postcmd doesn't start after upgrade

[lg1234]

Hi,


I need help to find out why postcmd was not execute. We are now at version 7.3 on backup server, but still 7.1 on every client.

I did a simple test, and only precmd was executed

type: savepnpc;
precmd: "echo hello";
pstcmd: "echo bye";
#timeout: "12:00:00";
abort precmd with group: Yes;

Thanks.

 

[haasemi]

I think, it´s a problem of the client version.

At 7.3 something with job id´s etc. changed, so that the communication for the pre/post command execution don´t work correctly.
Try this again after upgrading a client to 7.3.x

I hope, you are using 7.3.2 for your server. 7.3 versions lower than 7.3.2 had to many bugs...

 

[lg1234]

We are now 7.3.2 on server and 7.3 on a client
and we retried the same script and we got the same result..

Any suggestion is appreciate..

thanks.

 

[lg1234]

Hi Haasemi,

The pstcmd problem seems due to client version different from server. When a client is upgraded to 7.3.2, pstcmd seems to perform correctly.

Thank you for help, it was very useful!

 

[lg1234]

Do someone ever seen this message, It's appeared after installation of EBS 7.3.2 client ?

And what does it mean ?

nsrexecd: WARNING: NSR configuration database found invalid resource type
Please check your configuration!

 

[605]

Since NW 7.0 defective resource files are detected and isolated in a separate directory. Is it possible that - at this point - this applies to your groupname.res file?

 

[lg1234]

I don't understand your question, could you give more explanation please ?

 

[605]

Look whether there is a ..\nsrdb\dbg directory. This is where defective resources will be isolated.

 

[lg1234]

There is no nsrdb/dbg directory on server and client..

On server, we have this :

/opt/nsr/res/nsrdb
%ls -l
total 26
drwx------ 2 root other 1024 Oct 25 21:35 00
drwx------ 2 root other 1024 Oct 26 09:17 01
drwx------ 2 root other 2048 Oct 26 05:42 02
drwx------ 2 root other 2048 Oct 26 10:09 03
drwx------ 2 root other 1536 Oct 26 10:56 04
drwx------ 2 root other 1024 Oct 26 11:02 05
drwx------ 2 root other 1024 Oct 26 11:02 06
drwx------ 2 root other 1024 Oct 26 11:02 07
drwx------ 2 root other 1024 Oct 26 09:17 08
drwx------ 2 root other 1024 Oct 26 09:17 09

We performed a backup last night, there was 10 clients doing backup, 30 min after, we had 3 backups-client done, but the pstcmd didn't execute, it waits until the completion of the group.
And I have all these following messages and I don't know what does it mean ...
10/25/06 09:00:15 PM preclntsave: Starting up the precmds.
10/25/06 09:00:31 PM preclntsave: All command(s) ran successfully.
10/25/06 09:01:31 PM pstclntsave: Client is not active in the worklist.
10/25/06 09:01:31 PM pstclntsave: Client is not active in the worklist.
10/25/06 09:01:31 PM pstclntsave: Client is not active in the worklist.
10/25/06 09:01:31 PM pstclntsave: Client is not active in the worklist.
10/25/06 09:01:31 PM pstclntsave: Client is not active in the worklist.
10/25/06 09:01:31 PM pstclntsave: Client is not active in the worklist.
..... many message per seconds..

10/25/06 09:01:31 PM pstclntsave: Client is not active in the worklist.
10/25/06 09:01:31 PM pstclntsave: Client is not active in the worklist.
10/25/06 09:01:31 PM pstclntsave: Client is not active in the worklist.

0/25/06 09:01:32 PM pstclntsave: Client is still active in the worklist.
10/25/06 09:01:32 PM pstclntsave: Worklist not complete. Some saves are still running.
10/25/06 09:02:32 PM pstclntsave: Client is not active in the worklist.
10/25/06 09:02:32 PM pstclntsave: Client is not active in the worklist.
.... many output
10/25/06 09:02:32 PM pstclntsave: Client is still active in the worklist.
10/25/06 09:02:32 PM pstclntsave: Worklist not complete. Some saves are still running.
...many output...
10/25/06 09:04:33 PM pstclntsave: Client is still active in the worklist.
10/25/06 09:04:33 PM pstclntsave: Worklist not complete. Some saves are still running.
... many output...
10/25/06 09:08:37 PM pstclntsave: Client is still active in the worklist.
10/25/06 09:08:37 PM pstclntsave: Worklist not complete. Some saves are still running.
0/26/06 12:26:51 AM pstclntsave: Client is not active in the worklist.
10/26/06 12:26:51 AM pstclntsave: Client is not active in the worklist.
10/26/06 12:27:51 AM pstclntsave: Savegroup is not running.
10/26/06 12:27:51 AM pstclntsave: All savesets on the worklist are done.
10/26/06 12:27:51 AM pstclntsave: Starting up the pstcmds.
10/26/06 12:28:20 AM pstclntsave: All command(s) ran successfully.
10/26/06 12:28:20 AM pstclntsave: Exited.

In the previos version, pstcmd is exectuted after the backup of saveset's client, it is normal thatpstcmd in 7.3.2 is doing after the completion of the group ???? it's make no sense ! please help.

 

[Bidoulaloge]

Hi

I am also interested in this problem and I have not yet found any solution. Does anybody in the forum have encountered this situation???

Sun support and presumably, Legato's are very slow on this issue

Thanks

 

[605]

Do not forget the temporary files
/nsr/tmp/groupname.lck
/nsr/tmp/groupname.tmp

For more details please see the manpage for "preclntsave" & pstclntsave.

 

[Bidoulaloge]

I did the testing again, after deleting all the .lck and the .tmp files on the client side and the behavior of the pstclntsave is the same. It completes only after the whole group is finished, not only the client.

From Legato, I received the following answer:

Legato has reproduced the issue using a windows based system. The only workaround they can offer, for now, is to use the timeout variable in the .res file or place the clients in different groups.

I told them that neither is very convenient.