Back up completed under Job Status but not available for restore.

[db]

Hello all,

My company is having a strange problem that I am hopeful someone can assist us with.  First, a bit of background.

Software: Backup Exec v9.1

Server: Novell NetWare 6.0 SP5

Tape backup device: (Brand is HP, type varies, but is always a single stand-alone tape backup device)

Backup rotation:

(A single tape encompasses our incremental jobs from Mon-Thurs.)

Monday - erase the tape, run an incremental job on a single tape, set to over-write media

Tuesday - run an incremental job on the same tape, append to tape, terminate if no appendable media found

Wednesday - run an incremental job on the same tape, append to tape, terminate if no appendable media found

Thursday - run an incremental job on the same tape, append to tape, terminate if no appendable media found

(A single, separate tape, encompassess our full backup job on Friday.)

Friday - erase the tape, run a full job on a separate tape, set to over-write media

(I realize this is the BE for Windows forum, but I can't find a NW forum.)

The issue we are having is this.  Our backups complete successfully all of the time, and show up as "Normal" in the BE Administration Console -> Administration -> Job Logs screen.  However, when we go to do a restore, under Jobs -> Restore -> Make selections by volume -> find your server/volume -> the backup jobs for the incremental jobs (Mon-Thurs) do not show up.

In going through some various testing, we have found that the most recent incremental job appears as available for the restore.  In other words, if we looked at doing a restore on Friday, only the Thursday incremental job appears as being available.  If we looked on Wednesday, only the Tuesday incremental job is there.

The weekly, full jobs, are always available (we keep our catalogue history for 100 days). 

I further want to note that we are aware that under this system, that tape is erased every Monday, so we would only have 4 days worth of data, then on Monday, it gets over-written and started over again.  But if we wanted to do a restore on Friday...we should be able to see the Mon-Thurs jobs, shouldn't we? 

We have tried numerous things to correct this issue, including changing to differential backups, re-creating the job schedule/policies, re-installing BE, going through all of the configuration/options/settings on the NetWare server, and other miscellaneous things.  But nothing is correcting this rather perplexing issue.

If anyone out there has any suggestions, we would really appreciate the help.  We currently have about 25 server installations of BE v9.1 in our "remote" sites, and this issue is occurring across all 25 servers.

Thank you very much!

[Ken_Putnam]

I've never used BENW, but if this was happening on a BEWS system I'd say

1) double check the OPP of the daily media set?  It shold be two days (since it is reset every time the backup closes

2)  make sure that the global  media protection option is set to FULL or PARTIAL, not none

3)  Recreate the daily job and selection list from scratch and see what happens

(and yes, I agree that the location of the Netware Area makes absolutely no sense at all.  I've mentioned it several times via Email to WEBMaster and on the Enahancements page)

Message Edited by Ken Putnam on 06-13-2008 08:33 AM

[db]

Hi there,

Thank you very much for the prompt response. 

The options that you have described do not quite match up exactly with what we see on BE for NW.  However, they are close, so here is what I tried.

1. Unless I am not looking in the right spot, there is no options or configuration on how/where to configure the OPP of the daily media set.

2. Under Options -> Media Server -> Media -> Media Overwrite Protection Level.  This option was set to "None", which is kind of what you were referencing, above, in a somewhat combination of your points #1 and #2.  I set this to (first) Partial, then (second) Full, but neither made a difference.

3. I have also re-created the daily jobs from scratch, but also, that has not made any difference.

I am wondering if we are simply over-looking something, that this is simpler than we are making it out to be.  I know it is a long shot, but we are quite stumped. 

As the product we are running (v9.1) is/was end of life quite some time ago, I'd assume there is no post-patches or what not that may or may not address this?  Unlikely, but it's a long shot for us.

If you have any other suggestions, we would be most appreciative.  Failing that, when we migrate our servers from Netware 6.0 to OES2, the problem will likely fix itself.  However, that's 6-12 months away.... :)

Thanks again for the assistance!