Top
Systemwalker Operation Manager  Cluster Setup Guide for UNIX
FUJITSU Software
Chapter 6 Cluster Setup Procedure Specific to Systems Using Oracle Solaris Cluster > 6.1 Registering Systemwalker Operation Manager with the Oracle Solaris Cluster System > 6.1.1 Creating applications
PreviousNext

6.1.1 Creating applications

Create the following applications to control the starting and stopping of each Systemwalker Operation Manager daemon:

Create these application files on both the active and standby nodes, and place them at the same location in each node (use the same directory path). Do not create these files on the shared disk. Be sure to set up execution privileges after creating these applications.

Create fault monitor to monitor the daemons running on the active node of the cluster system, and use daemon termination as a trigger for cluster failover. There is no need to create fault monitor if it is not required.

The following sections describe examples of how to create each of these applications. Samples of each of these programs are provided with this product. Copy and customize these samples according to your operating environment.

6.1.1.1 Creating resource type registration files

Create resource type registration file (RTR file) to define the static properties of data service. Resource type registration file must be created. The following sample resource type registration file is provided:

/opt/FJSVJMCMN/etc/script/SC3.0/resource

The sample resource type registration file is for 1:1 active/standby configuration. This file must be modified for N:1 active/standby configuration and dual node mutual standby configuration.

Copy the sample file, and then modify it to match the cluster system operation. For 1:1 active/standby configuration, this sample file can be used without modification, but should be backed up anyway.

The procedure for modifying the resource type registration file is described below.

Example of how to modify the resource type registration file for N:1 active/standby configuration

  1. For N:1 active/standby configuration, create a separate resource type registration file for each active node (N files altogether) and assign a unique file name in each case. Modify the information in each file to match the respective environment.

    In the following example, there are three active nodes.

    • Resource type registration file 1: resource1

      RESOURCE_TYPE = "OMGR1";

(omitted)

RT_BASEDIR=/opt/FJSVJMCMN/etc/script/SC3.0;
PKGLIST = FJSVJMCMN;

START         = omgr_start1;
STOP          = omgr_stop1;

MONITOR_START = omgr_fstart1;
MONITOR_STOP  = omgr_fstop1;
MONITOR_CHECK = omgr_check1;

    • Resource type registration file 2: resource2

      RESOURCE_TYPE = "OMGR2";

(omitted)

RT_BASEDIR=/opt/FJSVJMCMN/etc/script/SC3.0;
PKGLIST = FJSVJMCMN;

START         = omgr_start2;
STOP          = omgr_stop2;

MONITOR_START = omgr_fstart2;
MONITOR_STOP  = omgr_fstop2;
MONITOR_CHECK = omgr_check2;

    • Resource type registration file 3: resource3

      RESOURCE_TYPE = "OMGR3";

(omitted)

RT_BASEDIR=/opt/FJSVJMCMN/etc/script/SC3.0;
PKGLIST = FJSVJMCMN;

START         = omgr_start3;
STOP          = omgr_stop3;

MONITOR_START = omgr_fstart3;
MONITOR_STOP  = omgr_fstop3;
MONITOR_CHECK = omgr_check3;

    If fault monitor is not required, comment out the "MONITOR_START", "MONITOR_STOP" and "MONITOR_CHECK" lines.

    If necessary, modify "RT_BASEDIR" to specify the location where the application files are actually stored.

  2. Place each resource type registration file on its respective active node, and then copy all the resource type registration files to the standby node, storing all N files in the same directory structure as the active nodes.

    The following example shows how to place these files when there are three active nodes:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/resource1
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/resource2
    Active node3: /opt/FJSVJMCMN/etc/script/SC3.0/resource3
    Standby node: /opt/FJSVJMCMN/etc/script/SC3.0/resource1
              /opt/FJSVJMCMN/etc/script/SC3.0/resource2
              /opt/FJSVJMCMN/etc/script/SC3.0/resource3

Example of how to modify the resource type registration file for dual node mutual standby configuration

  1. For dual node mutual standby configuration, create a separate resource type registration file for each active node (two files altogether) and assign a unique file name in each case. Modify the information in each file to match the respective environment.

