failover through Sun Cluster 3.1 when both I/O paths are disconnected

From: Syed, Shadman (Shadman.Syed_at_globalcrossing.com)
Date: 10/26/05

  • Next message: Dave Martini: "SUMMARY: Double click to open PDF's in Thunderbird" 
    Date: Wed, 26 Oct 2005 14:26:32 -0400
To: <sunmanagers@sunmanagers.org>

    I have a 2 node cluster running SC 3.1, on V880s with Solaris 9. Both
    systems are dual connected to HDS9970 through redundant brocade
    switches. We are using MPXIO for path failover and SVM for disk
    management. I have appropriate resource groups setup for failover with
    resources (devices for Informix DB, application filesystems) depending
    on SUNW.HAStoragePlus. Problem is that when I disconnect both paths to
    the storage from a host, it does not failover resources to the 2nd node
    on the cluster....the resource group remains online on the host with
    both fiber channels disconnected. Do I need to change some resource
    properties to get this failover working?

    Rest of cluster integrity tests (there is a list of nearly 40 tests to
    certify the cluster install.....bring a host down hard, soft,
    disconnecting public/private network, etc.) all ran fine.

    Thanks

    Shadman
    _______________________________________________
    sunmanagers mailing list
    sunmanagers@sunmanagers.org
    http://www.sunmanagers.org/mailman/listinfo/sunmanagers

  • Next message: Dave Martini: "SUMMARY: Double click to open PDF's in Thunderbird"

    Relevant Pages

    • Re: 2 node cluster, why failover test fail?
      ... Disconnect the HBA and you'll get a failover. ... there's no point in having just a group with only a disk. ... How can I get resource in "Group 0" failover? ... It seems I can't copy "Cluster ...
      (microsoft.public.windows.server.clustering)
    • Re: Network failure detection and recovery in Windows Server 2003
      ... the user bring the groups back online assuming the ... network AFTER the failover process has stopped, ... Support Escalation Engineer ... times and then the resource is failed and the default group and the sql ...
      (microsoft.public.windows.server.clustering)
    • Re: Network failure detection and recovery in Windows Server 2003
      ... In your scenario, the 10 failovers have already occurred, so the group stays offline until you, the user bring the groups back online assuming the network connectivity has been restored. ... As mentioned, if you reconnect the network AFTER the failover process has stopped, you will have to manually bring the resources back onlin. ... times and then the resource is failed and the default group and the sql group. ...
      (microsoft.public.windows.server.clustering)
    • Re: cannot delete some files on th cluster...weird situation
      ... same thing happens if i failover to the 2nd node or the same one. ... I have tried all the ideas you gave me...taking the resource share offline ... permissions, cluster only manages permissions on the root so I suspect ...
      (microsoft.public.windows.server.clustering)
    • Re: Failover Errors
      ... One thing you can do is take the DiskXtender resource in the groupOffline and then test a failover. ... Plus, in this case it is not the system event log info we would need to look at, it is the cluster log info to see what is happening when the cluster tries to take the disk offline in preparation for a failover. ...
      (microsoft.public.windows.server.clustering)