Re: [Networker] Client backup failure wiht RPC error

This is almost certainly due to the upgrade to 6.1.3 which has a bug where
the "rsh" function doesn't work.  This only affects "old-style" clients
which don't use the nsrexec functionality.

Thanks to Daniel Bass:
In Legato knowledge base these issues are mentioned as  LGTpa49915 /
LGTpa50173.
The fix contains patched version of nsrexec. Contact Legato Technical
support to get the fix.

On Fri, 9 May 2003 11:32:51 -0400, Matthew Temple
<mht AT RESEARCH.DFCI.HARVARD DOT EDU> wrote:

>We have a problem with an IRIX client which we've been backing up
>for almost 2 years.   There have been no OS changes.   We're running
>server version 6.1.3 on a RedHat Linux computer (7.3)
>
>The client has 2 ethernet cards, which I suspect is part of the
>problem, but I'm not sure.
>
>Until last week, server initiated backups always worked without
>a hitch.  We did once have a problem which is relavent to this,
>I think, and that is that we weren't able to do a recover initiated
>from the client.
>
>So, call the client zzz1.   It has a second ethernet card and there
>appears as zzz2.   Each has an address in our normal network address
>space.
>
>Right now, if I try to recover from the client, I get the following
>error:
>==============
>recover: Cannot start session with server tapes.dfci.harvard.edu: Unable
>to receive
>==============
>
>If I tell the host machine that the client has the address assigned to
>zzz2,  I get the following error message:
>==============
>recover: Cannot start session with server tapes.dfci.harvard.edu:
>Authentication error
>=====================================================================
> From the server, I get the following error message when backing up:
>
>* zzz1:All 2 retries attempted
>* zzz1:All nsrexec: authtype rsh
>* zzz1:All savefs: RPC error: Unable to receive
>* zzz1:All savefs: cannot retrieve client resources
>   savefs zzz1: failed.
>
>Our server isn't running any firewall, and, to my knowledge,
>neither is the client.   We moved our office at approximately
>the same time this started, we also upgraded from 6.1.1  (I
>think) to 6.1.3.
>
>During part of this time, we saw an ARP problem where it looked liker
>another machine had tried to take this client's IP address, but that
>has been resolved.
>
>=============================================================
>I've got another strange phenomenon.   What is this telling me?
>
>zzz1.dfci.harvard.edu# /usr/etc/nsrports
>nsrports: RAP error: Service not available.
>
>
>==============================================================
>Any help or pointer suggesting where to start would be deeply
>appreciated.
>
>                                        Matt Temple

--
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.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=