LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Linux - Enterprise (http://www.linuxquestions.org/questions/linux-enterprise-47/)
-   -   dsm hung after backup of a file failed (http://www.linuxquestions.org/questions/linux-enterprise-47/dsm-hung-after-backup-of-a-file-failed-4175441403/)

Rohit_4739 12-14-2012 01:20 AM

dsm hung after backup of a file failed
 
Hi Guys,

I need urgent hint or resolution on the ahead mentioned issue. We have are having an issue related to Backup on one of the client server runnig rhel 5.6 where the Scheduled Backup to TSM server missed. I was trying to look at the next schedule for backup using "dsmc q sched", however the command just hanged and didn't return any output.

Looking into the dsmerror logs i saw the below information.

Code:

ANS1228E Sending of object '/test/1_log' failed
12/10/12  17:30:07 ANS4037E Object '/test/1_log' changed during processing.  Object skipped.
12/10/12  17:30:13 ANS1802E Incremental backup of '/test' finished with 1 failure

I also tried to kill/restart the dsmc but again that is also not happening. The client is able to ping and telnet on the TSM server successfully, so the issue is on client. I have looked for error codes on the http://publib.boulder.ibm.com but having little hard time to resolve. I

So can anyone please suggest anything that might help me in troubleshooting the issue further and then may be resolving it since its bit urgent; so may be if somebody had encountered.

TB0ne 12-16-2012 09:36 AM

Quote:

Originally Posted by Rohit_4739 (Post 4849065)
Hi Guys,
I need urgent hint or resolution on the ahead mentioned issue. We have are having an issue related to Backup on one of the client server runnig rhel 5.6 where the Scheduled Backup to TSM server missed. I was trying to look at the next schedule for backup using "dsmc q sched", however the command just hanged and didn't return any output.

Looking into the dsmerror logs i saw the below information.
Code:

ANS1228E Sending of object '/test/1_log' failed
12/10/12  17:30:07 ANS4037E Object '/test/1_log' changed during processing.  Object skipped.
12/10/12  17:30:13 ANS1802E Incremental backup of '/test' finished with 1 failure

I also tried to kill/restart the dsmc but again that is also not happening. The client is able to ping and telnet on the TSM server successfully, so the issue is on client. I have looked for error codes on the http://publib.boulder.ibm.com but having little hard time to resolve. I

So can anyone please suggest anything that might help me in troubleshooting the issue further and then may be resolving it since its bit urgent; so may be if somebody had encountered.

If this is urgent, and you're using IBM Tivoli Storage Manager, on a RHEL server, you are in a supported configuration. You are PAYING for support from both IBM and RHEL...contact their tech support, since they're there 24/7 to help you with 'urgent' problems.

You say you tried to kill/restart dsmc, but tell us nothing besides "also not happening"....what, exactly, are you doing, and what results do you get, when? And are you the TSM administrator? This is seen ALL the time, and the error is fairly self-explanatory...it's telling you the file /test/1_log has CHANGED...that is, more data got written to it/deleted from it/SOME operation was performed on it, while TSM was backing it up. The AN codes on the TSM site tell you this too, and give you resolutions. You can either ignore it, or stop whatever process that's accessing that file before you backup...see the PRE and POST commands for TSM.
http://publib.boulder.ibm.com/infoce...2FANS1228E.htm
http://publib.boulder.ibm.com/infoce...2FANS1228E.htm
http://publib.boulder.ibm.com/infoce...2FANS1228E.htm

Rohit_4739 12-16-2012 11:56 PM

Quote:

Originally Posted by TB0ne (Post 4850401)
If this is urgent, and you're using IBM Tivoli Storage Manager, on a RHEL server, you are in a supported configuration. You are PAYING for support from both IBM and RHEL...contact their tech support, since they're there 24/7 to help you with 'urgent' problems.

You say you tried to kill/restart dsmc, but tell us nothing besides "also not happening"....what, exactly, are you doing, and what results do you get, when? And are you the TSM administrator? This is seen ALL the time, and the error is fairly self-explanatory...it's telling you the file /test/1_log has CHANGED...that is, more data got written to it/deleted from it/SOME operation was performed on it, while TSM was backing it up. The AN codes on the TSM site tell you this too, and give you resolutions. You can either ignore it, or stop whatever process that's accessing that file before you backup...see the PRE and POST commands for TSM.
http://publib.boulder.ibm.com/infoce...2FANS1228E.htm
http://publib.boulder.ibm.com/infoce...2FANS1228E.htm
http://publib.boulder.ibm.com/infoce...2FANS1228E.htm

Hi TB0ne , I did read the AN codes for the error from TSM site and understand what the error meant. So what i did was i excluded that file and restarted dsmc and it was successful but the problem is that when i run "dsmc q sched" to view the schedule or "dsmc q backup" to view backup listing from TSM it just hangs, in short running "dsmc" just hangs. And i am not a TSM administrator, i am a Linux admin. I

TB0ne 12-17-2012 07:22 AM

Quote:

Originally Posted by Rohit_4739 (Post 4850707)
Hi TB0ne , I did read the AN codes for the error from TSM site and understand what the error meant.

Ok...if you knew what they meant and how to resolve them, why open a thread?
Quote:

So what i did was i excluded that file and restarted dsmc and it was successful but the problem is that when i run "dsmc q sched" to view the schedule or "dsmc q backup" to view backup listing from TSM it just hangs, in short running "dsmc" just hangs. And i am not a TSM administrator, i am a Linux admin. I
Just run "dsmc", then from the "dsmc>" prompt, enter "q sched". Unless you put the correct command-line out there, it won't return.

Also, excluding that file is not a good thing to do...since now, you won't have ANY backups of that file, ever. If a user/process has it open and its being used, that would indicate that it's of some value. Unless you can definitively identify it as a low-value log file, you need to address the underlying cause.

And again....contact IBM/Tivoli support if you're having problems. You're paying for support from them.


All times are GMT -5. The time now is 11:57 PM.