Backup of Networker Client Fails after IP change

Usually it is enough to open just one thread to get answer to your question

Have you checked local hosts (on clients) and DNS entries?

I had this problem too one time. Networker was the only thing that couldn't see the client anymore, kept getting failed probe job messages if I recall. I never could fix it. Had to totally rebuild the clients from scratch.

OK, let's play then - pick one client which is causing a trouble. Go to backup server and execute following command:

echo print | nsradmin -p 390113 -i - -s

and post output here. Also, post daemon.log part showing failure for that client from both backup server and client.

Yes, I did. The backup server's IP did not change, so I did not need to change the host files on the clients. nslookup from both sides return the right IP address (new IP)

There is no need for rebuild, trust me...

Ok, here is the output as requested:

echo print | nsradmin -p 390113 -i - -s myclient.mydomain.co.za
nsradmin: RPC error: Remote system error
nsradmin: There does not appear to be a NetWorker nsrexecd server running on myclient.mydomain.co.za.


daemon.log:

05/09/07 13:41:22 nsrd: savegroup info: starting Test (with 1 client(s))
05/09/07 14:06:34 savegrp: command 'savefs -s mybackup.mydomain.co.za -c myclient.mydomain.co.za -g Test -p -l full -R -v ' for client myclient.mydomain.co.za exited with return code 1.
* myclient.mydomain.co.za:All Network connection could not be established with client 'myclient.mydomain.co.za'
05/09/07 14:06:34 savegrp: myclient.mydomain.co.za:probe will retry 1 more time(s)
05/09/07 14:31:46 savegrp: command 'savefs -s mybackup.mydomain.co.za -c myclient.mydomain.co.za -g Test -p -l full -R -v ' for client myclient.mydomain.co.za exited with return code 1.
* myclient.mydomain.co.za:All 1 retry attempted
* myclient.mydomain.co.za:All Network connection could not be established with client 'myclient.mydomain.co.za'
05/09/07 14:31:46 savegrp: myclient.mydomain.co.za:probe will retry 0 more time(s)
05/09/07 14:31:47 nsrd: mybackup.mydomain.co.za:index:myclient.mydomain.co.za saving to pool 'DailyCUT' (000061L2)
05/09/07 14:31:50 nsrd: mybackup.mydomain.co.za:index:myclient.mydomain.co.za done saving to pool 'DailyCUT' (000061L2)
05/09/07 14:31:50 nsrd: savegroup alert: Test completed, total 1 client(s), 0 Hostname(s) Unresolved, 1 Failed, 0 Succeeded. (myclient.mydomain.co.za Failed)
05/09/07 14:31:50 nsrd: runq: NSR group Test exited with return code 1.
05/09/07 14:32:32 nsrd: write completion notice: Writing to volume 000061L2 complete

There you go, you can't even establish duplex RPC connection with client. Now either:
a) RPC traffic is blocked
b) IP used is wrong
c) NetWorker on client is not running
d) client can't communicate back to server due to wrong routing (this could be your issue)

If the clients were moved to another network, you have to change the server network interface in each client ressource.

Hrvoje,

Spot on. Our network guys finally admits that one of our firewalls seems to be the culprit (seems like the firewall's database is corrupt), after I asked them to check for drops for the fourth time!

Thanks for all your trouble.

Lennel