I've experimented a crash of my -current system today:

$ uname -a
NetBSD titine.fr.eu.org 1.5Y NetBSD 1.5Y (TITINE-Current-20010918)\
#13: Mon Nov 12 18:33:28 CET 2001   \
root@titine.fr.eu.org:/usr/src/sys/arch/i386/compile/TITINE i386

Context:

I was trying to read news with Gnus and it seems that, at same time,
the news daily crontab start. Then, my system deadly froze and the
only solution was to press the big button.

The only messages about this crash (just before it, i think) are found
in log/messages:

Nov 14 18:25:17 titine /netbsd: wd1b: uncorrectable data error reading fsbn 5715
2 of 57152-57159 (wd1 bn 1148816; cn 1139 tn 11 sn 11), retrying
Nov 14 18:25:36 titine /netbsd: wd1b: uncorrectable data error reading fsbn 5715
2 of 57152-57159 (wd1 bn 1148816; cn 1139 tn 11 sn 11), retrying
Nov 14 18:25:36 titine /netbsd: wd1b: uncorrectable data error reading fsbn 5715
2 of 57152-57159 (wd1 bn 1148816; cn 1139 tn 11 sn 11), retrying
Nov 14 18:25:36 titine /netbsd: wd1: transfer error, downgrading to Ultra-DMA mo
de 2
Nov 14 18:25:36 titine /netbsd: wd0(pciide0:0:0): using PIO mode 4, Ultra-DMA mo
de 5 (Ultra/100) (using DMA data transfers)
(etc.)


More generally, i've found that, when cron jobs (like system daily
cron tasks) start, the system slow down _very_ noticeably (e.g. when
i'm typing some document during such a cron job, words take long time
to appear). I wonder if it not the same pb.

As i've a Athlon 1GHz/256Mb RAM and ATA100/66 disk, i can't imagine is
related to my system (this pb doesn't appears with, say, FreeBSD or
Solaris). May be a nice value to adjust ?
-- 
Éric Jacoboni, né il y a 1309200239 secondes