Home

Results 1 to 5 of 5

Thread: Cluster volumes go comatose on abend

Hybrid View

  1. #1
    jason.naubur@ddb.ca NNTP User

    Cluster volumes go comatose on abend

    I have a server that regularly abends in a 2 server cluster (I am working
    on it). When either server crashes / reboots the volumes go comatose. In
    the Cluster view I have to offline the drives and then online them to the
    remaining server manually. This netware cluster is not my strong suit and I
    didn't set it up so any advice on where to start looking would be
    appreciated. The Master IP Address resource does migrate properly though.


    Thanks in advance,

    Jason Naubur

  2. #2
    jason.naubur@ddb.ca NNTP User

    Cluster volumes go comatose on abend

    Here is the Cluster report:

    Date: April 25, 2006 2:05:59 PM PDT
    Current Nodes: JULIET
    ROMEO
    Epoch: 11
    Resources:
    Name State Location Lives: Up Since:
    Master_IP_Address_Resource Running ROMEO 4 Mar 22, 2006 9:04:25 AM
    APPS_SERVER Running ROMEO 4 Apr 25, 2006 1:37:58 PM
    DATA_SERVER Running ROMEO 4 Apr 25, 2006 1:38:08 PM
    STUDIO_SERVER Running ROMEO 4 Apr 25, 2006 1:38:14 PM
    Cluster Options
    Cluster:
    Quorum Triggers:
    Number of Nodes:
    2
    Timeout (seconds):
    2
    Protocol Setting
    Heartbeat (seconds):
    1
    Tolerance (seconds):
    8
    Master Watchdog (seconds):
    1
    Slave Watchdog (seconds):
    8
    Maximum Retransmits:
    30
    All Nodes:
    Name Number IP Address Distinguished Name
    JULIET 0 172.17.2.232
    ROMEO 1 172.17.2.234
    Resources:
    Master_IP_Address_Resource
    Policies:
    Ignore Quorum:
    Yes
    Start Mode:
    Auto
    Failover Mode:
    Auto
    Failback Mode:
    Disable
    Assigned Nodes: JULIET
    ROMEO
    Load Script

    IGNORE_ERROR set allow ip address duplicates = on
    IGNORE_ERROR CLUSTER CVSBIND ADD Vela 172.17.2.237
    IGNORE_ERROR NUDP ADD Vela 172.17.2.237
    IGNORE_ERROR add secondary ipaddress 172.17.2.237
    IGNORE_ERROR HTTPBIND 172.17.2.237 /KEYFILE:"SSL CertificateIP"
    IGNORE_ERROR set allow ip address duplicates = off

    Timeout (seconds):
    600
    Unload Script

    IGNORE_ERROR HTTPUNBIND 172.17.2.237
    IGNORE_ERROR del secondary ipaddress 172.17.2.237
    IGNORE_ERROR NUDP DEL Vela 172.17.2.237
    IGNORE_ERROR CLUSTER CVSBIND DEL Vela 172.17.2.237

    Timeout (seconds):
    600
    APPS_SERVER
    Policies:
    Ignore Quorum:
    No
    Start Mode:
    Auto
    Failover Mode:
    Auto
    Failback Mode:
    Disable
    Assigned Nodes: JULIET
    ROMEO
    Load Script

    nss /poolactivate=APPS
    mount APPS VOLID=254
    CLUSTER CVSBIND ADD VELA_APPS_SERVER 172.17.2.238
    NUDP ADD VELA_APPS_SERVER 172.17.2.238
    AFPBIND ADD VELA_APPS_SERVER 172.17.2.238
    CIFS ADD .CN=VELA_APPS_SERVER.OU=IS.L=Vancouver.C=CA.T=DDB.
    add secondary ipaddress 172.17.2.238

    Timeout (seconds):
    600
    Unload Script

    del secondary ipaddress 172.17.2.238
    CLUSTER CVSBIND DEL VELA_APPS_SERVER 172.17.2.238
    NUDP DEL VELA_APPS_SERVER 172.17.2.238
    AFPBIND DEL VELA_APPS_SERVER 172.17.2.238
    CIFS DEL .CN=VELA_APPS_SERVER.OU=IS.L=Vancouver.C=CA.T=DDB.
    nss /pooldeactivate=APPS /overridetype=question

