Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 02 Apr 2001 21:48:26 +0200
From:      alex <ml-freebsd-smp@phobgate.de>
To:        freebsd-smp@freebsd.org
Subject:   Re: IBM PC Server 704 ... dual pentium pro ...
Message-ID:  <2700595854.986248106@MOD>
In-Reply-To: <Pine.BSF.4.33.0104021600150.91376-100000@mobile.hub.org>

next in thread | previous in thread | raw e-mail | index | archive | help
hi,

i just can report that i had the same problem using gigabyte board with 
dual p3 800mhz and adaptec dual channel controller with 4 scsi disks.
this was in december last year and i installed 4.2 release and vinum 
(raid5, like your config). under heavy disk load (severals processes 
gziping and ungziping large archives) the box suddenly halted or rebooted.

i have another box with dual p3 400mhz and adaptec scsi (without vinum) 
which runs fine for ~5 months now. even under heavy disk load.

maybe it's the comination of smp kernel and vinum...?

alex

--On Montag, 2. April 2001 16:11 -0300 The Hermit Hacker <scrappy@hub.org> 
wrote:

>
> Anyone have any experiences iwth one of these?  we've had it here, running
> Dual with Solaris/x86, for the past two years.  They upgraded the machine,
> and plan on using this one for a database server for one of their
> projects, so we installed FreeBSD 4.3-RC onto it ... but it appears to
> 'hang' periodically ...
>
> funny thing is that dmesg and /var/run/dmesg.boot show pretty much
> 'garbage':
>
> aittdb# dmesg
> num/raid:
> fxp0: <Intel Pro 10/100B/100+ Ethernet> port 0xff00-0xff3f mem
> 0xfe900000-0xfe9fffff,0xfeaff000-0xfeafffff irq 2 at device 11.0 on pci0
> fxp0: Ethernet address 00:d0:b7:bb:8c:63
> aittdb# cat /var/run/dmesg.boot
> OWNER=root MODE=100644
> aittdb#
>
> /var/log/messages shows from a few days ago shows, if anything stands out
> in here?
>
>
> and we're using vinum, in the following configuration, to give us a RAID5
> array:
>
> the last 'hang' happened when we tried to install jdk12-beta ... I'd say
> its whenever there is heavy disk I/O, but when it first started happened,
> and we'd run top, all the numbers came in pretty much as zero ...
>
> thoughts as to where to look?  known problems with these machiens?
>


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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?2700595854.986248106>