IBM Support

db2start fails with SQL1517N in a Purescale environment

Troubleshooting


Problem

db2start fails with SQL1517N in a Purescale environment when there are inconsistencies between the db2nodes.cfg file and the cluster manager.

Symptom

db2start
SQL1517N db2start failed because the cluster manager resource states are inconsistent.

Cause

This error indicates that there are inconsistencies between the db2nodes.cfg file and the cluster manager.

Diagnosing The Problem

- Check the db2nodes.cfg file on all machines in the cluster.

- Review the output of lssam to check for corresponding resource groups and states.

- Review the output of db2instance -list in a IBM DB2 PureScale setup.

- Refer to db2diag.log for information about the problem.

- Verify the resources for all the nodes using :


    db2cluster -cm -verify -resources

Resolving The Problem

Recreating the resource groups to match the changed configuration file will resolve the error.
Steps to resolve the error are :


    1) Edit the db2nodes.cfg file on all machines to reflect the correct configuration.

    2) Run ipclean.

    3) Check the ps -ef output to ensure all db2 related processes are gone.

        ps -ef | grep db2
    4) As ROOT - stop and start the domain using these commands:
      stoprpdomain -f <domain_name>  
      Monitor the domain is down by using lsrpnode to check all nodes are in OFFLINE state.
      startrpdomain <domain_name>
      Monitor the domain comes up by using lsrpnode to check all nodes are in ONLINE state.

    5) As INSTANCE OWNER - recreate cluster resources and this can be done in a single step using 'db2cluster -cm -repair -resources'

    If this command returns an error as below, try the following commands to delete and create new resources.


      db2cluster -cm -repair -resources
      Resources could not be repaired for the DB2 instance, and the resource state is now inconsistent.  Refer to the db2diag.log for more details.
      db2cluster -cm -delete -resources
      Check the lssam output to verify all resources have been deleted.

      db2cluster -cm -create -resources
      Check the lssam output to verify all resources have been created.

    6) Verify resources
     db2cluster -cm -verify -resources  

    7) Run db2start

[{"Product":{"code":"SSEPGG","label":"Db2 for Linux, UNIX and Windows"},"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Component":"High Availability - PureScale","Platform":[{"code":"PF002","label":"AIX"},{"code":"PF016","label":"Linux"},{"code":"PF027","label":"Solaris"}],"Version":"9.8;10.1","Edition":"","Line of Business":{"code":"LOB10","label":"Data and AI"}}]

Document Information

Modified date:
16 June 2018

UID

swg21611499

Page Feedback