/12.1.0/root.sh script in Step 14, log in to another terminal window as the root user and enter the following command to run the root.sh script:

./root.sh


Appendix A Installing and Patching Oracle Downloading and Installing the Required Oracle 12c Patch for 64-Bit Linux Platforms

Step 17 Because the client static library is not generated during installation, you must do the following to generate and link your applications to the client static library:

a. Enter the following command to log in to the Oracle workstation as the oracle user:

su - cporacle

b. Enter the following command to generate the client static library:

$ORACLE_HOME/bin/genclntst

Downloading and Installing the Required Oracle 12c Patch for 64-Bit Linux Platforms

See the Oracle installation documentation to install the patches listed in Table A-1. You must install the Oracle patches in the exact order shown.

Table A-1 Oracle 12c Patches for Solaris and Linux

Installation Order Patch Number Package Name DescriptionLinux1 6880880 p6880880_121010_Linux-x86-64.zip Installs Oracle 12.1.0.1.8

2 20831110 p20831110_121020_Linux-x86-64.zip Install the Oracle Security Patch (PSU:12.1.0.2.4)

3 21068507 p21068507_121020_Linux-x86-64.zip Install the Oracle Java VM Patch (PSU:12.1.0.2.4)


Appendix A Installing and Patching Oracle Downloading and Installing the Required Oracle 12c Patch for 64-Bit Linux Platforms


cisco prime optical high availability installation …...installation guide for use with cisco prime...

Documents