Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 12 Dec 2001 21:44:56 +0200
From:      Alex Popa <razor@ldc.ro>
To:        Morsal Roudbay <morsal@swipnet.se>
Cc:        Brady Montz <bradym@mail.hydrologue.com>, stable@FreeBSD.ORG
Subject:   Re: 4.4-STABLE crashes - suspects new ata-driver over wd-driver
Message-ID:  <20011212214455.C10147@ldc.ro>
In-Reply-To: <20011210184258.GC25562@zigman.2y.net>; from morsal@swipnet.se on Mon, Dec 10, 2001 at 07:42:58PM %2B0100
References:  <200112050033.fB50X6G12476@mail.hydrologue.com> <20011210184258.GC25562@zigman.2y.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Dec 10, 2001 at 07:42:58PM +0100, Morsal Roudbay wrote:
> On Tue, Dec 04, 2001 at 04:33:06PM -0800, Brady Montz wrote:
> > Morsal Roudbay <morsal@swipnet.se> writes: 
> >  
> > > On Sat, Dec 01, 2001 at 02:45:18AM -0500, Chris BeHanna wrote: 
> > I've been recently noticing the same thing with my machine. I don't 
> > have the specs handy since I'm at work, but I am using IDE drives. 
> > The trouble started last week, using a 4.4-STABLE kernel built on Nov 3. 
> > I reinstalled world yesterday from the latest 4.4-STABLE,
> > and it hasn't gotten better. 
> >  
> > For me, it only happens during heavy disk access, such as when backing 
> > up my hard drive to a huge tar file, or when doing build or clean. My 
> > machine will either lock up or reboot. I haven't seen any messages in 
> > any of my log files. The MTBF under such load (like portupgrade mozilla) seems
> > to be about an hour or so.
> > 
> > Other data poiints: I ran memtest86 overnight, and it didn't find any 
> > memory errors, and tripwire hasn't noticed any filesystem corruption.
> 
> I ran some checks on my harddrives (IDE ones) and RAM, no indications of failure.
> 
> If I upgrade several ports at the same time the machine shuts down.
> 
> This sever of mine worked very well before the upgrade to 4.4-STABLE.
> 
> Morsal

I still have the same kernel I reported the freezes with, but for now
the machine has been running for about 8 days, without crashes. 

I even ran bonnie++ over the weekend. It ran on the partition (actually
vinum mirrored volume) on which the system was writing when I had the
initial crashes.  Arguments to bonnie++ were "-s 3g -n 128:16384:0:16".
I ran bonnie++ twice, first on the volume w/o softupdates, then w/
softupdates, and I got no crashes.

During the bonnie++ runs iostat showed some 22M/s activity on each disk,
so I am beginning to suspect there has to be a correlation between heavy
disk I/O and heavy network I/O (my initial crashes occured when someone
was making a large backup over SMB to the server, at about 9M/s disk
activity) in order to get the crashes.

Also there might be something wrong with some part of the disks, because
when running bonnie++, the disks were about 12G more occupied.

Again, I am running 4.4-RELEASE, kernel config is based on GENERIC, but
only with 686 CPU support, maxusers 128, no INET6, with IPFILTER, no
SCSI, no pccard.

The next thing I am going to try (when I get physically to the server)
is to run bonnie++ in conjunction with some TCP/IP stress, first over
loopback, then over ethernet.

Have Fun!

------------+------------------------------------------
Alex Popa,  |  "Artificial Intelligence is
razor@ldc.ro|         no match for Natural Stupidity"
------------+------------------------------------------
"It took the computing power of three C-64s to fly to the Moon.
It takes a 486 to run Windows 95. Something is wrong here."

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20011212214455.C10147>