Legato RAP errors after crash/TCP/IP # updates

[JasonPartridge]

Hey all,
We had a little problem with our A5000 Photon Sun RAID since then, our indexes are whacked! We have attempted to restore the indexes with mmrecover, but now we are getting datastream errors. When we try to run jbconfig, we get RAP errors. We have a service contract and a priority 1 incident with Legato, but we all know how that works.

Any help or ideas would be greatly appreciated.

TIA

Jason Partridge
PC/LAN Manager
Graphic Impressions, Inc
812-334-2323 X6014

 

[ag100]

Hey...

Ok, so I'm assuming all SCSI problems are resolved? Right? Did you have to blow away /dev/rmt? What RAP errors do you get when you try the jbconfig?

Assuming they are, you performed an mmrecov to recover the media database.. I assume you re-installed NetWorker? Are the client indexes still intact? What version of NetWorker are you using? I ask, because, nsrck is not a very reliable means of checking indexes in versions prior to 6. Have you reinstalled the same version of NetWorker you were using??

Let me know, and I'll help if I can...

-ag100

 

[JasonPartridge]

OK, Here's the deal, our Photon went down Hard.... Sun came and "fixed" it... basically they made it worse... it went down again, and took the kernal with it. Sun came back, and insisted on patching the OS to the current level for 2.6. These patches inclueded some SCSI patches. During this mess, the server did not cleanly unmount /apps before it re-booted (/apps is where Legato resides) After the dust settled on the Photon mess, now our Qualstar TLS-4480 blows rap errors if we try to run JBConfig. if we try to mount a tape it gives us an invalid tape argument. Our indexes as far as we (we meaning our IT manager, me, and Legato) can tell the indexes are OK, although none of us want to bet money on it. It just seems that Legato doesn't want to talk to the Jukebox.

Help!

 

[ag100]

Ok.. You're getting a bunch of weird things happening.. Try doing a pkgrm on all the Legato packages installed. When all packages are uninstalled, reinstall the same version of NetWorker. I've never seen an invalid argument when attempting to mount a tape.. What command are you using (nsrjb -m -f /dev/rmt/xxx?)

Also,

What NetWorker processes are running now (ps -ef | grep nsr)? What version of NetWorker are you using?

 

[JasonPartridge]

yep, we are definetly getting a bunch of weird things happening! I'm not sure if an un-install/re-install will be viable, but hell at this point we may have to!
We are mounting the tape via the GUI (nwadmin &)
These are the processes that are running:
root 4721 4632 0 11:40:05 ? 0:00 /usr/sbin/nsrmmd -n 2
root 4727 4632 0 11:40:07 ? 0:00 /usr/sbin/nsrmmd -n 3
root 5498 4632 0 11:42:34 ? 0:00 /usr/sbin/ansrd gi-dam.graphicimpressions-usa.com:root recovering
root 14003 4706 0 12:11:50 ? 0:00 /usr/sbin/nsrindexd -a ADD gi-dam.graphicimpressions-usa.com index for backup o
root 22448 22258 0 12:35:45 pts/9 0:00 grep nsr
root 4731 4632 0 11:40:09 ? 0:00 /usr/sbin/nsrmmd -n 4
root 4632 1 0 11:39:56 ? 0:04 /usr/sbin/nsrd
root 4610 1 0 11:39:52 ? 0:01 /usr/sbin/nsrexecd -s gi-dam.graphicimpressions-usa.com
root 4718 4632 0 11:40:03 ? 0:00 /usr/sbin/nsrmmd -n 1
root 4706 4632 0 11:40:00 ? 0:00 /usr/sbin/nsrindexd
root 5500 4632 0 11:42:35 ? 0:00 /usr/sbin/nsrmmd -n 6
root 4615 1 0 11:39:53 ? 0:01 /usr/sbin/lgtolmd -p /nsr/lic -n 1
root 5076 4632 0 11:41:12 ? 0:00 /usr/sbin/nsrmmd -n 5
root 4660 4632 0 11:39:58 ? 0:08 /usr/sbin/nsrmmdbd
root 4614 4610 0 11:39:53 ? 0:01 /usr/sbin/nsrexecd -s gi-dam.graphicimpressions-usa.com

We are running 6.0.1 Build 174

 

[ag100]

Ok, the first thing I would try is uninstalling, then reinstalling.. If nothing else, definitely the LGTOnode and LGTOdrvr packages. If any of the drivers got corrupted or stepped on with the problems you had, I could see that hurting you.

Also, you may want to check Legato's website, and check to make sure your jb is supported with that OS and NetWorker version.

Let me know how things go.


Good Luck!

 

[JasonPartridge]

Yeah, I'm thinking that that may have happened... the drivers being corrupted. I know that our JB is supported under Networker and the OS, so maybe that is the problem.

I'll try and talk the IT manager into killing it and re-installing. The way it works around here, he has control over the server, but I'm in charge of all the backups!

 

[JasonPartridge]

Hey all,
Legato is back up and happy, it ended up being a solaris patch issue, one of the installed patches changed the underlying SCSI drivers, as a result of that, it wrote a default st.conf to the OS. This was not good! What was sad about this is that we have a service contract with both Legato and Sun. Legato took almost a week, and could not resolve the issue, our IT manager called Sun out of desperation, 5 minutes later the appropriate changes were re-applied, and after a re-boot, NetWorker took right off! Hmmm, why do we all pay Legato's insanly expensice servic/support contract prices? Oh yeah for all those updates to the software, that they would otherwise charge for! I can't believe that Legato's support is so worthless!

Grrr,

Anyway, thanks for all the suggestions.

Jason Partridge

PC/LAN Manager