Re: [Networker] Problem with NDMP backups

Most NAS servers have some way to run a backup locally, or at least to check
the health of the backup processes.  Can you 'log in' to your Mirapoint
server, and do something like this?  If you can remove NetWorker from the
equation, you can figure out where the problem lies.

tl

-----Original Message-----
From: Stan Sander [mailto:ssande AT SANDIA DOT GOV]
Sent: Monday, September 29, 2003 6:03 PM
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Subject: Re: [Networker] Problem with NDMP backups


Stan Horwitz wrote:

>On Mon, 29 Sep 2003, Stan Sander wrote:
>
>
>>Yes, I have seen problems with subsequent ndmp backups after one gets
>>interrupted mid-stream.  However, in my experience it only affects the
>>client that was interrupted, not others.  Do you get any messages that
>>you can share?  I'm sure you've tried this, but have you rebooted the
>>Legato server?  Have you tried labeling fresh tapes for the ndmp pool?
>>
>>
>
>Yup, we've tried rebooting our Legato server..
>
>We are attempting to test this out by running backups on our smallest
>(least data) NDMP client (which was not even being backed up at the
>time this interruption ocurred). We also rebooted the NDMP server. We
>get errors that look like:
>
>--- Unsuccessful Save Sets ---
>
>* po-e.temple.edu:/usr/store savegrp: suppressed 4 lines of output
>* po-e.temple.edu:/usr/store NDMP Service Log: imagedump: smalloc_init:
>shmget failed
>* po-e.temple.edu:/usr/store NDMP Service Log: Cannot allocate memory
>* po-e.temple.edu:/usr/store NDMP Service Log: ndmp-image: abnormal exit:
status = 6
>* po-e.temple.edu:/usr/store
>******************************************************************
>* po-e.temple.edu:/usr/store  ------      E R R O R     -------
>* po-e.temple.edu:/usr/store Data server halted: Error during the backup.
>* po-e.temple.edu:/usr/store
>******************************************************************
>* po-e.temple.edu:/usr/store
>******************************************************************
>* po-e.temple.edu:/usr/store  ------      E R R O R     -------
>* po-e.temple.edu:/usr/store Error running backup. status=3
>* po-e.temple.edu:/usr/store
>******************************************************************
>* po-e.temple.edu:/usr/store Error during NDMP backup
>
>--- Successful Save Sets ---
>
>  bootz: index:po-e.temple.edu      level=full,     13 KB 00:00:26     11
files
>
>
OK, I figured you had done that, but sometimes you just gotta ask :-).

Here is what I know about this.  The messages that contain the NDMP Service
Log:  are just lines that are logging the stderr from the ndmp client.  I
sure would like to have a look at the 4 lines that got
suppressed.   Only way I know of to see that is to run a client manually
from the command line.

Example:

savegrp -c client -G ndmp

--
Stan Sander - CSU Special Projects
Unix Systems/Server Administrator
(505)284-4915

--
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.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
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.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=