Continental cluster

[smanikrajan]

Hi

I was trying to start the primary cluster of a continental after maintenence but I got the below error

ERROR: Could not determine if package XXXX is enabled

ERROR: Could not determine status of package XXXX 

If you have determined that XXXX is not running on the

QA cluster, the XXXX can be forced to start on the

PRD cluster using the cmforceconcl command.

Cannot start primary package PRD/XXXX because recovery package QA/XXXX conflicts

Error: Cannot start package XXXX:  Disallowed by the ContinentalClusters product

cmrunpkg: Unable to start some package or package instances

But I am able to start the package successfully using cmforceconcl command

Below is the output of cmviewconcl

cmviewconcl -v
ERROR: Could not determine status of package XXXX

CONTINENTAL CLUSTER  concl_XX

RECOVERY CLUSTER  QA

    PRIMARY CLUSTER       STATUS           EVENT LEVEL  POLLING INTERVAL
    PRD                                  Unmonitored      unmonitored  1 min

        CONFIGURED EVENT  STATUS         DURATION    LAST NOTIFICATION SENT
        alert             unreachable    5 min       --
        alert             unreachable    10 min      --
        alarm             unreachable    15 min      --
        alert             down           5 min       --
        alert             down           10 min      --
        alarm             down           15 min      --
        alert             error          0 sec       --
        alert             up             0 sec       --

PACKAGE RECOVERY GROUP  XXXX

    PACKAGE                           ROLE           STATUS
    PRD/XXXX                              primary        up
    QA/XXXX                                 recovery       unknown

Please let me know what could be the problem here

[VK2COT]

Hello,

It looks like your monitor package is having some problems.

a) Run "cmviewcl -v" on both clusters. Check, for example, that:

  1. All primary packages are running on the Primary Cluster,

  2. Data sender packages (if any) are running on the Primary Cluster,

  3. Data receiver packages (if any) are running on the Recovery Cluster,

  4. Monitor package that monitors the Primary Cluster is running on the Recovery Cluster,

  5. Monitor package is running  on the Primary Cluster.

b) Check for any possible errors in /var/adm/syslog/syslog.log.

c) Verify the correct operation of the Continental Cluster daemon:

  # /opt/cmom/tools/bin/cmreadlog -f /var/adm/cmconcl/sentryd.log

d) Wild guess, but you might need to remove/rename logs in /var/adm/cmconcl:

sentryd.all

oldsentryd.all

e) Verify that /etc/inittab has correct entries.

f) And while you are at it, check the ckluster configuration too. For example:

  # cmcheckconcl -v -C  /etc/cmcluster/cmconcl.config

Hoepfully, this should guive you a good start...

Cheers,

VK2COT

[Emil Velez_2]

might want to make sure you do a

su - conclusr

csshsetup -r <host1>  

do this for all of the nodes in the cluster.. might be a communications problem.