Problem with schedlued backups

[EGR]

Hi,

we have got a scheduled backup for a Solaris 9 Server running.
Other Solaris 9 backups work without big problems.

But this server is not backed up anymore.

Following entries in dsmerror.log:

"11/07/05 22:07:36 ANS1999E Incremental processing of '/raid0' stopped.

11/07/05 22:07:40 ANS1017E Session rejected: TCP/IP connection failure

11/07/05 22:07:40 ANS1512E Scheduled event 'NODENAME' failed. Return code = 12.
11/08/05 22:07:30 ANS1999E Incremental processing of '/raid0' stopped.

11/08/05 22:07:34 ANS1017E Session rejected: TCP/IP connection failure

11/08/05 22:07:34 ANS1512E Scheduled event 'NODENAME' failed. Return code = 12.
11/09/05 22:07:32 ANS1999E Incremental processing of '/raid0' stopped.

11/09/05 22:07:37 ANS1017E Session rejected: TCP/IP connection failure

11/09/05 22:07:37 ANS1512E Scheduled event 'NODENAME' failed. Return code = 12.

In dsmsched.log are no entries since several days...

When I start an incremental backup manually it works without problems...

Any help is really appreciated.

Regards

/noodles

 

[ProbablyDown]

What does the dsmsched.log say?

 

[EGR]

Hi,

following update:
dsmsched.log does not have any entries.

We updated (new installation) the client on the Solaris 9 machine to 5.3.2.

We changed MTU settings as described in

http://www-1.ibm.com/support/docvie...7E&uid=swg21203302&loc=en_US&cs=utf-8&lang=en

No success!

We get following errors in dsmerror.log:

"11/23/05 22:08:00 ANS1999E Incremental processing of '/mountpoint'' stopped.

11/23/05 22:08:04 ANS1017E Session rejected: TCP/IP connection failure

11/23/05 22:08:04 ANS1512E Scheduled event 'SERVER' failed. Return code = 12."

In dsmsched.log there are no entries at all.

The actlog says following:

ANR2561I Schedule prompter contacting SERVER (session 2517) to start a scheduled operation. (SESSION: 28) Nov 23, 2005 10:00:01 PM
ANR0406I Session 2518 started for node SERVER (SUN SOLARIS) (Tcp/Ip KATALOG(51524)). (SESSION: 2518) Nov 23, 2005 10:00:16 PM
ANR0407I Session 2519 started for administrator ADMIN_CENTER (DSMAPI) (Tcp/Ip TSM.emg-hh.com(4549)). (SESSION: 2519) Nov 23, 2005 10:00:59 PM
ANR0403I Session 2517 ended for node SERVER (). (SESSION: 28) Nov 23, 2005 10:15:15 PM
ANR0482W Session 2518 for node SERVER (SUN SOLARIS) terminated - idle for more than 15 minutes. (SESSION: 2518) Nov 23, 2005 10:16:30 PM

ANR0406I Session 2522 started for node SERVER (SUN SOLARIS) (Tcp/Ip KATALOG(51544)). (SESSION: 2522) Nov 23, 2005 10:19:49 PM
ANE4952I Total number of objects inspected: 6 (SESSION: 2522) Nov 23, 2005 10:19:49 PM
ANE4954I Total number of objects backed up: 0 (SESSION: 2522) Nov 23, 2005 10:19:49 PM
ANE4958I Total number of objects updated: 0 (SESSION: 2522) Nov 23, 2005 10:19:49 PM
ANE4960I Total number of objects rebound: 0 (SESSION: 2522) Nov 23, 2005 10:19:49 PM
ANE4957I Total number of objects deleted: 0 (SESSION: 2522)
ANE4970I Total number of objects expired: 0 (SESSION: 2522) Nov 23, 2005 10:19:49 PM
ANE4959I Total number of objects failed: 0 (SESSION: 2522) Nov 23, 2005 10:19:49 PM
ANE4961I Total number of bytes transferred: 0 B (SESSION: 2522) Nov 23, 2005 10:19:49 PM
ANE4963I Data transfer time: 0.00 sec (SESSION: 2522) Nov 23, 2005 10:19:49 PM
ANE4966I Network data transfer rate: 0.00 KB/sec (SESSION: 2522) Nov 23, 2005 10:19:49 PM
ANE4967I Aggregate data transfer rate: 0.00 KB/sec (SESSION: 2522) Nov 23, 2005 10:19:49 PM

ANR2579E Schedule DAILY in domain STANDARD for node SERVER failed (return code 12). (SESSION: 2523) Nov 23, 2005 10:19:54 PM



Manual Backups work without problems.

Help is really appreciated.

Regards
/noodles

 

[LED888]

If there are no entry in the dsmsched.log, is the log being redirected to another location or name other than dsmched.log?

Is the backup beging started by a cron job?

Or do we have the parameter 'quite' set in the dsm.sys file?

If we are using the TSM Scheduler.
From the client\ba\bin issue: dsmc sched
Do we see schedule information or an error message?

From the message in the dsmerror.log:

ANS1999E Incremental processing of '/mountpoint'' stopped.

Are we able to manually backup /mountpoint' ?
In the original posting there was a mention that we were able to perform a manual backup. If we are able to perform a manual backup.
Is this a new schedule backup?
Post the complete schedule.

ANS1017E Session rejected: TCP/IP connection failure
From this message, for some reason the TSM Client can not connect with the TSM Server.
Possible network issue or password issue.

From the actlog:

ANR0482W Session 2518 for node SERVER (SUN SOLARIS) terminated - idle for more than 15 minutes. (SESSION: 2518)

For some reason the TSM Server did not receive any data, and it termiated the session.

In the dsm.sys file, update the schedmode opposit to its current setting.

If the schedmode is set to polling update this to prompted
and visa versa.

If we need to set the schemode to prompted, also need to set the following in the dsm.sys file also.

TCPCLIENTADDRESS ip.address.of.client
TCPCLIENTPORT 1500 <- We'll use the same port as the TCPPORT.

In the past I was able to resolve strange schedule backup issue just by changing the schedmode opposit to its current setting.

Once we have the dsm.sys file update, stop/restart the schedule daemon so that the new updates are read into memory.

Good Luck,
Sias