From: "Tom Sightler" <ttsig@tuxyturvy.com>
To: "Igor Mozetic" <igor.mozetic@uni-mb.si>, <linux-kernel@vger.kernel.org>
Subject: Re: Any stable 2.4 kernel?
Date: Sat, 27 Oct 2001 15:00:19 -0400 [thread overview]
Message-ID: <002801c15f19$a4833d50$3a01a8c0@zeusinc.com> (raw)
In-Reply-To: <15322.33513.293148.371409@cmb1-3.dial-up.arnes.si>
> I wonder if anybody has found a stable kernel for the following
> hardware: C440GX+, dual Xeon 550, 2GB RAM, 1GB swap, aic7xxx.
> Usage pattern is load > 2, highmem, not much I/O (maybe swap?).
> Some of our jobs take weeks, so stable means months between reboots.
> I found anything beyond 2.4.10 useless - lockups after a few days.
> Currently I run 2.4.3 with varying degree of success - initial lifespan
> was 4 months, but last reincarnation survived for 3 weeks only.
I'll offer a few suggestions.
What particular distribution are you using? If using a commercial
distribution like SuSE or Redhat have you tried the kernels that they
provide. I've generally found these kernels to be very stable as both of
these vendors perform fairly rigorous testing before making a kernel release
to the public.
The next suggestion would be to actually attempt to determine the reason for
the crash and see if it can be fixed. I would generally agree that your
problem is probably with highmem, and since 2.4.10 was when Andrea's VM went
in (i think) maybe that's why kernels after that point are worse for you.
Perhaps try a recent -ac kernel and see what they act like.
> Any recommendation for 2.4 or should I consider going back to 2.2 ?
> I don't need any fancy features (apart to SMP and highmem),
> only stability is important.
Well, I still run 2.2 on all the machines that need 100% reliability, and
while I also have several 2.4.x machines that have achieved 200+ day
uptimes, they are all single processor <512MB RAM machines. The 2.2 servers
are all dual or quad processor with >2GB RAM and all seven achived 150+ day
uptimes, although I'll finally have to reboot them all soon to apply the
recent SuSE kernel updates for the recent security exploits (even though I'm
fairly low risk for them I thought it was a good excuse to schedule
maintanence). I'm currently testing one of the machines with the 2.4.7
kernel from SuSE, but I'm still not ready to make the jump on the really
loaded machines.
Later,
Tom
next prev parent reply other threads:[~2001-10-27 19:13 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <Pine.LNX.4.33.0110271212070.446-100000@mikeg.weiden.de>
2001-10-27 9:48 ` Igor Mozetic
2001-10-27 9:56 ` Rik van Riel
2001-10-27 10:01 ` Dan Hollis
2001-10-27 10:03 ` Jeff Garzik
2001-10-27 14:16 ` Eric W. Biederman
2001-10-27 14:12 ` Eric W. Biederman
2001-10-27 15:40 ` Alan Cox
2001-10-27 20:51 ` Rik van Riel
2001-10-27 10:24 ` Igor Mozetic
2001-10-27 11:50 ` Igor Mozetic
2001-10-27 19:00 ` Tom Sightler [this message]
2001-10-28 14:08 Steve.Batson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='002801c15f19$a4833d50$3a01a8c0@zeusinc.com' \
--to=ttsig@tuxyturvy.com \
--cc=igor.mozetic@uni-mb.si \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®