Legato on HPUX 11.0

[Richard Guiborat]

Two weeks ago Legato stopped talking to the clients. Nwadmin keeps locking up. I have three N class servers each with two DLT 8000 tape drives and am running Legato 5.5.1. Any suggestions on how to solve the problem. I have restored the Legato software back to befor the problem occurred.

[John Payne_2]

In /nsr/logs, there is a file called 'messages' and another called 'summary'. Do either of these files give you any information like errors, etc.? Mybe there is something like 'server license has expired' or something like that. We run our backups over a data lan. If you do also, maybe you have a blown NIC.

Also, generally when something like this happens, we contact Legato and one of their engineers can figure things out pretty quick. (Like if your server license expire for some reason...) If they ask for a patch, they ask for a 2+ year old patch and you have to explain that you have a superceeded patch applied already, however.

Hope it helps.

John

[Helen French]

Hi Richard:

Try these things:

1) Run 'rpcinfo -p' on Server and Client to see that 'nsr' daemons are registered there.
2) Run 'nwrecover -s server_name' to recover the server connection if lost
3) Check the logs under /nsr/logs/daemon.log and messages for error messages
4) Start and stop nsr daemons on client and server:

# nsr_shutdown
# /sbin/init.d/networker start
5) Post any error message you get.

Also 5.5.1 is little old, So I would suggest you to upgrade to 6.X or 5.5.4

HTH,
Shiju

[David Burgess]

I've had nwadmin lock up legato 6.01 when the robot in the jukebox got confused and locked up. Maybe there is a problem with a tape drive or scsi bus connection. Take a look at /nsr/logs/messages /nsr/logs/summary and /nsr/logs/daemon.log. Look for scsi lbolts in /var/adm/syslog.log

This also caused the nsrjb command to core dump.

HTH

Dave.

[Stefan Schulz]

Hi,

we had several "lockups" with our Networker 5.5.36. First it seemed that there was a network problem because the clients couldn't backup or restore. But everything pointed to a problem with our ATL P1000 Jukebox.

The workaround was always to shutdown the Jukebox then shutdown networker and restart everything again. I think that there has been some sort of timeout involved. The networker server lost the connection to the jukebox and hang up. As the server was still up (but hung) the clients couldn't do anything.

I don't know if this is the same as your problem, but perhaps you could make sure that the lost client connection is no secundary problem.

Since we updated to Networker 6.1.1 we haven't had any problems like this again. Hope this is solved now.

Hope this helps


Regards Stefan