[OpenAFS] 1.3.80 server strangeness (kernel 2.6.11-gentoo-r3)

[Kevin]

On Fri, 2005-03-25 at 14:37 -0500, Derrick J Brashear wrote:
> On Fri, 25 Mar 2005, Kevin wrote:

> > What are the implications of being in the state where all of the server
> > instances are running but the bosserver is not?  Is that bad news?
> > Should I kill -9 the other servers ASAP (since I seem unable to do so
> > with bos stop from any server or client) or is continuing to run for a
> > while ok?  I've noticed that if I run bosserver again (without killing
> > anything), I get two copies of all these server instances.  Is that bad?
> 
> kill -9 is bad. a clean shutdown is better. they can run until you want to 
> start a new bosserver, but don't kill -9. don't try to outclever them, you 
> will be sad.
> 

I'll heed that advice.  Thanks for pointing it out.

Aside from what I've noted in this thread (and one other thing which
I'll get to shortly), 1.3.80 server-mode only seems to be working fine
for me with a gentoo 2.6.11-r3 kernel.  To my knowledge, it's handled
several copy operations to the volumes it's hosting without problems.
But without explanation this morning at 0400, all of the server
instances on this machine just shut down.  The bosserver was still up,
but log entries in /usr/afs/logs showed what seemed to be a normal
termination for all the others; only I didn't order it.  I'm 100%
certain that foul play is not a factor because the network hosting the
the cell is (for the time being), not even accessible from without.  I
tried a:
# bos status server -long -localauth
bos: failed to contact host's bosserver (communications failure (-1)).

but no explanation there.

Similarly, nothing in /var/log/messages.

Any ideas on where else to look for a reason for the shutdown?

Also, I've recently installed valgrind.  Should I try to use it to
isolate the cause for the glibc error noted above in this thread?
Should afs be rebuilt with debugging information first?

TIA.


-- 
-Kevin
http://www.gnosys.us