IBM Support

IJ45674: GLVM VGS MAY NOT BE SYNCD AT CLUSTER START

Subscribe to this APAR

By subscribing, you receive periodic emails alerting you to the status of the APAR, along with a link to the fix after it becomes available. You can track this item individually or track all items by product.

Notify me when this APAR changes.

Notify me when an APAR for this component changes.

 

APAR status

  • Closed as program error.

Error description

  • When a PowerHA/GLVM cluster is inactive all
    Remote Physical Volume Client and Server
    devices are in Defined state on all nodes.
    
    When the cluster is started and the first
    node, which joins, acquires a RG with
    GMVG_REP_RESOURCE resources, it starts
    a cl_sync_vgs process after the GLVM
    is varied on, even if a syncvg process
    can't succeed while the RPV server devices
    at the remote node are still in Defined
    state.
    
    Once a remote cluster nodes starts
    and acquires the RG into 'ONLINE SECONDARY'
    state (RPV server devices are made
    Available), the primary node is supposed
    to start another cl_sync_vgs process.
    
    But in large environments it can happen that
    the first cl_sync_vgs process is still running
    (serveral syncvg processes have already failed)
    and so no new cl_sync_vgs process is started:
    +[RG]:glvm_join_cleanup[285] /usr/bin/ps -eo pid,args
    +[RG]:glvm_join_cleanup[285] grep -w cl_sync_vgs
    +[RG]:glvm_join_cleanup[285] grep -vw grep
    +[RG]:glvm_join_cleanup[285] grep -w datavg
    +[RG]:glvm_join_cleanup[285] 1> /dev/null 2>& 1
    +[RG]:glvm_join_cleanup[286] ((0!=0))
    
    
    If that happens LVs of the GLVM VG remain
    unsynchronized (stale).
    

Local fix

  • Manually sync stale LVs, if there are any.
    

Problem summary

  • In larger environment, on cluster start with GLVM configuration,
     once the primary node is online cl_sync_vgs is triggered
    without checking the RPV servers are in defined state in remote
    nodes leading to cl_sync_vgs failed and LVs remain
    unsynchronized.
    

Problem conclusion

  • As fix checking the state of RG in cl_sync_vgs to ACQUIRING
    SECONDARY or ONLINE SECONDARY before proceeding to sync to avoid
     the fail scenario.
    

Temporary fix

Comments

APAR Information

  • APAR number

    IJ45674

  • Reported component name

    POWERHA SYSMIR

  • Reported component ID

    5765H3900

  • Reported release

    727

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2023-03-02

  • Closed date

    2023-03-02

  • Last modified date

    2023-03-02

  • APAR is sysrouted FROM one or more of the following:

    IJ44306

  • APAR is sysrouted TO one or more of the following:

Fix information

  • Fixed component name

    POWERHA SYSMIR

  • Fixed component ID

    5765H3900

Applicable component levels

[{"Business Unit":{"code":"BU008","label":"Security"},"Product":{"code":"SGL4G4","label":"PowerHA"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"727"}]

Document Information

Modified date:
02 March 2023