1.2.9 unstable ? (was [OpenAFS] inaccessibble volume - please help)

[Derrick J Brashear]

On Thu, 8 May 2003, Rudolph T Maceyko wrote:

> --On Thursday, May 08, 2003 09:08:57 +0200 jarausch@igpm.rwth-aachen.de
> wrote:
>
> > Has anybody experienced similar problems with 1.2.9 ?
>
> I haven't quite tracked this down yet, but we were seeing shutdown
> hangs in afsd when processes were still accessing /afs at shutdown time.
>
> This was under Red Hat 7.3, OpenAFS RPMs built for Red Hat's
> 2.4.18-27.7.x kernel.

Can you still cmdebug the cache manager at that point? If so, does it shut
down at that point?

It's conceivable something changed about the killing routines
we're using to shut down the rx listener. The idea is that we need to wake
it up so it can die. Might be smarter to switch to what kolya did for
Solaris (I think) which was to fake a packet to the listener so it would
notice we were shutting down, and do so.