savefs: RPC error: Unable to send, cannot retrieve client resources

[libove]

I recently renumbered the IP addresses on my network (changes ISPs and netblocks) and everything stayed working fine except that my Networker server can no longer contact one particular of my backup clients.

The server is 6.2 Build 109 on Windows 2000 Advanced Server SP3. The clients that work are all Windows XP. The one client that no longer works (but used to work fine) is RedHat Linux v7.3 running a Linux kernel v2.4.18 and Legato client RPM v6.1.3-1.

Here's the error output from the savegroup completion messages:

--- Unsuccessful Save Sets ---
* panther4e:All 2 retries attempted
* panther4e:All nsrexec: authtype nsrexec
* panther4e:All savefs: RPC error: Unable to send
* panther4e:All savefs: cannot retrieve client resources
savefs panther4e: failed.

This happens with NO local hosts table entries on either client or server, and also happens with complete host table entries on both client and server.

Ideas please?

Thanks.
-Jay

 

[ag100]

Hi,

From the client, try typing save -s backup_server

Do you get any errors? If so, what are they?

-ag100

 

[libove]

I just get this:

NSR server cheshire4.felines.org: Unable to receive
save: waiting 30 seconds then retrying

.. until it gives up after a few tries.

I looked on the NT 4.0 Server hosting the Networker server and verified that the RPC services are running.

What else should I look for to make sure that it is running on the NSR server?

Thanks!
-Jay

 

[libove]

Ugh. The problem as demonstrated by a tcpdump, once I remembered that the IP address and therefore hostname to look for is from a different ethernet interface on that particular client, was simply that the server told the client to start the backup, the client contacted the server, the server performed a reverse DNS lookup on the client's IP address, and found that the client's name as returned by the PTR record from the reverse DNS lookup did not match any of the aliases of the client as registered with Networker.

This was because, as part of the re-IP numbering, I also incremented the number in the hostname of this system...

It would have been nice if Networker provided a more descriptive error.

Thanks for the help!
-Jay