How to configure a UNIX Event Agent to forward all messages to a Windows Manager

Document ID : KB000056312
Last Modified Date : 14/02/2018
Show Technical Document Details

On the UNIX Event Agent:

1) Make sure that the REMOTEDB entries in $CAIGLBL0000/scripts/envset point to the Windows manager node:

CAI_CAL_REMOTEDB=<manager node> CAI_OPR_REMOTEDB=<manager node>

If changes are made you must source the envset script:

. $CAIGLBL0000/scripts/envset

You can the check these changes are in the environment by issuing the UNIX command:

    env | grep REMOTEDB

    And then and recycle Unicenter

    2) Make sure that there is a remote entry in the $CAIGLBL0000/cci/config/<nodename>/ccirmtd.prf file for the Windows manager.

    If you need add an entry you must recycle Unicenter.

    On the Windows Manager:

    3) Create a Message Record & Action to forward all messages to the Windows manager node:

    Message Record: Message Id: *
            Eval node: <Agent Nodename>
            Search all messages: <should be checked>

    Message Action: Sequence number: 10
            Action: Forward         Node: <Windows Manager Nodename>
    4) Make sure that the ccirmtd.rc has a remote line for the UNIX host.

    If you need to make changes to this file you will need to stop and restart CCI remote:

        ccicntrl stop rmt
        ccicntrl start rmt

    On both Manager and Agent:

    5) Now run an opreload, first on the Windows Manager node, then on the Unix Agent node.

    On the Windows side you should see confirmation that the Opr database has been reloaded in the console,

    On the UNIX side you should see a confirmation that Opr database has been reloaded:

        # oprcmd opreload

        .CAOP_I_DBCACHED Database (re)loaded on Fri Feb 7 15:35:17 2003

    This message may take several seconds to be displayed.

    All messages should now be forwarded to the Windows manager. You can test this by issuing a cawto message locally on the UNIX agent. You should see this message on the console of the Windows manager.

    Troubleshooting

    1) First make sure that you can ping both ways by hostname.

    2) Then check CCI communication is working both ways using ccir/ccis. Run ccir on one host and on the other host run:

    ccis &l    t;remote hostname> 3

    You should see 3 test messages on the host where ccir is running. Repeat this process in reverse by running ccir/ccis in the opposite direction.

    If CCI is not working check that CCI ccirmtd configuration file on both sides are correct and have remote lines for the remote host. This file is:

    Windows: \TND\CAIUSER\ccirmtd.rc (3.0)
        \TNG\CAIUSER\ccirmtd.rc (2.x)

    UNIX: $CAIGLBL0000/cci/config/<nodename>/ccirmtd.prf

    Follow procedures outlined above to refresh CCI after making changes.

    3) You can also check Opr is available and running using oprping from the remote host.

    4) Check that the Message Record and Action for forwarding has correct entries and has been saved from the Windows GUI.

    On the UNIX side you can check the timestamps of the rdw database files stored in $CAIGLBL0000/tmp to make sure the message record and action database has been updated:

    # ls -l total 6 -rw-rw-rw- 1 root other 652 Feb 7 15:35 msgactem.rdw -rw-rw-rw- root other 795 Feb 7 15:35 msgrecem.rdw -rw-r--r-- 1 root other 5 Feb 7 12:54 stardaemon.pid # 

    These timestamps should be the same as the message received after running the opreload command:

    # oprcmd opreload

      .CAOP_I_DBCACHED Database (re)loaded on Fri Feb 7 15:35:17 2003

      5) Make sure on the UNIX host that the Manager node has been correctly set:

      env | grep REMOTEDB

      Check that the following environment variables have been correctly set:

      CAI_CAL_REMOTEDB=<manager node>
      CAI_OPR_REMOTEDB=<manager node>

      If you need to make changes follow the procedure outlined above.