From owner-freebsd-current Wed Mar 4 11:19:27 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA01587 for freebsd-current-outgoing; Wed, 4 Mar 1998 11:19:27 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from critter.freebsd.dk (critter.freebsd.dk [195.8.129.14]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA01567 for ; Wed, 4 Mar 1998 11:19:19 -0800 (PST) (envelope-from phk@critter.freebsd.dk) Received: from critter.freebsd.dk (localhost [127.0.0.1]) by critter.freebsd.dk (8.8.7/8.8.5) with ESMTP id UAA06269; Wed, 4 Mar 1998 20:12:36 +0100 (CET) cc: Kevin Day , current@FreeBSD.ORG Subject: Re: Donations. In-reply-to: Your message of "Wed, 04 Mar 1998 19:33:43 +0100." <6135.889036423@critter.freebsd.dk> Date: Wed, 04 Mar 1998 20:12:35 +0100 Message-ID: <6267.889038755@critter.freebsd.dk> From: Poul-Henning Kamp Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG In message <6135.889036423@critter.freebsd.dk>, Poul-Henning Kamp writes: >In message <199803041708.LAA03872@home.dragondata.com>, Kevin Day writes: > >>Seriously, most of the bugs can be reproduced by having /usr/src over NFS, >>and doing the following: >> >>cp /dev/zero /usr/src/blah & >>make buildworld -j4 > >I just did this on a laptop that happens to be here on a visit. >only special thing here is that the laptop runs soft-updates... > >ALT-F2 > mount phk:/usr/src /usr/src > cd /usr/src ; make world >ALT-F1 > cd /usr/src ; dd if=/dev/zero bs=32k of=foo > >and it hangs solid :-( > >I'll ponder the packet trace for a moment and see if I can >figure something out. The laptop is stoned and I can't get >it into DDB :-( I hate this kind of problem, I ran over the packet trace, which btw was made on the NFS server. At end of the crash the client (the laptop) has five outstanding write requests which hasn't been answered or retransmitted, one of which is the very first one. Interestingly, the client issues a "commit" on the request it still has outstanding about 1.4 second after it was sent, and the commit is acked... Now, running with "nfsiod -n 8" I can't seem to provoke the hang, but dd will die saying that it wrote 641433600 bytes, (3132 x 200k), but the file on my disk is only 5128192 bytes (25*200k) Man I wish I had time to look into this problem :-( But it seems that raising the number of nfsiods may be a workaround and lowering it will probably make the problem even worse. No panics though... -- Poul-Henning Kamp FreeBSD coreteam member phk@FreeBSD.ORG "Real hackers run -current on their laptop." "Drink MONO-tonic, it goes down but it will NEVER come back up!" To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message