Date: Mon, 27 Jul 1998 05:58:37 +0200 (MET DST) From: Sascha Schumann <sas@www.schell.de> To: David Greenman <dg@root.com> Cc: Peter Mutsaers <plm@xs4all.nl>, freebsd-questions@FreeBSD.ORG Subject: Re: FreeBSD or LINUX??? - Which one should I choose? Message-ID: <Pine.LNX.3.96.980727054711.10769B-100000@www.schell.de> In-Reply-To: <199807270250.TAA18903@implode.root.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 26 Jul 1998, David Greenman wrote:
> >> Maybe, but from my experience, FreeBSD (even -current) is more stable
> >> than Linux.
> >
> >I'm running FreeBSD-stable (2.2.7) now for four days and it crashed three
> >times on me. The first time, I copied a 2MB file to a clean ext2fs
> >partition - the system hang (I could still switch between terminals),
> >but the partition was mixed up - lots of errors while running e2fsck. The
> >second time happened while hammering the FreeBSD machine with lots of web
> >request. The system froze (=totally dead) after ~2M requests. The third
> >time was again disk related, "Freeing free block" and system reboot within
> >15 seconds while installing a new kernel image.
>
> Ext2fs doesn't get a lot of testing and has been known in the past to be
> a little buggy. I wouldn't be surprised if the 3rd panic was indirectly
The ext2 partitions contain my normal working files. There is sth else
with it - if I don't unmount the ext2 partitions manually, a 'shutdown -r
now' will unmount zero partitions - instead it will be "giving up" and
starts fsck'ing the next time I start it up. Probably some kind of mistake
in the scripts, I didn't look into this up to now. (hints welcome ;)
> caused by things that happend in the first. The second panic sounds like
> a kernel misconfiguration - for busy WWW servers, you need to be careful
> about how you many network buffers you configure. This has been talked
> about to death in our lists, but nonetheless, you can probably fix the
> problem with something like
>
> options "NMBCLUSTERS=10000"
I had set it to 4096 according to a online article on configuring Apache
and FreeBSD for high performance.
> ...in your kernel config file. If the system runs out it will eventually
> panic. The only solution is to configure enough buffers to handle the peak
> usage.
The disk I copied to (a old 500MB Conner, UFS) has/had bad blocks. I
played with bad144/badsect/fsck shortly before that and dd'ed some test
files around the partitions. That was probably the cause.
Bye,
Sascha
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.3.96.980727054711.10769B-100000>