Networker UNIX/Solaris recover problem - RPC error

Question

Hi - any help/input appreciated

trying to recover to a solaris host from another solaris host.

network connectivity is fine - server/storage node/destination host/source host can all see each other - pingable, DNS entries OK.

Host can backup OK but can t get this damn recover to work.

Nick

recover: Total estimated disk space needed for recover is 4889 MB.
Recovering 60113 files within /ctron_grp/ into /ctron_grp/ctronbel
Requesting 60113 file(s), this may take a while...
recover: recover failed: recover: RPC error: RPC receive operation failed. A network connection could not be established with the host.
Recover completion time: Mon Sep 4 11:38:38 2006

dboucha · Answer

Hito leave the Network server interface field empty in the two hosts,that normally will regulate the probléme.djamal

ble1 · Answer

It could be a blind shot, but I would try to get those clients to the same version (7.2.2). If it fails then, try fresh backup on aigunx199 and then try that to restore on aigunx10 (this is to rule out any version gothas). I assume aigunx10 is defined as client too, right?

nicbone · Answer

djamal - thanks but we never make any entry in that field so this is always blank.

nicbone · Answer

root : aigunx199 [/opt/nsr/res] #more nsrla.res
administrator: root, "user=root,host=aigunx199";
disable directed recover: No;

This host is 7.2 build 172

The other host - aigunx10 - has no mention of directed recover in its nsrla.res file
aigunx10 is 6.1.1 client

ble1 · Answer

Even error doesn't indicate that did you check remote access permissions? Also check nsrla.res if directed restores are enabled or not.

nicbone · Answer

Even error doesn't indicate that did you check remote
access permissions? Also check nsrla.res if directed
restores are enabled or not.

Yes - I have ensured that the requesting client aigunx199 has following entry in remote access field of the source client aigunx10.

Have never had this issue before & set up over 50 unix & Windows clients over the past year.

ble1 · Answer

Have you checked their nsrla.res that this is enabled indeed?

ble1 · Answer

Oh, I guess support will wish to verify network connection so have something ready to support your claim that everything is ok. ping is not enough. Have ready output of scp test (in both directions) between your target host and storage node/ backup server.

nicbone · Answer

Yeah - I would prefer if both were same rev but aigunx10 is Solaris 2.6 & is not certified with Legato 7.2. It probably would be OK but unless I am forced to dont want to risk it.

Interesting thing is the save from aigunx199 is recoverable on aigunx10.

Yes - aigunx10 is a defined legato client in the same Networker datazone.

Nick

ble1 · Answer

Then open a case with support - they could give you the same suggestion which might force you then (official justification). I'm not sure if this version mismatch might be the reason or not, but it looks as something I would test first. Also, by sending resource db to support they might check your configuration - there might be something overlooked.

nicbone · Answer

did a recover from a solaris 2.6 legato 6.1.4 client OK so thinking of upgrading aigunx10 to 6.1.4. Only hassle is removing the Sybase module which is a tricky to reinstall on our sybase installations.

NetWorker

Networker UNIX/Solaris recover problem - RPC error

Was this post helpful?