On Thu, 3 Apr 2003, Levi Bjørge wrote:

LB> we experienced the same problem, also on NetWorker 6.1.3, but Solaris 2.8.
LB> We have a STK L700e library with 678 slots (firmware 3.03.02).
LB> And the symptom seems to be the same, specially when a tape is marked "Full"
LB> and the system need a new one.
LB> Another symptom was that a nsrjb or nsrjb -v was very slow when the jukebox
LB> had a lot of emty slots or new blank tapes
LB> (unknown to the mediadb).
LB>
LB> I'll got a fix from Legato - LGTpa45147:
LB>
LB> Here are some information about the problem:
LB>
LB> "From your description of the problem this issue seems to be very close to a
LB> known problem with nsrjb of NetWorker 6.1.x, if lots of slots in the
LB> jukebox/roboter are still empty.
LB>
LB> nsrjb ?C runs into ENOMEM, if there is a huge library with many empty slots
LB> and many volumes in the media db. For occupied slots the info is only about
LB> the requested volume.
LB> For every empty slot heap memory of appr. 160k is used, for occupied slots
LB> only about 250 bytes."

Hm, that's not exactly the issue here. Are you sure this is the correct
patch? nsrjb eats all CPU (loops). I have received information OffList
that I need LGTpa50956 - and I found an nsrjb file in /pub/outgoing on the
FTP server, however, it didn't seem to be a binary executable at all. It
could have been compiled for some other platform, but it wasn't even an
ELF file. Anyway, I've sent a mail to support AT legato DOT com and requested
these patches, let's see what happens. :)

//Oscar

Ps. There's some weird shit on legato's FTP site:
ftp://ftp.legato.com/pub/outgoing/Tim/testimony
Quite entertaining. ;)

--
Note: To sign off this list, send a "signoff networker" 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.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=