Backup jobs hanging

[Nelson_Silva]

Hi all

My problem has been persisiting for some weeks now. once a backup job is has started, it hangs at a certain point, or fails outright. The media is then stuck in the drive, and does not respond to any ejecting procedures, be they manual or electronic. The only safe way to get the media out is to shutdown and then restart the server. We use 2 sets of backup media, and I tried erasing the tapes to see if this would alleviate the problem. The first set of erased media worked perfectly, completing all jobs and then ejecting afterwards. However, as we entered the new week of backups, I found the second "monday" tape to be hanging on a job for over sixteen hours, as opposed to the usual 2, with no noticeable tranfer of data.

Is it something to do with the actual Veritas program, or simply an old set of tapes?

Any help would be appreciated.
Thank you.

[Christian_Borth]

Hi,
Try a new tape and check the logs in BE and Event viewer for error messages!

[Rick_Dolan]

I am also having a similar issue. My back up jobs hang indefinitely after a certain point. The only way to stop them is to shutdown the server. I have a robotic tape drive so I don’t really have to eject anything though. I have a new set of tapes for the next scheduled full back up this weekend and will see how that goes. I might try to make back ups of the tape catalogs and then delete them and then re-catalog the tapes and see if that helps. I’m holding off on taking any major attempts and it because I’m playing phone tag with tech support.


Rick

[Christian_Borth]

Both of you will probably find error messages in BE job log and/or event viewer.

It will help a lot if you post'em here ;)

[Rick_Dolan]

Hope I’m not hijacking this tread.
The Job itself has not returned any alerts as it thinks it’s still running.
The event viewer I get this and other similar catalog management Events. There are no Errors yet.

Event Type:Information
Event Source:Backup Exec Catalog Server
Event Category:None
Event ID:34305
Date:10/31/2005
Time:2:12:10 PM
User:N/A
Computer:CASSIOPIEA
Description:
Catalog server database has been synchronized with catalog files in C:\Program Files\VERITAS\Backup Exec\NT\Catalogs\.
Number of backup sets added to the database are 6.
Number of backup sets deleted from the database are 0.
Size of the catalog folder is 727.41 MB.
Size of data processed in the catalog files during synchronization is 4.5227 MB.
Number of non catalog files in the catalog folder are 40.

For more information, click the following link:
http://eventlookup.veritas.com/eventlookup/EventLookup.jhtml

Or Event Type:Information
Event Source:MSSQL$BKUPEXEC
Event Category:(2)
Event ID:17177
Date:11/1/2005
Time:12:00:42 AM
User:N/A
Computer:CASSIOPIEA
Description:
This instance of SQL Server has been using a process id of 972 since 10/31/2005 2:11:40 PM (local) 10/31/2005 8:11:40 PM (UTC).
Data:
0000: 19 43 00 00 0a 00 00 00 .C......
0008: 14 00 00 00 43 00 41 00 ....C.A.
0010: 53 00 53 00 49 00 4f 00 S.S.I.O.
0018: 50 00 49 00 45 00 41 00 P.I.E.A.
0020: 5c 00 42 00 4b 00 55 00 \.B.K.U.
0028: 50 00 45 00 58 00 45 00 P.E.X.E.
0030: 43 00 00 00 00 00 00 00 C.......

[Christian_Borth]

try updating Veritas drivers and SCSI driver. Eventually upgrade to the latest BE 10 build (10d rev5629)

5629 --> http://seer.support.veritas.com/docs/279332.htm

[Ajit_Kulkarni]

Hello Nelson,


I would like to know ...when the bakup job hangs at certain point, does it generates any alerts in the Alerts tab in Backup Exec ?

Also, are there any event ids generated in the Application and System event log related to that backup ?

Have you tried creating a new backup job and checked the results ?

What type of backup it is ? - is it a local backup or remote servers' backup ?

Kindly update.


Regards

NOTE : If we do not receive your reply within two business days, this post would be marked "assumed answered" and would be moved to "answered questions" pool.

[Rick_Dolan]

These are some of the events listed in the windows event view relating to BUexec. This hanging problem occurs on both full and incremental back ups, but not on other things like inventory, and such. Each backups type has its own set of tapes the device switches between. I do have some soft read write errors on the old media. The jobs seem to hang at the end of back up very near the end from the amount of data I see has been written.


17055,17177,34113,57612,57620,34305,34305.
These are after I just tried recataloging the tapes.
I get some interesting errors from the cataloging process, mostly inconsistent data on media and what concerns me is the 57612 event which tells me that I have a tape format issue. I do have a new set of tapes set to run this weekend for full back up. Before I too much I'm going to see how those run as to determine if the tapes are the cause.

[R___Andy__Ander]

Has anyone determined the cause of this problem. We are having a similar issue. The Job Engine dies right after midnight once a week. I found this post because I was searching on the event ID: 17177. It is bizarre since the day of the week is random. We just upgraded to 10d hoping this would resolve the issue, but it has not. We still get this just after midnight, once a week:

The Backup Exec Job Engine service terminated unexpectedly. Event ID: 7034

Any ideas!