From: "J. Bruce Fields" Subject: Re: Syslog flooded: "NFSD: preprocess_seqid_op: magic stateid!" Date: Wed, 16 Jun 2010 22:32:57 -0400 Message-ID: <20100617023257.GB20645@fieldses.org> References: <501A322854D4C944BCAF0528CFC7B65E3B30E7F5EE@BALZ00SW-MBX.ads.mwn.de> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Cc: "linux-nfs@vger.kernel.org" To: Joschi Brauchle Return-path: Received: from fieldses.org ([174.143.236.118]:44797 "EHLO fieldses.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759287Ab0FQCc5 (ORCPT ); Wed, 16 Jun 2010 22:32:57 -0400 In-Reply-To: <501A322854D4C944BCAF0528CFC7B65E3B30E7F5EE-7sgYbYbIMHsZ3TY+6YjGZ+/kjxG1y2if@public.gmane.org> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Wed, Jun 16, 2010 at 10:20:15AM +0200, Joschi Brauchle wrote: > Second problem: > We recently downgraded from a patched nfs-utils 1.0.9 (patched to acc= ept the =E2=80=9C-h=E2=80=9D option for different hostname in Kerberos = service creds in a high availability environment) to the original Novel= l RPM providing nfs-utils 1.0.7 (now using the undocumented =E2=80=9C-c= =E2=80=9D switch to read the accepted creds from keytab file). (Note: t= he first problem has already occurred before this downgrade) >=20 > Since this change, we are experiencing massive lags (1-10 sec freezes= , irregularly, but a couple times an hour) on our client machines, whic= h are mounting their homes via NFSv4. There seem to be no related messa= ges or warnings on the server nor the clients. >=20 > We come to the conclusion that 1.0.9 was the more stable version. The= resulting question is: Which nfs-utils version would be best suited fo= r SLES10SP3 (see kernel version above). Can we go straight to the lates= t 1.2.2 available or should we stick with 1.0.xx version? No idea. We'd have to figure out the cause of the lags. --b.