Date: Thu, 8 Jul 2004 02:26:40 -0700 (PDT) From: Don Lewis <truckman@FreeBSD.org> To: mich@FreeBSD.org Cc: current@FreeBSD.org Subject: Re: today's CURRENT lockups Message-ID: <200407080926.i689QeS0008918@gw.catspoiler.org> In-Reply-To: <20040708091051.GC50232@mich2.itxmarket.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 8 Jul, Michael L. Hostbaek wrote: > Don Lewis (truckman) writes: >> >> I got bit by this problem when I tried to do some browsing on my >> Thinkpad R40 after my primary (4.10-STABLE) machine lost a disk. >> >> It's a July 5th kernel with SCHED_ULE, DDB, INVARIANTS, WITNESS, and >> ACPI. > > Try to cvsup your sources, and rebuild your kernel. rwatson's commit to > sched_ule.c (rev. 1.113) fix this issue for me. I use cvsup in CVS mode maintain a local copy of the repository, and then use cvs to update the local machines from that repository. My copy of the cvs repository lived on the dead disk. I fired up mozilla on my laptop to go shopping for a replacement disk ... There's a hole in the bucket, dear Liza, dear Liza There's a hole in the bucket, dear Liza, there's a hole. I was able to salvage all the data and scrounged up a couple of replacement disks, and when I loaded the data onto the new disks it became apparent that there are some severe performace problems in the filesystem code, probably in the inode and/or block allocators. Restoring 5 GB of data to the cyrus imap spool directory that was also on that disk took about 6 hours ...
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200407080926.i689QeS0008918>