    Timeout (seconds):
    600
    Protocols:
    Virtual Server Name:
    VELA_APPS_SERVER
    CIFS Server Name:
    Vela_APPS
    IP Address:
    172.17.2.238
    Advertising Protocols:
    afp,cifs,ncp
    DATA_SERVER
    Policies:
    Ignore Quorum:
    No
    Start Mode:
    Auto
    Failover Mode:
    Auto
    Failback Mode:
    Disable
    Assigned Nodes: JULIET
    ROMEO
    Load Script

    nss /poolactivate=DATA
    mount DATA VOLID=253
    CLUSTER CVSBIND ADD VELA_DATA_SERVER 172.17.2.231
    NUDP ADD VELA_DATA_SERVER 172.17.2.231
    AFPBIND ADD VELA_DATA_SERVER 172.17.2.231
    CIFS ADD .CN=VELA_DATA_SERVER.OU=IS.L=Vancouver.C=CA.T=DDB.
    add secondary ipaddress 172.17.2.231

    Timeout (seconds):
    600
    Unload Script

    del secondary ipaddress 172.17.2.231
    CLUSTER CVSBIND DEL VELA_DATA_SERVER 172.17.2.231
    NUDP DEL VELA_DATA_SERVER 172.17.2.231
    AFPBIND DEL VELA_DATA_SERVER 172.17.2.231
    CIFS DEL .CN=VELA_DATA_SERVER.OU=IS.L=Vancouver.C=CA.T=DDB.
    nss /pooldeactivate=DATA /overridetype=question

    Timeout (seconds):
    600
    Protocols:
    Virtual Server Name:
    VELA_DATA_SERVER
    CIFS Server Name:
    Vela_DATA
    IP Address:
    172.17.2.231
    Advertising Protocols:
    afp,cifs,ncp
    STUDIO_SERVER
    Policies:
    Ignore Quorum:
    No
    Start Mode:
    Auto
    Failover Mode:
    Auto
    Failback Mode:
    Disable
    Assigned Nodes: JULIET
    ROMEO
    Load Script

    nss /poolactivate=STUDIO
    mount STUDIO VOLID=252
    CLUSTER CVSBIND ADD VELA_STUDIO_SERVER 172.17.2.230
    NUDP ADD VELA_STUDIO_SERVER 172.17.2.230
    AFPBIND ADD VELA_STUDIO_SERVER 172.17.2.230
    CIFS ADD .CN=VELA_STUDIO_SERVER.OU=IS.L=Vancouver.C=CA.T=DD B.
    add secondary ipaddress 172.17.2.230

    Timeout (seconds):
    600
    Unload Script

    del secondary ipaddress 172.17.2.230
    CLUSTER CVSBIND DEL VELA_STUDIO_SERVER 172.17.2.230
    NUDP DEL VELA_STUDIO_SERVER 172.17.2.230
    AFPBIND DEL VELA_STUDIO_SERVER 172.17.2.230
    CIFS DEL .CN=VELA_STUDIO_SERVER.OU=IS.L=Vancouver.C=CA.T=DD B.
    nss /pooldeactivate=STUDIO /overridetype=question

    Timeout (seconds):
    600
    Protocols:
    Virtual Server Name:
    VELA_STUDIO_SERVER
    CIFS Server Name:
    Vela_STUDIO
    IP Address:
    172.17.2.230
    Advertising Protocols:
    afp,cifs,ncp

  3. #3
    Andrew C Taubman NNTP User

    Re: Cluster volumes go comatose on abend

    Pls see TID 10092217
    --
    Andrew C Taubman
    Novell Support Forums Volunteer SysOp
    http://support.novell.com/forums
    (Sorry, support is not provided via e-mail)

    Opinions expressed above are not
    necessarily those of Novell Inc.

  4. #4
    jason.naubur@ddb.ca NNTP User

    Re: Cluster volumes go comatose on abend

    OK. That describes the problem pretty well. So now for the server crashing,
    perhaps it is best to not implement this until I discover the cause of the
    failures.

    Thanks,

    Jason

  5. #5
    Andrew C Taubman NNTP User

    Re: Cluster volumes go comatose on abend

    I'd implement it immediately.

    You can start a new thread in the abends-hangs forum on the crashes, pls
    post full Abend 1 details from the abend log there.
    --
    Andrew C Taubman
    Novell Support Forums Volunteer SysOp
    http://support.novell.com/forums
    (Sorry, support is not provided via e-mail)

    Opinions expressed above are not
    necessarily those of Novell Inc.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •