Results 1 to 4 of 4
  1. #1
    Join Date
    Sep 2006
    Posts
    33

    Question Unanswered: Changing device name and paths while migrating using install/load approach

    Hi all,

    AFAIK I can change the device names and their paths when I am migrating a Sybase server using install/load technique.
    However, I am not sure if this will hurt my disaster recovery. I mean is it necessary to have devices with same name and path for recovery in a new migrated environment to continue with the earlier disaster recovery mechanism? Please help.

    Thanks in advance.

  2. #2
    Join Date
    Jan 2003
    Location
    Geneva, Switzerland
    Posts
    353
    I'm not sure that I understand the question.

    The device paths are specific to the database instance. The database dumps that you want to load aren't linked to specific devices, so as long as the target databases have been create with the right sizes things should work fine.

    What disaster recovery issue do you see with changing the device paths ?

    Michael

  3. #3
    Join Date
    Sep 2006
    Posts
    33
    Sorry Michael for complicating things unnecessarily.

    Actually I want to know what disaster recovery mechanisms can be in place. One method of achieving this, as I understand is to maintain dumps of the databases. This will not be impacted by any change in the device name or paths as I understand. However, I am not sure of other methods of recovery and whether they will be impacted if the device names and their paths change in the migrated environment.
    Please hekp.

  4. #4
    Join Date
    Jan 2003
    Location
    Geneva, Switzerland
    Posts
    353
    Let's put it this way. Disaster recovery means being able to rebuild your database / dataserver if something goes wrong (you lose a disk, etc).

    In order to do that you need to have copies of the dataserver software (i.e. ASE), the scripts needed to recreate the devices (disk init..., etc.) and database and transaction log dumps that you can use to reload the most recent version of the data. I usually also store bcp OUT copies of some system tables (syslogins, sysdevices, sysusages, sysloginroles and possibly a few others). These need to be stored so that they are safe, and also so that you can have access to them easily.

    You may of course also have requirements that go beyond the above depending on the application(s) that the database serves.

    From my perspective this is really unrelated to the migration you are working on, or to changing the device path, etc, apart from adapting your database dump scripts with the new paths as necessary.

    Michael

Posting Permissions

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