I concur with Ed. By default I don't exclude anything as I don't control the applications and changes made to them and have found that covering my back is most important. The issue does arise, as Ed mentioned, when they add new drives in a cluster but I think that the new 6.5 client addresses that issue with a Netbackup cluster resource.
Regards. From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ed Wilts Sent: Thursday, November 06, 2008 2:45 PM To: VERITAS-BU@mailman.eng.auburn.edu Subject: Re: [Veritas-bu] What are your thoughts on filesystem monitoring? On Thu, Nov 6, 2008 at 2:58 PM, MPish44 <[EMAIL PROTECTED]> wrote: My thought is while that is great perhaps its time to move to an "excludes" type of system, granted its a manual effort across hundreds of Unix servers and we will probably have to manage the excludes in many the same ways... By default, a file system usually needs to be backed up. You may have bad backups if it happens to contain something like an Oracle database, and you won't have an application-consistent backup, but at least you'll have something. It definitely gets harder with active/passive clusters since NetBackup doesn't really support them very well, but something is better than nothing. When your admins tell you how the file system really needs to be handled, you can adjust your backup policies to do the right thing. By if they forget to tell you anything at all, at worst case you'll have a generic file system backup. We tend to do most of our backups with a generic all_local_drives type of policy coupled with some exclude for things like database volumes, and then use application-specific policies to cover them. Simple is always good. The less you have to muck with it, the more likely you are to be successful. .../Ed Ed Wilts, RHCE, BCFP, BCSD, SCSP, SCSE [EMAIL PROTECTED]
_______________________________________________ Veritas-bu maillist - Veritas-bu@mailman.eng.auburn.edu http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
