From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ian Kent Subject: Re: clients suddenly start hanging (was: (no subject)) Date: Thu, 12 Jun 2008 11:20:29 +0800 Message-ID: <1213240830.4994.82.camel@raven.themaw.net> References: <20080423185018.122C53C3B1@xena.cft.ca.us> <1211510100.15090.27.camel@raven.themaw.net> <20080526003442.F3E27211250@simba.math.ucla.edu> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20080526003442.F3E27211250@simba.math.ucla.edu> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: autofs-bounces@linux.kernel.org Errors-To: autofs-bounces@linux.kernel.org To: Jim Carter Cc: autofs@linux.kernel.org On Sun, 2008-05-25 at 17:34 -0700, Jim Carter wrote: > I applied autofs-5.0.3-expire-thread-create-cond-handling.patch. There > is still hanging, but I have the impression that it may be a bit less > than before: 4 hangs in 3+ hours, having mounted 247 filesystems. Below > I've included backtraces from all 4 hangs (oink, oink). > > There was one confusing incident in this test run: after one of the > filesystems was automounted, the host machine crashed (or more likely, > the owner pulled the Ethernet cable so he could plug in his laptop). > This caused every access to that machine to be very slow, which the test > program interpreted as hung automount processes. I snipped those out. > Likely you won't see any evidence -- I don't see the string > /net/rangers/m1 in any argument list -- but if you notice a thread > working on that filesystem, ignore it. > > (Re. Heads Up: I'm using the SuSE copy of the source, not GIT. The > patch went on with no complaints.) Hi Jim, I've been working on this. I believe there is still a problem somewhere but I'm going to send over another test snapshot for testing. Ian