FAQ
Hello all,

I've been assigned to build a data warehouse-type database on an Itanium server running HPUX 11.23. This server is currently the passive side of a Veritas Cluster active-passive configuration. RDBMS version 11.1.0.7 will be used. RAC will not be used. The idea is that if one of the servers goes down, the database will be migrated to the active server and started up by the cluster.

I suggested that we use Automatic Storage Management, but the UNIX/storage admins seem to feel that this is not compatible with VCS. Personally I would think that it shouldn't really matter whether the database is using ASM or not. If VCS can move the diskgroups from one server to another, then it be fine. Perhaps assuming that ASM diskgroups can be included in a VCS failover configuration is presumptuous?

Can anyone tell me if this will work or not?
If so, any links to documentation would be greatly appreciated.

Thanks,
Lou Avrami

Search Discussions

  • Matthew Zito at May 15, 2009 at 6:50 pm
    It'll work fine, just make sure the ASM instance is configured in the
    resource group as well, and add it as a dependent resource for the
    Oracle instance. Your failover time will be increased, obviously, as
    you will have to wait for ASM instance to start before VCS can start up
    the database.

    There doesn't seem to be a huge advantage to using ASM in this config,
    though, since you could easily put the data on a cooked filesystem on
    top of an LVM, which would give you the disk concatenation.

    Thanks,
    Matt

    -----Original Message-----
    From: oracle-l-bounce_at_freelists.org
    On Behalf Of Lou Avrami
    Sent: Friday, May 15, 2009 2:43 PM
    To: oracle-l@freelists.org
    Subject: ASM in a Veritas Cluster Software environment

    Hello all,

    I've been assigned to build a data warehouse-type database on an Itanium
    server running HPUX 11.23. This server is currently the passive side of
    a Veritas Cluster active-passive configuration. RDBMS version 11.1.0.7
    will be used. RAC will not be used. The idea is that if one of the
    servers goes down, the database will be migrated to the active server
    and started up by the cluster.

    I suggested that we use Automatic Storage Management, but the
    UNIX/storage admins seem to feel that this is not compatible with VCS.
    Personally I would think that it shouldn't really matter whether the
    database is using ASM or not. If VCS can move the diskgroups from one
    server to another, then it be fine. Perhaps assuming that ASM diskgroups
    can be included in a VCS failover configuration is presumptuous?

    Can anyone tell me if this will work or not?
    If so, any links to documentation would be greatly appreciated.

    Thanks,
    Lou Avrami

    --
    http://www.freelists.org/webpage/oracle-l

    --
    http://www.freelists.org/webpage/oracle-l
  • Matthew Zito at May 15, 2009 at 9:57 pm
    Rajeev,

    You don't want to do any of that. Just configure the ASM instance to be shut down as part of the failover. That will clean everything up. Then it will be started up on the destination node, and the diskgroups mounted automatically.

    Thanks,
    Matt

    --
    Matthew Zito
    Chief Scientist
    GridApp Systems
    P: 646-452-4090
    mzito_at_gridapp.com
    http://www.gridapp.com

    -----Original Message-----
    From: oracle-l-bounce_at_freelists.org on behalf of Rajeev Prabhakar
    Sent: Fri 5/15/2009 5:53 PM
    To: avramil_at_concentric.net
    Cc: oracle-l@freelists.org
    Subject: Re: ASM in a Veritas Cluster Software environment

    Hello Lou,

    Please ensure that the resource switchover/failover procedures take care of the
    dismounting and mounting of appropriate diskgroups on the relevant environments
    (going-to-be-passive and going-to-be-active node). So, no dangling
    file descriptors
    (howsoever remote a possibility ) or anything that prevents exclusive access to
    those disk groups on the surviving node).

    -Rajeev
    On Fri, May 15, 2009 at 2:42 PM, Lou Avrami wrote:
    Hello all,

    I suggested that we use Automatic Storage Management, but the UNIX/storage admins seem to feel that this is not compatible with VCS. Personally I would think that it shouldn't really matter whether the database is using ASM or not. If VCS can move the diskgroups from one server to another, then it be fine. Perhaps assuming that ASM diskgroups can be included in a VCS failover configuration is presumptuous?
    --
    http://www.freelists.org/webpage/oracle-l

    --
    http://www.freelists.org/webpage/oracle-l

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
grouporacle-l @
categoriesoracle
postedMay 15, '09 at 6:42p
activeMay 15, '09 at 9:57p
posts3
users2
websiteoracle.com

2 users in discussion

Matthew Zito: 2 posts Lou Avrami: 1 post

People

Translate

site design / logo © 2022 Grokbase