    The following example shows how to change these files:

    • Resource type registration file 1: resource1

      RESOURCE_TYPE = "OMGR1";

(omitted)

RT_BASEDIR=/opt/FJSVJMCMN/etc/script/SC3.0;
PKGLIST = FJSVJMCMN;

START         = omgr_start1;
STOP          = omgr_stop1;

MONITOR_START = omgr_fstart1;
MONITOR_STOP  = omgr_fstop1;
MONITOR_CHECK = omgr_check1;

    • Resource type registration file 2: resource2

      RESOURCE_TYPE = "OMGR2";

(omitted)

RT_BASEDIR=/opt/FJSVJMCMN/etc/script/SC3.0;
PKGLIST = FJSVJMCMN;

START         = omgr_start2;
STOP          = omgr_stop2;

MONITOR_START = omgr_fstart2;
MONITOR_STOP  = omgr_fstop2;
MONITOR_CHECK = omgr_check2;

    If fault monitor is not required, comment out the "MONITOR_START", "MONITOR_STOP" and "MONITOR_CHECK" lines.

    If necessary, modify "RT_BASEDIR" to specify the location where the application files are actually stored.

  2. Copy the two resource type registration files to the same directory on both the active node and the standby node.

    The following example shows how to place these files:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/resource1
  (Standby node2)/opt/FJSVJMCMN/etc/script/SC3.0/resource2
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/resource2
  (Standby node1)/opt/FJSVJMCMN/etc/script/SC3.0/resource1

6.1.1.2 Creating START callback methods

Create a START callback method to start the daemons for Jobscheduler and Job Execution Control. The following sample START callback method is provided:

/opt/FJSVJMCMN/etc/script/SC3.0/omgr_start

This sample START callback method is for 1:1 active/standby configuration. This START callback method must be modified for N:1 active/standby configuration and dual node mutual standby configuration.

Note also that in the sample START callback method, the mount point for the shared disk is "/disk1". Modify the mount point so that the shared disk is referred to by the correct name.

Copy the sample file, and then modify it to match the cluster system operation. For 1:1 active/standby configuration, this sample file can be used without modification, but should be backed up anyway.

The following examples show how to modify the START callback method.

Example of how to modify the START callback method for N:1 active/standby configuration

  1. Remove the comment for the "Make symbolic links. (if N:1 standby)" section. (That is, remove the "#" from the lines of code.)

    [Before]

    # Make symbolic links.(if N:1 standby)
# ACL Manager
#if [ ! "(" -h "/var/opt/FJSVfwseo/JM" -o -f "/var/opt/FJSVfwseo/JM" ")" ]
#then
#      ln -s /disk1/FJSVfwseo/JM /var/opt/FJSVfwseo/JM
#fi
# Job Scheduler
#if [ ! "(" -h "/var/opt/FJSVJOBSC" -o -f "/var/opt/FJSVJOBSC" ")" ]
#then
#      ln -s /disk1/FJSVJOBSC /var/opt/FJSVJOBSC
#fi
# Job Execution Control
#if [ ! "(" -h "/var/spool/mjes" -o -f "/var/spool/mjes" ")" ]
#then
#      ln -s /disk1/FJSVMJS/var/spool/mjes /var/spool/mjes
#fi
#if [ ! "(" -h "/etc/mjes" -o -f "/etc/mjes" ")" ]
#then
#      ln -s /disk1/FJSVMJS/etc/mjes /etc/mjes
#fi
# Calendar
#if [ ! "(" -h "/var/opt/FJSVjmcal/post" -o -f "/var/opt/FJSVjmcal/post" ")" ]
#then
#      ln -s /disk1/FJSVjmcal/post /var/opt/FJSVjmcal/post
#fi
# Stem
#if [ ! "(" -h "/var/opt/FJSVstem" -o -f "/var/opt/FJSVstem" ")" ]
#then
#      ln -s /disk1/FJSVstem /var/opt/FJSVstem
#fi
# - 1:1 standby, N:1 standby

    [After]

    # Make symbolic links.(if N:1 standby)
# ACL Manager
if [ ! "(" -h "/var/opt/FJSVfwseo/JM" -o -f "/var/opt/FJSVfwseo/JM" ")" ]
then
      ln -s /disk1/FJSVfwseo/JM /var/opt/FJSVfwseo/JM
fi
# Job Scheduler
if [ ! "(" -h "/var/opt/FJSVJOBSC" -o -f "/var/opt/FJSVJOBSC" ")" ]
then
      ln -s /disk1/FJSVJOBSC /var/opt/FJSVJOBSC
fi
# Job Execution Control
if [ ! "(" -h "/var/spool/mjes" -o -f "/var/spool/mjes" ")" ]
then
      ln -s /disk1/FJSVMJS/var/spool/mjes /var/spool/mjes
fi
if [ ! "(" -h "/etc/mjes" -o -f "/etc/mjes" ")" ]
then
      ln -s /disk1/FJSVMJS/etc/mjes /etc/mjes
fi
# Calendar
if [ ! "(" -h "/var/opt/FJSVjmcal/post" -o -f "/var/opt/FJSVjmcal/post" ")" ]
then
      ln -s /disk1/FJSVjmcal/post /var/opt/FJSVjmcal/post
fi
# Stem  (*1)
if [ ! "(" -h "/var/opt/FJSVstem" -o -f "/var/opt/FJSVstem" ")" ]
then
      ln -s /disk1/FJSVstem /var/opt/FJSVstem
fi
# - 1:1 standby, N:1 standby

    *1: Remove this comment only if the Master Schedule Management function is enabled.

  2. Prepare N copies of the START callback method (one for each active node), assign a unique name to each copy. Change the directory ("/disk1" in the example) where symbolic links will be created to match the shared disk of each active node.

    Use the same file names as listed in the resource type registration files.

  3. Place each START callback method on its respective active node, and then copy all the START callback methods to the standby node, storing all N methods in the same directory structure as the active nodes.

    The following example shows how to place these files when there are three active nodes:

    Active node1: opt/FJSVJMCMN/etc/script/SC3.0/omgr_start1
    Active node2: opt/FJSVJMCMN/etc/script/SC3.0/omgr_start2
    Active node3: opt/FJSVJMCMN/etc/script/SC3.0/omgr_start3
    Standby node: opt/FJSVJMCMN/etc/script/SC3.0/omgr_start1
             /opt/FJSVJMCMN/etc/script/SC3.0/omgr_start2
             /opt/FJSVJMCMN/etc/script/SC3.0/omgr_start3

Example of how to modify the START callback method for dual node mutual standby configuration

  1. Create a START callback method by copying the sample and changing the file name, and then change the value of the "SUBSYSTEM" variable to "1".

    [Before]

    SUBSYSTEM="0"

    [After]

    SUBSYSTEM="1"

  2. Modify the "- 1:1 standby, N:1 standby" and "- 2 nodes mutual standby" sections for dual node mutual standby. (Change the positions of "#".)

    [Before]

    # - 1:1 standby, N:1 standby
/opt/FJSVfwseo/bin/mpaclcls
sh /opt/FJSVfwseo/bin/jmacltrn.sh
# - 2 nodes mutual standby
#/opt/FJSVfwseo/bin/mpaclcls -s $SUBSYSTEM
#sh /opt/FJSVfwseo/bin/jmacltrn.sh $SUBSYSTEM
 
# - 1:1 standby, N:1 standby, 2 nodes mutual standby

    [After]

    # - 1:1 standby, N:1 standby
#/opt/FJSVfwseo/bin/mpaclcls
#sh /opt/FJSVfwseo/bin/jmacltrn.sh
# - 2 nodes mutual standby
/opt/FJSVfwseo/bin/mpaclcls -s $SUBSYSTEM
sh /opt/FJSVfwseo/bin/jmacltrn.sh $SUBSYSTEM
 
# - 1:1 standby, N:1 standby, 2 nodes mutual standby

  3. Create one more START callback method (change the file name), and change the value of the "SUBSYSTEM" variable to "2".

    [Before]

    SUBSYSTEM="1"

    [After]

    SUBSYSTEM="2"

  4. Place the two START callback methods on the active and standby nodes, using the same directory path in both cases.

    Use the same file names as listed in the resource type registration files.

    The following example shows how to place these files:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_start1
  (Standby node2)/opt/FJSVJMCMN/etc/script/SC3.0/omgr_start2
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_start2
  (Standby node1)/opt/FJSVJMCMN/etc/script/SC3.0/omgr_start1

6.1.1.3 Creating STOP callback methods

Create a STOP callback method to stop the daemons for Jobscheduler and Job Execution Control. The following sample STOP callback method is provided:

/opt/FJSVJMCMN/etc/script/SC3.0/omgr_stop

This sample STOP callback method is for 1:1 active/standby configuration. This STOP callback method must be modified for N:1 active/standby configuration and dual node mutual standby configuration.

