Maintenance Plans to Local Storage in a Clustered environment

  • Hi, I’m hoping somebody can be of help or point me to a similar post:

    I have an Active/Active SQL Server 2008 R2 Cluster with two servers, lets call them ServerA and ServerB. On top of this cluster, two SQL Server Instances, Instance1 and Instance2.

    The servers have local storage and we also have Central Storage Configured which contains the data and log files for the various databases in the cluster everything works fine from a cluster point of view.

    We also have in place, two separate Backup routines

    1) Local Backups to local storage, these are Daily and Weekly Maintenance Plans and they backup to the Servers/Active Nodes respective local disks (some of you may be onto my problem at this point already).

    2) Backups to Tape via TSM’s TDPSQL.

    The Tape backups are fine, even in the event of a Cluster Failover, however the issue I have comes with the local Backups in the event of a failover.

    The Daily and Weekly Maintenance Plans are set to backup to \\ServerA\backups$\ and \\ServerB\backups$\ on each physical server respectively. Lets say that ServerB crashes entirely, Instance2 Fails over automatically and the SQL Services continue, however at this point, the path for the Maintenance Plans for Instance2 will be unavailable and the backups will fail as \\ServerB\backups$ will not be available.

    The environment is still covered due to the additional Tape Backup which we consider our primary backup, but I would still like to ensure, if possible, that my local Maintenance Plan backups will continue on their own in the event of a cluster failover, short of having to go to Management Studio and modifying the relevant Maintenance Plan steps manually.

    Bear in mind that I do not have sufficient clustered storage to run the Maintenance Plans to Clustered Drives.

    Any ideas/suggestions?

  • The best thing to do is to set up a disk as a clustered resource in the same group as your instance. That way, if SQL Server fails over, the disk fails with it.

    The next best option is to have a disk with the same driver letter (let's call it X) on both nodes. Then you can set your backups to be made to X:\backups or something similar.

    John

Viewing 2 posts - 1 through 1 (of 1 total)

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