Backup looks completed but reports Failed.

[Snowdaddy]

Using v8.6, scheduled backups have been running fine for months. Past 2 have reported failed. Checked logs, no '^^^^^^' to mark error, everything looks fine, even looks like completes backup, but I get:

Backup completed on 6/3/2002 at 10:23:11 PM.
Backed up 1999 files in 79 directories.
Processed 261,139,975 bytes in 2 minutes and 22 seconds.
Throughput rate: 105.2 MB/min
----------------------------------------------------------

==========================================================
Job ended: Monday, June 03, 2002 at 10:25:06 PM
Job completion status: Failed
==========================================================

Running on W2K server, no active directory, Compaq DL380 Server, Compaq 35/70 GB DLT.....PLEASE HELP!!

 

[mot98]

Check to make sure that there are no files that are being skipped due to "in use" or "no permission" state.

There are a number of reasons it could fail..but at least the data is being backed up.

HTH's... mot98..

"Where's the beer?"

 

[Snowdaddy]

The log shows no "no permission" states, and in use files are minimal, in addition, the in use files have not affected my previous 5 months of 'good' backups. So I think that there has to be another avenue to explore, but I also thank you for your suggestion.

Oddly enough, I can restore files from the "failed backup" like any other "good backup". It is like the program is just misreporting the end of the backup.

The one thing that I did notice is that the failure notice always seems to come about 2 1/2 minutes after the last backup command of the job has completed, like it is not being able to 'close the session' properly, but there is no indication of this.

All comments welcome. Although it appears like this is not really affecting my systems, it bothers me highly. Thanks in advance for all your help.

 

[tnrcc11]

Our B/up was indicating files/folders were all backed up but the Job History would display "Operation ended on tape #1 in partition 'HA:0_ID_:5_LUN:0_QUANTUM_DLT4000'. Job completion status : Error" at the end. The Verification process would not run after the error lines. A cleaning tape was used and no more errors were reported.

It's a thought.

NW v5.1, BE v8.5

 

[dmm2222]

I had a similar problem until I installed the Remote Agent on the client machines. That cleared up the "Failed" flag as well as significantly speeding up the backup.

 

[nttechiesince93]

I'm with dmm2222, you should check all your builds of remote agents and make sure you have the latest ones.
And, having been in tech support for servers for a long time, my #1 problem with calls were typically related to backups. Most of the times, the solution was to clean the drive. The tape drive manufacturers state you should only have to clean once a month or so, but I always suggested creating a Clean Job and running it weekly.
A tape drive is no different than video players, DVD players or anything else that has to "read" media and therefore must be cleaned regularly if used regularly. Also, have you also had any software changes? Applied any SPs or anything like that?

 

[protos]

Should be
^ ^ ^ ^ ^

5 carots to search for in BE 8.6

or look for keywords,
unable, access, denied
error

worse case, call support or email your whole log file
to support@veritas.com
or paste the whole logfile in here and i'll check it out.

 

[Snowdaddy]

Thank you all for your suggestions. I have cleaned the tape drive. Remote agents are running (although I should check for newer versions). And I think that I have narrowed it down to a bad port in the switch, or a bad NIC card on the backup server. Both Gigabit Fiber and should not go bad, but the server loses connectivity (if even temporarily) and that portion of the back up fails. Thanks again for your help.