[2390] in linux-net channel archive
Re: Internet Provider?
daemon@ATHENA.MIT.EDU (Jon Lewis)
Sun Apr 7 18:46:47 1996
Date: Sun, 7 Apr 1996 18:25:19 -0400 (EDT)
From: Jon Lewis <jlewis@inorganic5.fdt.net>
To: Alan Cox <alan@lxorguk.ukuu.org.uk>
cc: linux-net@vger.rutgers.edu
In-Reply-To: <m0u5sZy-0005FYC@lightning.swansea.linux.org.uk>
On Sun, 7 Apr 1996, Alan Cox wrote:
> Having a pile of heavily battered terminal servers doesn't prove its solid
> but it sure as hell proves its when its faulty, and the existing races can
> be measured in seriousness by uptimes and the occasional Oops log to analyse.
> Thats what showed up other problems after I thought I had it fixed, and Ted
> has found other bugs since, so some are very well hidden indeed.
I suspect there are additional races in the driver. I read about the one
mentioned in the rocket.o 1.10 release notes about writing to the port
while closing the port...but most of my Oops/panics have been during
the startup of a PPP session. I'm still using 1.08 + a small patch from
Ted that keeps the ports from "soft hanging". Upgrading the motherboard
from dx4-100 to P100 seems to have helped. It's currently tied with its
longest uptime since going beyond 50 ports.
ewok:~# uptime
6:19pm up 9 days, 5:21, 45 users, load average: 0.01, 0.02, 0.00
It's running all 64 ports now on 2 RA32 boards.
------------------------------------------------------------------
Jon Lewis | Mime attachments are OK
jlewis@inorganic5.fdt.net | But please ask before sending
http://inorganic5.fdt.net | unsolicited huge files.
________Finger jlewis@inorganic5.fdt.net for PGP public key_______
