AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:
CFG:
PT:0
S:andy.sharp@lsi.com
RQ:
SSV:mhbs.lsil.com
NSV:
SSH:
R:<Shun.Yagishita@lsi.com>
MAID:2
X-Sylpheed-Privacy-System:
X-Sylpheed-Sign:0
SCF:#mh/Mailbox/sent
RMID:#imap/LSI/INBOX	0	98A4690930008C419FBEB2F0407A5C09DDDC2A18@sikmail02.lsi.com
X-Sylpheed-End-Special-Headers: 1
Date: Fri, 21 Aug 2009 11:27:37 -0700
From: Andrew Sharp <andy.sharp@lsi.com>
To: "Yagishita, Shun" <Shun.Yagishita@lsi.com>
Subject: Re: NFS lockf() or fcntl()
Message-ID: <20090821112737.19828cf2@ripper.onstor.net>
In-Reply-To: <98A4690930008C419FBEB2F0407A5C09DDDC2A18@sikmail02.lsi.com>
References: <98A4690930008C419FBEB2F0407A5C09DDD3B5AC@sikmail02.lsi.com>
	<20090820101403.1afbcebb@ripper.onstor.net>
	<98A4690930008C419FBEB2F0407A5C09DDDC2A18@sikmail02.lsi.com>
Organization: LSI
X-Mailer: Sylpheed-Claws 2.6.0 (GTK+ 2.8.20; x86_64-pc-linux-gnu)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

Ah, if the client is Redhat, this could be a Redhat bug.  Certain
versions of Redhat have a bug where the kernel silently drops the mount
when using automount.  We have a build server here that suffers from
that problem, so we changed it to static (fstab) mounts.  I have never
seen this problem with other distros, only certain versions of Redhat.

If it's not Redhat, then possibly the problem can be tracked down by
running tcpdump -W -C <n> on the client, which will allow it to run
continuously until the problem happens, and then have a trace of the
network at the time, without using too much disk space.

Cheers,

a


On Fri, 21 Aug 2009 02:07:39 -0600 "Yagishita, Shun"
<Shun.Yagishita@lsi.com> wrote:

> Andy san,
> 
> It hungs completely until rebooting of DEEPMail and server OS(RedHat).
> The vsvr doesn't failover.
> 
> They used the automount, but they try fstab from today.
> 
> Thanks.
> Yagi
>     
> 
> -----Original Message-----
> From: Andrew Sharp [mailto:andy.sharp@lsi.com] 
> Sent: Friday, August 21, 2009 2:14 AM
> To: Yagishita, Shun
> Subject: Re: NFS lockf() or fcntl()
> 
> On Wed, 19 Aug 2009 20:15:55 -0600 "Yagishita, Shun"
> <Shun.Yagishita@lsi.com> wrote:
> 
> > IHAC that using the DEEPMail(DEEPSoft). They have two DEEPMail
> > servers and access to Cougar(4.0.2.3) via NFS v3. We are facing NFS
> > hung sometimes. Does Cougar support NFS lockf or fcntl? Two
> > DEEPMail servers use one shared DB on Cougar via NFS.
> > 
> > Thanks.
> > Yagi
> 
> Hi Yagi-san,
> 
> Can you characterize the hang the customer is experiencing?  Does it
> hang for a minute or two and then continue, or does it hang
> completely and the vsvr fails over, or something else?  I'm trying to
> see if this matches a known bug.
> 
> Thanks,
> 
> a