Note also that in the sample STOP callback method, the mount point for the shared disk is "/disk1". Modify the STOP callback method so that the shared disk is referred to by the correct name.

Copy the sample file, and then modify it to match the cluster system operation. For 1:1 active/standby configuration, this sample file can be used without modification, but should be backed up anyway.

The following examples show how to modify the STOP callback method.

Example of how to modify the STOP callback method for N:1 active/standby configuration

  1. Remove the comment for the "remove symbolic links. (if N:1 standby)" section. (That is, remove the "#" from the lines of code.)

    [Before]

    # remove symbolic links.(if N:1 standby)
# Job Scheduler
#if [ -h "/var/opt/FJSVJOBSC" ]
#then
#      rm /var/opt/FJSVJOBSC
#fi
# Job Execution Control
#if [ -h "/var/spool/mjes" ]
#then
#      rm /var/spool/mjes
#fi
#if [ -h "/etc/mjes" ]
#then
#      rm /etc/mjes
#fi
# ACL Manager
#/opt/FJSVfwseo/bin/mpaclcls -u
#if [ -h "/var/opt/FJSVfwseo/JM" ]
#then
#      rm /var/opt/FJSVfwseo/JM
#fi
# Calendar
#if [ -h "/var/opt/FJSVjmcal/post" ]
#then
#      rm /var/opt/FJSVjmcal/post
#fi
# Stem
#if [ -h "/var/opt/FJSVstem" ]
#then
#      rm /var/opt/FJSVstem
#fi
 
exit 0

    [After]

    # remove symbolic links.(if N:1 standby)
# Job Scheduler
if [ -h "/var/opt/FJSVJOBSC" ]
then
      rm /var/opt/FJSVJOBSC
fi
# Job Execution Control
if [ -h "/var/spool/mjes" ]
then
      rm /var/spool/mjes
fi
if [ -h "/etc/mjes" ]
then
      rm /etc/mjes
fi
# ACL Manager
/opt/FJSVfwseo/bin/mpaclcls -u
if [ -h "/var/opt/FJSVfwseo/JM" ]
then
      rm /var/opt/FJSVfwseo/JM
fi
# Calendar
if [ -h "/var/opt/FJSVjmcal/post" ]
then
      rm /var/opt/FJSVjmcal/post
fi
# Stem  (*1)
if [ -h "/var/opt/FJSVstem" ]
then
      rm /var/opt/FJSVstem
fi
 
exit 0

    *1: Remove this comment only if the Master Schedule Management function is enabled.

  2. Prepare N copies of the STOP callback method (one for each active node), assigning a unique name to each copy.

    Use the same file names as listed in the resource type registration files.

  3. Place each STOP callback method on its respective active node, and then copy all of the STOP callback methods to the standby node, storing all N files in the same directory structure as the active nodes.

    The following example shows how to place these files when there are three active nodes:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_stop1
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_stop2
    Active node3: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_stop3
    Standby node: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_stop1
              /opt/FJSVJMCMN/etc/script/SC3.0/omgr_stop2
              /opt/FJSVJMCMN/etc/script/SC3.0/omgr_stop3

Example of how to modify the STOP callback method for dual node mutual standby configuration

  1. Create a STOP callback method by copying the sample and changing the file name, and then change the value of the "SUBSYSTEM" variable to "1".

    [Before]

    SUBSYSTEM="0"

    [After]

    SUBSYSTEM="1"

  2. Create one more STOP callback method (change the file name), and change the value of the "SUBSYSTEM" variable to "2".

    [Before]

    SUBSYSTEM="1"

    [After]

    SUBSYSTEM="2"

  3. Place the two STOP callback methods on the active and standby nodes, using the same directory path in both cases.

    Use the same file names as listed in the resource type registration files.

    The following example shows how to place these files:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_stop1
  (Standby node2)/opt/FJSVJMCMN/etc/script/SC3.0/omgr_stop2
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_stop2
  (Standby node1)/opt/FJSVJMCMN/etc/script/SC3.0/omgr_stop1

6.1.1.4 Creating fault monitors

Creating and registering fault monitor allows users to notify cluster system of failover triggers by monitoring network resources and application-specific processes running on the cluster system.

Create the following programs:

Fault monitor has to be created only if application-specific failover triggers are to be set up.

Resource monitoring program

Create a resource monitoring program.

