Problem with Sun machines hanging

[Colleen Hayase]

Thanks to everyone who responded to my inquiry. Bob Cook's response (below)
looks very promising since Sun's analysis of our system also pointed to a
problem with AFS. Our Solaris 2.7 machines are at AFS 3.6 so we'll have to
upgrade them ASAP. Our 2.5.1 machines are at 3.4a version 5.74 and also
display the hanging behavior. Another group I work with run version 5.55 on
their 2.5.1 machines and haven't run into the same problem.

Colleen Hayase, EDA LAN Support
IBM East Fishkill, Bldg 334 2L14-423
hayase@us.ibm.com
(t/l) 533-8976, 845-894-8976


Bob Cook <bobcook@SLAC.Stanford.EDU> on 03/07/2001 04:21:29 PM

Please respond to bobcook@SLAC.Stanford.EDU

To:   Colleen Hayase/Fishkill/IBM@IBMUS
cc:   info-afs@transarc.com
Subject:  Re: Problem with Sun machines hanging



>This is probably a Solaris problem but I thought I'd throw it out to the
>AFS community in case it  rings a bell...

It definitely rings a bell, but it is an AFS bell rather than a Solaris
bell.

We reported this to Transarc last year as their problem TR-53970.  It was
(supposedly) identified as

      * Defect 12418

       (Solaris only) This fix corrects a problem that occurred when a
       process attempted to access the proc file system.

and fixed in level afs3.6 2.5.  There was also a fix for afs3.5, but I
don't
know which level it went into, or even if it did.

The "supposedly" above is simply because it's very difficult to determine
if
it fixed the problem; we could never reliably recreate the problem.  I can
say that we have no confirmed cases of it occurring on a machine running
the
fixed client.