Date: Sat, 17 Oct 1998 09:51:05 +1100 (EDT) From: Andrew MacIntyre <andymac@bullseye.apana.org.au> To: Jaye Mathisen <mrcpu@internetcds.com> Cc: stable@FreeBSD.ORG Subject: Re: Annoying >2.2.5 oddity on reboot. Message-ID: <Pine.OS2.3.95.981017091545.181B-100000@CENTRAL> In-Reply-To: <Pine.NEB.3.95.981016090250.2118M-100000@schizo.cdsnet.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 16 Oct 1998, Jaye Mathisen wrote:
{...}
> They cold boot perfectly. Just won't warm reboot, w/o hitting the reset
> button.
>
> On Fri, 16 Oct 1998, Chris Johnson wrote:
>
> > On Fri, Oct 16, 1998 at 08:27:08AM -0700, Jaye Mathisen wrote:
> > >
> > >
> > > I have a bunch of DEC ZX and HX 6000 P6 boxes. Have been running 2.2.5
> > > for heap 'um big long time, zero troubles.
> > >
> > > So I upgraded 2 of them to -stable as of yesterday.
> > >
> > > Now neither of them will reboot properly. They shutdown, and get to the
> > > stage where they should start reloading, and they just hang.
{...}
This is not an answer to your problem, but another datapoint for review.
The machine concerned is a HP Netserver LM/60 (EISA/ISA only, no PCI),
fitted with 32M and 5x1GB mechs on one channel of the internal EISA SCSI
controller (Adaptec, basically a 2742T AFAICT).
I had an Intel EtherExpress/16 (ie0 driver) installed when I did the
initial install (2.2.7-RELEASE, after several of the initial problems were
fixed), and I had no problems with doing a passive FTP install. The
intitial reboot after FTP went OK - at least there were no error messages.
At that point I left it for a while, and came back some time later to do
the package installation.
Despite the fact the ifconfig reported the interface up, I could not
access the network - eventually it timed out (in sendto??? vague
recollection here). Not to mention the fact it worked fine during the
install.
A subsequent shutdown -r hung immediately after the "going down" message,
very similar to your symptoms, and required a reset to restart.
I found an Intel EtherExpress Pro/10 ISA card (ex driver), and swapped it
for the EE16, mainly to get rid of some problems the ie driver was
reporting (card/packet errors of some sort). Same thing - detected fine &
ifconfig'ed, but no viable connection.
I can't recall the exact process that followed, but I did find that
building a custom kernel produced weird results:-
certain combinations of the ie and ex drivers along with other kernel
settings (no of users I think) produced kernels that would work with one
of the drivers if the other driver was present but disabled. Also
some kernels had the rebooting hang and others were OK.
The kernel that I'm currently using is working just fine with the EEPro10
(the ex driver is in the kernel but disabled), and reboots fine, but
displays the hang symptoms on a halt (it never gets to say "system halted"
- the last thing on the screen is the "going down" message.
The one apparent commonality with your problem is older hardware with EISA
support. Do your boxes have EISA SCSI controllers onboard and are you
using them? (some of the HP Netservers had EISA SCSI even when they had
PCI buses - I guess because there weren't many reliable PCI-PCI bridge
chips in the early PCI days)
I hadn't tried to install older versions on to this box, so I can't
compare behaviour.
--
Andrew I MacIntyre "These thoughts are mine alone..."
E-mail: andrew.macintyre@aba.gov.au (work) | Snail: PO Box 370
andymac@bullseye.apana.org.au (play) | Belconnen ACT 2616
Fido: Andrew MacIntyre, 3:620/243.18 | Australia
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?Pine.OS2.3.95.981017091545.181B-100000>