On Wed, 2008-06-25 at 11:33 +0800, Ian Kent wrote:
> On Tue, 2008-06-24 at 16:39 -0700, Jim Carter wrote:
> > On Tue, 24 Jun 2008, Ian Kent wrote:
> > 
> > > > > There is a pattern of failure that may not be automount's fault.  On 
> > > > > almost exactly 0.1% of the attempted mounts, the readdir eventually 
> > > > > fails 
> > > > > with ENOENT....
> > > 
> > > This could be related to the expire issue I saw in my tests so we need
> > > to test my slightly updated patch and see before we go into this.
> > > Hopefully I haven't regressed from the last test.
> > > 
> > > So here is my updated revision 8 of the patch against the SuSE package
> > > as usual.
> > 
> > OK, I ran it for 4 hrs 19 mins, doing 5900 mount-expire cycles.  No threads 
> > hung.  There were 5 ENOENT events -- the rate was not statistically 
> > different from the previous long test.  
> 
> Would it be to difficult to get a debug log for this.
> I need to find out whether this is happening in the kernel or in the
> daemon.

And if possible an indication of when a couple of these events occurred
would be really useful, so I can locate them in the log.

Ian

_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs

Reply via email to