ADSM-L

Scheduled backup failing

2001-03-14 15:02:31
Subject: Scheduled backup failing
From: "Gregory J. Ira" <gji01 AT HEALTH.STATE.NY DOT US>
Date: Wed, 14 Mar 2001 15:05:02 -0500
We've been working on this problem for months now with no success and I'm hoping
someone else has ran into this and found a solution.  One of our clients is
trying to do a scheduled backup of a Sun server.   It always fails with an
"unable to contact client" error when using prompted or "client has missed it's
scheduled window" when using polling.  If we try bringing down and restarting
their server it sometimes works (not often) the first time but never a second.
They are able to initiate a backup manually and have been doing so since we
registered the node, however they would prefer to be able to run the backup
during off-hours.  We've upgraded the client on their server twice to no avail.
They've also upgraded the code on their machine a few times as well.  We've
added/double checked the DSM_DIR in the environment variables.  Examining a core
file dump from their system showed the following messages:

ANS0101E NLInit: Unable to open message repository '%s'.
ANS0110E LogMsg: Unable to open error log file '%s' for output.
ANS0118E LogSchedMsg: Unable to open schedule log file '%s' for output.
ANS0105E ReadIndex: Error trying to read index for message %u from repository
'%s'.
ANS0106E ReadIndex: Message index not found for message %u.
ANS1111M Error retrieving message %d from dsmclientV3.cat.
ANS0109E ReadMsg: Insufficient memory for allocation of message structure.
ANS0107E ReadMsg: Error trying to read header for message %u from repository
'%s'.

We tried completly wiping out the client and reinstalling it and verifying that
the various items mentioned above (message
repository...dsmsched...dsmerror..etc) were available. We tried changing the
scheduled backup time,used a sniffer to make sure ADSM was reaching the problem
server(it was), and the hand full of usuall checks for simple things like the
scheduler not running etc..  The odd thing is that we can get the schedule
backup to work once every now and then.  I should point out that the users
running this server are no rookies to ADSM and have about a dozen other servers
all running with no problems, some with the same exact setup as this problem
server. I found no closed APARs refering to this problem and before I go and
open one with Tivoli I thought I'd give this list a try.  Any help would be
apprectiated.

Some info:


ADSM Server Version 3, Release 1, Level 2.20

Client server info ( i took out server name and replaced with *)
SunOS ****** 5.7 Generic_106541-14 sun4u sparc SUNW,Ultra-Enterprise-10000

adstar client info
Command Line Backup Client Interface - Version 3, Release 1, Level 0.8



Greg Ira
ADSTAR/Tivoli/MVS Systems
NYS Dept. Health/BHNSM
Albany, N.Y.  12237
gji01 AT health.state.ny DOT us

I do not fear computers.  I fear the lack of them.  - Isaac Asimov
<Prev in Thread] Current Thread [Next in Thread>
  • Scheduled backup failing, Gregory J. Ira <=