It is normal for NetBackup to retry backup failures.  This can be configured
by selecting NetBackup Configuration from the NB admin screen.  Error 196's
will happen if you try to backup after a window has closed.  But it also
intermittently appears when the backup is done writing to one tape, calls
for a second to continue, and notices that the backup window has closed.


Ebon

-----Original Message-----
From: Jeffrey Dykzeul [mailto:Jeffrey_Dykzeul AT raytheon DOT com]
Sent: Monday, October 28, 2002 3:27 PM
To: veritas-bu AT mailman.eng.auburn DOT edu
Subject: [Veritas-bu] Q re NBU write errors and restarts


NBU 3.2 running on Solaris 2.5.1


We experienced some tape drive write errors (exit code 84) during backups
over the weekend. I have some questions about how NBU handled this.

If I use the job monitor GUI to look at the full report for a backup job
which finishes normally, the last two lines in the status pane are

      dd/mm/yy hh:mm:ss - end writing; write time: hhh:mm:ss
      (0) the requested operation was successfully completed

In the full report for each of the jobs that experienced write errors, the
last two lines are

      10/26/02 12:32:41 - end writing; write time: 004:27:46
      (196) client backup was not attempted because backup window closed

The "end writing" message suggests to me that the backup data had completed
its writing to tape, even though there is no "successfully completed"
message. Would it be incorrect to interpret it that way?  There are no
references to write errors in this pane, but if I look in the "all log
entries" report for this time period, there are write errors starting at
12:32:45, immediately after the "end writing" timestamp, one for each of
the backup jobs which were running at that time. Immediately after the
write error messages in the "all log entries" report, there are messages
indicating that client backups were not attempted because the backup window
had closed. Again, there is one such message for each of the backup jobs
which had been running. It looks like NBU tried to restart each job,
perhaps in response to the write errors.

 Is it normal behavior for jobs to restart after failures such as write
errors? Is this configurable somewhere?


Jeff Dykzeul
Raytheon


_______________________________________________
Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu



The contents of this e-mail are intended for the named addressee only. It
contains information that may be confidential. Unless you are the named
addressee or an authorized designee, you may not copy or use it, or disclose
it to anyone else. If you received it in error please notify us immediately
and then destroy it.