Results 1 to 9 of 9
  1. #1
    Join Date
    Jul 2004
    Posts
    306

    Unanswered: HADR container paths etc.

    9.7 on Linux

    Hey guys,

    If I have to productions servers used for slightly different functions but with some of the same DBs can I set up HADR for each to use the other as the standby?

    eg.
    Server A
    instancea

    clientdb
    sample
    samptst

    Server B
    instanceb

    clientdb
    samptst
    csmrdb

    I have 2 difficulties (as far as I can see).

    #1 Some DBs have the same name so clearly I need to have a new instance for HADR. AFAIK this is supported so I'll just create a db2hadr instance and restore the DBs in there. Will that work?

    #2 The DBs have the same DMS container paths on both servers (eg db2data, db2index, db2blob). I know redirecting containers is not supported, is there something else I can do to get around this?

    Are there other gotchas I've missed?

  2. #2
    Join Date
    Aug 2001
    Location
    UK
    Posts
    4,650
    Quote Originally Posted by meehange View Post

    #1 Some DBs have the same name so clearly I need to have a new instance for HADR. AFAIK this is supported so I'll just create a db2hadr instance and restore the DBs in there. Will that work?
    You can setup HADR between databases on the same instance.
    The database names have to be the same on primary and standby.

    #2 The DBs have the same DMS container paths on both servers (eg db2data, db2index, db2blob). I know redirecting containers is not supported, is there something else I can do to get around this?
    If using automatic storage you may be able to get away using different storage paths ... Never have tested it myself and personally, I do not recommend this for a production database
    Visit the new-look IDUG Website , register to gain access to the excellent content.

  3. #3
    Join Date
    Jul 2004
    Posts
    306
    Thanks for the reply

    For #1 What I was actually saying is that I have databases on both servers with the same names so my HADR standbys can't be in the same instance as their identically named but functionally different neighbours. If that makes sense...

    As for #2 - It's a testing environment but it still needs to be highly available (hence the work I'm doing). We're not on automatic storage. I considered this as an option but I was kinda hoping for a configuration or filesystem workaround...

  4. #4
    Join Date
    May 2003
    Location
    USA
    Posts
    5,737
    Quote Originally Posted by meehange View Post
    As for #2 - It's a testing environment but it still needs to be highly available (hence the work I'm doing). We're not on automatic storage. I considered this as an option but I was kinda hoping for a configuration or filesystem workaround...
    HADR will still work, but if container path names are different on the primary and standby, and you submit any DDL that references specific container path names, you will get an error on the standby.
    M. A. Feldman
    IBM Certified DBA on DB2 for Linux, UNIX, and Windows
    IBM Certified DBA on DB2 for z/OS and OS/390

  5. #5
    Join Date
    Jul 2004
    Posts
    306
    Quote Originally Posted by Marcus_A View Post
    HADR will still work, but if container path names are different on the primary and standby, and you submit any DDL that references specific container path names, you will get an error on the standby.
    Ah ok... since this aren't critical HA systems (and production will have a dedicated HADR standby) this might be workable....

    So just do a redirected restore for the standby DBs then, I assume.

  6. #6
    Join Date
    May 2003
    Location
    USA
    Posts
    5,737
    Quote Originally Posted by meehange View Post
    Ah ok... since this aren't critical HA systems (and production will have a dedicated HADR standby) this might be workable....

    So just do a redirected restore for the standby DBs then, I assume.
    That should work, and just remember to not submit any DDL that references specific container path names.
    M. A. Feldman
    IBM Certified DBA on DB2 for Linux, UNIX, and Windows
    IBM Certified DBA on DB2 for z/OS and OS/390

  7. #7
    Join Date
    Aug 2001
    Location
    UK
    Posts
    4,650
    Quote Originally Posted by Marcus_A View Post
    That should work, and just remember to not submit any DDL that references specific container path names.
    Will
    alter tablespace dmstsp1 extend (all 1G)
    work ?
    Visit the new-look IDUG Website , register to gain access to the excellent content.

  8. #8
    Join Date
    May 2003
    Location
    USA
    Posts
    5,737
    I am not sure if the above will work if container paths are different, since the SQL itself is not replicated, only the transaction log is replicated, which may have specific container paths in it for the above command. If there is an error, it should show up in the db2diag.log.
    M. A. Feldman
    IBM Certified DBA on DB2 for Linux, UNIX, and Windows
    IBM Certified DBA on DB2 for z/OS and OS/390

  9. #9
    Join Date
    Jul 2004
    Posts
    306
    If I set it up this way I'll test the above and report back

Posting Permissions

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