The following sample program is provided:

/opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor

The sample program monitors the execution status of the Jobscheduler and Job Execution Control daemon processes, and triggers a failover if both daemon processes have stopped. For operation monitoring only one of these daemon processes, modify the sample program as appropriate.

This sample program is for 1:1 active/standby configuration. This resource monitoring program must be modified for N:1 active/standby configuration and dual node mutual standby configuration.

Copy the sample file, and then modify it to match the cluster system operation. For 1:1 active/standby configuration, this sample file can be used without modification, but should be backed up anyway.

The following examples show how to modify the resource monitoring program.

Example of how to modify the resource monitoring program for N:1 active/standby configuration

  1. Prepare N copies of the resource monitoring program (one for each active node), assign a unique name to each copy.

  2. Place each resource monitoring program on its respective active node, and then copy all the resource monitoring programs to the standby node, storing all N files in the same directory structure as the active nodes.

    The following example shows how to place these files when there are three active nodes:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor1
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor2
    Active node3: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor3
    Standby node: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor1
              /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor2
              /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor3

Example of how to modify the resource monitoring program for dual node mutual standby configuration

  1. Create a resource monitoring program by copying the sample and changing the file name, and then change the value of the "SUBSYSTEM" variable to "1".

    [Before]

    SUBSYSTEM="0"

    [After]

    SUBSYSTEM="1"

  2. Create one more resource monitoring program (change the file name), and change the value of the "SUBSYSTEM" variable to "2".

    [Before]

    SUBSYSTEM="1"

    [After]

    SUBSYSTEM="2"

  3. Place the two resource monitoring programs on the active and standby nodes, using the same directory path in both cases.

    The following example shows how to place these files:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor1
  (Standby node2)/opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor2
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor2
  (Standby node1)/opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor1

Callback method to start the monitoring program

Create a callback method for starting the monitoring program.

The following sample program is provided:

/opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstart

This sample program is for 1:1 active/standby configuration. It must be modified for N:1 active/standby configuration and dual node mutual standby configuration.

Copy the sample file, and then modify it to match the cluster system operation. For 1:1 active/standby configuration, this sample file can be used without modification, but should be backed up anyway.

The following examples show how to modify the callback method to start the monitoring program.

Example of how to modify the callback method to start the monitoring program for N:1 active/standby configuration

  1. For N:1 active/standby configuration, create N callback methods to start the monitoring program (one for each active node), and change the "OperationMGR" (an arbitrary string) and "omgr_fmonitor" sections of each file to match the environment.

    Use the same file names as listed in the resource type registration files.

    Change "omgr_fmonitor" to match the name of the file created. The sample program assumes that the monitoring program is in the "opt/FJSVJMCMN/etc/script/SC3.0" directory. If necessary, modify this directory to the directory where the monitoring program is actually located.

    The following example shows how to place these files when there are three active nodes:

    • Callback method 1: omgr_fstart1

      (omitted)

pmfadm -c OperationMGR1 /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor1 $*

    • Callback method 2: omgr_fstart2

      (omitted)

pmfadm -c OperationMGR2 /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor2 $*

    • Callback method 3: omgr_fstart3

      (omitted)

pmfadm -c OperationMGR3 /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor3 $*

  2. Place each callback method for starting the monitoring program on its respective active node, and then copy all of the callback methods to the standby node, storing all N files in the same directory structure as the active nodes.

    The following example shows how to place these files when there are three active nodes:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstart1
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstart2
    Active node3: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstart3
    Standby node: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstart1
              /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstart2
              /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstart3

Example of how to modify the callback method to start the monitoring program for dual node mutual standby configuration

  1. Create a separate callback method for starting the monitoring program for each node (2 files altogether), changing the file name in each case. Modify the information in each file to match the environment.

    The following examples show how to make changes.

    • Callback method 1: omgr_fstart1

      (omitted)

pmfadm -c OperationMGR1 /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor1 $*

    • Callback method 2: omgr_fstart2

      (omitted)

pmfadm -c OperationMGR2 /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fmonitor2 $*

  2. Place the two callback methods for starting the monitoring program on the active and standby nodes, using the same directory path in both cases.

    Use the same file names as listed in the resource type registration files.

    The following example shows how to place these file:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstart1
  (Standby node2)/opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstart2
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstart2
  (Standby node1)/opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstart1

Callback method to stop the monitoring program

Create a callback method to stop the monitoring program.

The following sample program is provided:

/opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstop

This sample program is for 1:1 active/standby configuration. This must be modified for N:1 active/standby configuration and dual node mutual standby configuration.

Copy the sample file, and then modify it to match the cluster system operation. For 1:1 active/standby configuration, this sample file can be used without modification, but should be backed up anyway.

The following examples show how to modify the callback method to stop the monitoring program.

Example of how to modify the callback method to stop the monitoring program for N:1 active/standby configuration

  1. For N:1 active/standby configuration, create N callback methods to stop the monitoring program (one for each active node), and change the "OperationMGR" (an arbitrary string) sections of each file to match the environment.

    The following examples show how to place these files when there are three active nodes:

    • Callback method 1: omgr_fstop1

      (omitted)

pmfadm -s OperationMGR1 TERM

    • Callback method 2: omgr_fstop2

      (omitted)

pmfadm -s OperationMGR2 TERM

    • Callback method 3: omgr_fstop3

      (omitted)

pmfadm -s OperationMGR3 TERM

  2. Place each callback method to stop the monitoring program on its respective active node, and then copy all of the callback methods to the standby node, storing all N files in the same directory structure as the active nodes.

    The following example shows how to place these files when there are three active nodes:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstop1
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstop2
    Active node3: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstop3
    Standby node: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstop1
              /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstop2
              /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstop3

Example of how to modify the callback method to stop the monitoring program for dual node mutual standby configuration

  1. Create a separate callback method to stop the monitoring program for each node (2 files altogether), changing the file name in each case. Modify the information in each file to match the environment.

    The following example shows how to make changes.

    • Callback method 1: omgr_fstop1

      (omitted)

pmfadm -s OperationMGR1 TERM

    • Callback method 2: omgr_fstop2

      (omitted)

pmfadm -s OperationMGR2 TERM

  2. Place the two callback methods to stop the monitoring program on the active and standby nodes, using the same directory path in both cases.

    Use the same file names as listed in the resource type registration files.

    The following example shows how to place these files:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstop1
  (Standby node2)/opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstop2
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstop2
  (Standby node1)/opt/FJSVJMCMN/etc/script/SC3.0/omgr_fstop1

Callback method to check whether resources are active in the event of a failover

Create a callback method to check whether resources are active in the event of a failover.

The following sample program is provided:

/opt/FJSVJMCMN/etc/script/SC3.0/omgr_check

This sample program is for 1:1 active/standby configuration. This must be modified for N:1 active/standby configuration and dual node mutual standby configuration.

The following examples show how to make changes.

Example of how to modify the callback method to check whether resources are active in the event of a failover for N:1 active/standby configuration

  1. Prepare N copies of the callback method to check whether resources are active in the event of a failover (one for each active node), assigning a unique name to each copy.

    Use the same file names as listed in the resource type registration files.

  2. Place each callback method to check whether resources are active in the event of a failover on its respective active node, and then copy all of the callback methods to the standby node, storing all N files in the same directory structure as the active nodes.

    The following example shows how to place these files when there are three active nodes:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_check1
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_check2
    Active node3: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_check3
    Standby node: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_check1
              /opt/FJSVJMCMN/etc/script/SC3.0/omgr_check2
              /opt/FJSVJMCMN/etc/script/SC3.0/omgr_check3

Example of how to modify the callback method to check whether resources are active in the event of a failover for dual node mutual standby configuration

  1. Copy the sample program and then change the value of the "SUBSYSTEM" variable to "1".

    [Before]

    SUBSYSTEM="0"

    [After]

    SUBSYSTEM="1"

  2. Create one more callback method to check whether resources are active in the event of a failover (changing the file name), and change the value of the "SUBSYSTEM" variable to "2".

    [Before]

    SUBSYSTEM="1"

    [After]

    SUBSYSTEM="2"

  3. Place the two callback methods to check whether resources are active in the event of a failover on the active and standby nodes, using the same directory path in both cases.

    Use the same file names as listed in the resource type registration files.

    The following example shows how to place these files:

    Active node1: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_check1
  (Standby node2)/opt/FJSVJMCMN/etc/script/SC3.0/omgr_check 2
    Active node2: /opt/FJSVJMCMN/etc/script/SC3.0/omgr_check2
  (Standby node1)/opt/FJSVJMCMN/etc/script/SC3.0/omgr_check1
PreviousNext
Copyright 1995-2022 FUJITSU LIMITED