"NSR Server Busy" Problem.... 1

[pdaemon]

Hi,

We are running DG/UX Unix...Using Legato Networking as our backup software. Trying to recover a directory from one of our previous backups through nwrecover I get the following error.

NSR Server 'servername' busy
mwrecover: waiting 30 seconds then retrying.

I have tried stoping the Networker daemons by using nsr_stop and removing any temporary files out of the /nsr/tmp directory but this hasn't helped.

Does anyone have any suggestions?

Thanks for the help.

 

[ag100]

I've seen this happen a lot when there are other pending tape mounts on the system. I would suggest ensuring the correct tape is mounted in a drive prior to starting the recover. Let me know how you make out..

Good Luck!!!

 

[pdaemon]

Thanks for the reply...

Before I tried to recover the directory, I used nwrecover to check for different versions of the file and on which tape the file was on. I then subsequently mounted this tape through nwadmin. I changed time to recover the correct version and marked the directory for recovery. The strange thing was that it said some of the files were on tape 394...which was currently mounted and other files on tape 391. Could this be part of the problem??

Thanks...

 

[ag100]

Possibly... You should always ensure that NetWorker asks for the tapes you expect it to.. If it's asking for something else, you need to look into why.

What tape does it ask for first? If it wants 391 and it has 394, you would most definitely see that error. If you're not restoring from the full, it will want to use the full first, then lay down the changed data.

In any case, it sounds like you're on the right track. One last thing, when you launch NetWorker, make sure you aren't marking anything for recover before changing the browse time and selecting the files you need.

Keep me updated, and I'll help if I can.

Good luck!!!

 

[pdaemon]

The problem was that one of the files on the directory I wanted to restore was on tape 391...but the system couldn't find it...the other files in the directory on tape 394 were sucessfully restored. Now I just have to get tape 391 out of our safe!! heheh

Thanks you very much for you help..
Greatly appeciated.

 

[x2shy]

I saw this problem as well when Legato wasn't writing the indexs of the storage node to a tape. It would sit with that darn server busy message. Drove me insane until I worked out what was wrong. Just for the record. If you have a storage node that is in an autochanger you are running separate pools for the storage node. You will need ot ensure that one of the device connected to the backup server (ie not the SN) are in the allowed devices.
That is what I found anyway.