Port-vax archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Boot failure on 4000/500A



On 2019-04-12 21:43, George Harvey wrote:
On Fri, 12 Apr 2019 00:19:12 +0200
Johnny Billquist <bqt%update.uu.se@localhost> wrote:

On 2019-04-11 22:38, George Harvey wrote:
On Tue, 9 Apr 2019 21:24:07 +0100 George Harvey wrote:
On Tue, 9 Apr 2019 16:30:41 +0100 David Brownlee wrote:
On Mon, 8 Apr 2019 at 23:42, George Harvey wrote:

I have a VAX 4000/500A with a CQD-223/TM QBUS SCSI card. It was
previously running NetBSD 6.1.5 on an external SCSI disk and has
recently been upgraded to NetBSD 8.0. The 8.0 install kernel was
netbooted, found the SCSI disk (ra0) and completed the installation
normally. After rebooting, the generic kernel fails to find the CQD
card. Included below are the console outputs from the install and
generic kernels.
[ snip ]
My current state is that I can netboot both the install and generic
kernels with the -a option, the kernels see my CQD card as mscpbus1, I
can mount the SCSI disk (ra0) as root and the system boots normally.

All attempts to boot from disk result in failure to detect mscpbus1.

I've seen the posts about it being a memory size problem but if that is
the case, why does it work when I netboot?

Don't know. So you are booting the same kernel over network as on the
disk, and it works for you when you netboot?

Yes, works when I netboot, fails when I disk boot. Same kernel.

I can definitely show that for me, booting from disk, if I use one
amount of disk, it does not detect mscpbus1, but if I just configure the
system with less memory, it do detect mscpbus1. Same kernel, same
everything else. Literally, the only thing I change is the memory size.

I noticed something new today. After a cold start, I tried booting from
disk before doing anything else, and it worked! Found mscpbus1 and
booted normally. I then did a 'shutdown -h' tried booting again from
disk and it was back to failing to find mscpbus1. Netbooting still
works but all subsequent attempts to disk boot fail. Looks like some
difference in the hardware state between cold and warm starts.

That sounds even more weird. In my tests, it is entirely reproducable, and the same every time. Too much memory, and it won't detect the mscpbus.
However, the limit is different for different kernels.

  Johnny

--
Johnny Billquist                  || "I'm on a bus
                                  ||  on a psychedelic trip
email: bqt%softjar.se@localhost             ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol


Home | Main Index | Thread Index | Old Index