Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!orion.oac.uci.edu!sun.acs.uci.edu!cbell
From: cbell@sun.acs.uci.edu (Chris Bell)
Newsgroups: comp.protocols.appletalk
Subject: Problem with aufs timeouts
Summary: tickle timeouts
Keywords: aufs timeout tickle CAP
Message-ID: <27F8D0DB.20481@orion.oac.uci.edu>
Date: 2 Apr 91 18:43:39 GMT
Organization: University of California, Irvine
Lines: 15
Nntp-Posting-Host: hydra.acs.uci.edu

We are using the Rutgers CAP, on a Sun Sparcstation 2 under 4.1.1, and
having a problem with aufs.  About once a day, all of the open
sessions will suddenly drop.  The aufs log is excerpted below.  Is
there a timeout define that needs to be relaxed? How can we tell where
the problem is, because the log file says "probable a tickle timeout".
We are using the enet filter, and the macs are using ethertalk.  We
also have fastpaths, and a somewhat busy ethernet.  Has anyone else
seen this problem?

Thanks,

Chris Bell
Distributed Computing Support  University of California, Irvine
cbell@uci.edu        cbell@uci.bitnet       uunet!uci.edu!cbell
===============================================================