Has anyone else encountered the following problem with Power Edition 7.1.2
on a Solaris 9 system? I did just open up a case with Legato's tech
support about this, but I am wondering if other people experienced this
problem.
This morning, I needed to made a change to a recently added client. I
deleted the client entry and modified its /etc/hosts entry on our Legato
server. The client is Windows 2000 and it is new so it has never been
backed up, and the initial attempt to back it up during our nightly
schedule failed. I wanted to change the client resource name for this
client so I deleted it and then changed this client's entry in the
/etc/hosts file on our NetWorker server. After I entered the new client
entry, nsradmin came up and told me about the client-id. I didn't care
about preserving anything because it had never been backed up before so I
just pressed the Return key to create the entry.
Unfortunately, our DNS is not all that well managed, and I have no control
over it. As a result, I find that I have to maintain a fair number of
entries in the /etc/hosts file to overcome the DNS management issues and
frequently, those who ask me to add a Windows server to our nightly backup
schedule have no idea what their server's real hostname is. This means I
typically have to try one or two times before I can get a successful
backup of a new client. This is what I attempted to do this morning with
disastrous results.
As soon as I added the new client entry via nsradmin, nsr asked me to put
in the client-id or press The Return key to continue. As soon as I pressed
Return and then pressed Escape and "Yes" to save the new entry, nsrd
crashed. All the other processes were running.
A colleague was in the process of doing a bare metal recover of around
100GB of data at this time, for testing purposes. As such, I did not want
to interrupt that recover process so I tried to restart nsrd, but that
didn't work, so I shut down the normal way and then restarted. That
worked fine. Most of the backups from the previous night had already
completed and the others restarted on their own, accept for an NDMP
backup, which I restarted manually.
What I want to know is why nsrd crashed. How could a reverse DNS error
fail for an entry in the /etc/hosts file. It looks like I found a bug in
the software. Here is the relevant section of the /nsr/logs/daemon.log
file for your review:
11/06/04 05:50:27 nsrd: Diagnostic: Reverse DNS lookup failed for hostrobin,
address 155.247.252.39
11/06/04 05:50:41 nsrck: /nsr/index/robin/.nsr is missing; re-creating
11/06/04 05:52:15 savegrp: RPC error: Unable to send
11/06/04 05:52:35 savegrp: RPC error: Unable to send
11/06/04 05:52:35 savegrp: Failed to update server, aborting Savegroup.
11/06/04 05:54:49 savegrp: RPC error: Unable to send
11/06/04 05:54:50 savegrp: RPC error: Unable to send
11/06/04 05:55:04 savegrp: RPC error: Unable to send
11/06/04 05:55:05 savegrp: RPC error: Unable to send
11/06/04 05:55:10 savegrp: RPC error: Unable to send
11/06/04 05:55:10 savegrp: Failed to update server, aborting Savegroup.
11/06/04 05:55:10 savegrp: RPC error: Unable to send
11/06/04 05:55:10 savegrp: Failed to update server, aborting Savegroup.
11/06/04 05:55:25 savegrp: RPC error: Unable to send
11/06/04 05:55:25 savegrp: Failed to update server, aborting Savegroup.
11/06/04 05:55:25 savegrp: RPC error: Unable to send
11/06/04 05:55:25 savegrp: Failed to update server, aborting Savegroup.
11/06/04 05:55:45 savegrp: RPC error: Unable to send
11/06/04 05:56:05 savegrp: RPC error: Unable to send
11/06/04 05:56:05 savegrp: Failed to update server, aborting Savegroup.
--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list. Questions regarding this list
should be sent to stan AT temple DOT edu
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
|
