Disaster recovery plan

  • I have a production server and a log shipping server which are acting as a primary and secondary server respectively in the log shipping setting.

    If I create a disaster recovery plan for my production server, what all steps should I include

    wrt to log shipping?

    Regards
    Sushant Kumar
    MCTS,MCP

  • We have setup weekly full backup with daily diff. backup on the primary server. You can try the same.

    "More Green More Oxygen !! Plant a tree today"

  • Also you can schedule the job which will copy your backups regulary to some safe place and can be used in time of recovery.

    ----------
    Ashish

  • SKYBVI (7/28/2010)


    I have a production server and a log shipping server which are acting as a primary and secondary server respectively in the log shipping setting.

    If I create a disaster recovery plan for my production server, what all steps should I include wrt to log shipping?

    If secondary-server is off site then that's your DR platform - provided you have in place the network and security setup needed to switch your apps from Primary to Secondary site in case of a Disaster.

    T-Log backup as well as log shipping have to be in line with what the company is willing to loose in case of a disaster meaning, if the company is willing to loose the last 15 minutes of activity then T-Log backups have to occur every 15 minutes and logs have to be shipped immediately to secondary site.

    Last but not least, please do not forget to plan how are you going to move back from DR site to Primary site once the Disaster situation is remediated.

    _____________________________________
    Pablo (Paul) Berzukov

    Author of Understanding Database Administration available at Amazon and other bookstores.

    Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.
  • @ ashish

    What is the use of copying backups to a safe place when already my log shipping is in place.

    If some disaster happens to my production server, then can't I restore the databases on the secondary server where logs are being shipped.?

    What can be the procedure to move the production server to the log shipped server.

    The databases on the log shipped server are in stand by/read only mode so i should restore the transaction logs there so that the databases become usable?

    Regards,

    Sushant

    Regards
    Sushant Kumar
    MCTS,MCP

  • hi Sushant,

    copying backup file to safe location I suggested for , incase something happen your secondary and get corrupted.

    Anyway, yes so to bring your seconday as primary, you need to restore the last log with recovery which will make your database as opertation on secondary.

    But before last log restoration, just make sure that the one you restoring is definately the last and final log from primary.

    ----------
    Ashish

  • Okk..

    Why just the last log has to be restored? Not the logs before that??

    Assume, i have a full backup nightly and have transaction logs every 15 during day time.

    Then, if i want to restore the database at any time during day, then i should restore the last .bak file

    and then all the transaction logs since the backup time to the last transaction log sequentially...

    Is it fine?

    Regards,

    Sushant

    Regards
    Sushant Kumar
    MCTS,MCP

  • then what is the meaning of your secondary if you have to restore this much?

    ----------
    Ashish

  • I think you're confused a little on log shipping.

    The way log shipping works on the primary

    - Make a full backup

    - Make log backups

    - Make periodic full backups

    It just runs.

    On the secondary:

    - Restore full backup with NORECOVERY

    - copy logs from the primary every xx time (hours, minutes, days)

    - restore logs with NORECOVERY after yyy delay.

    At any point in time, the secondary ought to be fairly close to the primary in terms of data updates. Some people restore logs right away, maybe copying them every 5 minutes. In this case, the secondary typically has one log to restore in the event the primary fails.

    Some people delay the log restores by an hour, to catch any issues on the primary (like a delete without a WHERE). In this case, you'd need to restore all the log backups you make in that hour to catch up.

    There is not automatic switchover for clients, unlike for mirroring. So you need to "redirect" all clients to the secondary server. If you use the IP or Windows host name, then each client has to change that. If you use DNS, you can change the host entry, but you might need to flush the entries on clients. You could rename the secondary to the primary as well.

    No matter what, you might have network ARP issues as well, or other items that you have to flush from the caches.

  • Is your log shipping server in the same office as your primary server?

    If so you may need an offsite backup in case of flood, fire, terrorist attack.

  • hi,

    you can try to replication method . It can instantly take the backup

Viewing 11 posts - 1 through 10 (of 10 total)

You must be logged in to reply to this topic. Login to reply