On Thu, 14 Nov 2002, Tim Mooney wrote:
>
> Stan-
>
> It's in situations like this (NetWorker "stuck" accessing a tape drive)
> where doing a proper shutdown:
>
>       /sbin/init.d/NSRstartstop stop
>
> caused media database corruption for us on 3 separate occassions under
> 6.1.1.  Our server is also Tru64 UNIX 4.0f.  Reports from another
> gentleman on the list indicated that he has seen the same problem on
> Solaris at 6.1.1.
>
> It hasn't happened since we upgraded to 6.1.2, but then again I haven't
> had a problem with NetWorker thinking it was accessing a tape when it
> wasn't since upgrading to 6.1.2 either.

That's interesting. When I tried running 6.1.2 a few months ago, it was a
disaster. For some reason, nsrd processes would spawn off like nobody's
business. The number of nsrd processes eventually consumed nearly all
available swap space, which brought our backup server to its knees. I had
to revert back to 6.1.1. Some people suggested I enable reverse lookups,
but it has already been enabled; its always been enabled on our server's
NIC.

Last night, the server ran low on swap space again, which I rarely see
with 6.1.1 so I shut down NetWorker via nsr_shutdown and than removed all
the tape library temporary files from /nrs/tmp, then rebooted the system.
After the reboot, I issued the "nsrjb -HE" command for the library in
question. Then I reran a bunch of backup requests that were pending at the
time of the reboot. Everything appears fine now, which is good because I
will be out of the office all day tomorrow.

--
Note: To sign off this list, send a "signoff" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=