From owner-freebsd-bugs@FreeBSD.ORG Mon Nov 24 18:10:10 2008 Return-Path: Delivered-To: freebsd-bugs@hub.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id EE2AD106568D for ; Mon, 24 Nov 2008 18:10:09 +0000 (UTC) (envelope-from gnats@FreeBSD.org) Received: from freefall.freebsd.org (freefall.freebsd.org [IPv6:2001:4f8:fff6::28]) by mx1.freebsd.org (Postfix) with ESMTP id DFC088FC24 for ; Mon, 24 Nov 2008 18:10:09 +0000 (UTC) (envelope-from gnats@FreeBSD.org) Received: from freefall.freebsd.org (gnats@localhost [127.0.0.1]) by freefall.freebsd.org (8.14.3/8.14.3) with ESMTP id mAOIA9jb048945 for ; Mon, 24 Nov 2008 18:10:09 GMT (envelope-from gnats@freefall.freebsd.org) Received: (from gnats@localhost) by freefall.freebsd.org (8.14.3/8.14.3/Submit) id mAOIA9mS048944; Mon, 24 Nov 2008 18:10:09 GMT (envelope-from gnats) Date: Mon, 24 Nov 2008 18:10:09 GMT Message-Id: <200811241810.mAOIA9mS048944@freefall.freebsd.org> To: freebsd-bugs@FreeBSD.org From: Clarence Chu Cc: Subject: RE: kern/129132: 7.1-Beta2 nfsd cannot lock X-BeenThere: freebsd-bugs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: Clarence Chu List-Id: Bug reports List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 24 Nov 2008 18:10:10 -0000 The following reply was made to PR kern/129132; it has been noted by GNATS. From: Clarence Chu To: Cc: , Subject: RE: kern/129132: 7.1-Beta2 nfsd cannot lock Date: Tue, 25 Nov 2008 01:50:54 +0800 --_65d8bc99-85e7-4783-a5c6-8c11502bcac6_ Content-Type: text/plain; charset="big5" Content-Transfer-Encoding: 8bit Dear Remko, > > Which results in the question: do you have rpc.statd and rpc.lockd running > on the machine? > definitely I had started rpc.statd and rpc.lockd, what appears strange to me is the setting when flipped between 7.0-p5 and 7.1-BETA2 behave differently (via # make installworld; make installkernel; reboot). so, I can assure you that it is not my server setting issue. I had actually spare a box for the test and report it as FreeBSD is the only FOSS doing well with lockd other than Solaris. Yes, _ALL_ linuxes failed that, AFAIK. I had switched to 7.1-PRERELEASE for a while and am to upgrade all my nodes (about 20) to 7.1 soon. > > I copied in Doug Rabson, who did a lot of work on NFS recently and might > have more information wrt. Locking etc. (or not, then I was mistaken). > well, the senerio is repeatable as described in the PR, too. Will watch for csup logs for any delta to nfs-related code. Also, I'm not sure, three of my SMP/Core2duo nodes failed to "make buildworld; make installworld" both for i386 and amd64, the cc generated dump core, and seems no problems for UP systems. the numerous device drivers additions and CTF capability is attractive, provided -RCs are more reliable than the -BETAs. Best wishes, Clarence CHU _________________________________________________________________ Invite your mail contacts to join your friends list with Windows Live Spaces. It's easy! http://spaces.live.com/spacesapi.aspx?wx_action=create&wx_url=/friends.aspx&mkt=en-us --_65d8bc99-85e7-4783-a5c6-8c11502bcac6_ Content-Type: text/html; charset="big5" Content-Transfer-Encoding: 8bit

Dear Remko,

>
> Which results in the question: do you have rpc.statd and rpc.lockd running
> on the machine?
>

definitely I had started rpc.statd and rpc.lockd, what appears strange to
me is the setting when flipped between 7.0-p5 and 7.1-BETA2 behave
differently (via #  make installworld; make installkernel; reboot).  so,
I can assure you that it is not my server setting issue.

I had actually spare a box for the test and report it as Fr eeBSD
is the only FOSS doing well with lockd other than Solaris.
Yes, _ALL_ linuxes failed that, AFAIK.

I had switched to 7.1-PRERELEASE for a while and am to upgrade all
my nodes (about 20) to 7.1 soon.

>
> I copied in Doug Rabson, who did a lot of work on NFS recently and might
> have more information wrt. Locking etc. (or not, then I was mistaken).
>


well, the senerio is repeatable as described in the PR, too.
Will watch for csup logs for any delta to nfs-related code.

Also, I'm not sure, three of my SMP/Core2duo nodes failed
to "make buildworld; make installworld" both for i386 and
amd64, the cc generated dump core, and seems no problems
for UP systems.

the numerous device drivers additions and CTF capability is
attractive, provided -RCs are more reliable than the -BETAs.

Best wishes,

Clarence CHU


Invite your mail contacts to join your friends list w ith Windows Live Spaces. It's easy! Try it! --_65d8bc99-85e7-4783-a5c6-8c11502bcac6_--