Because the list manager has failed to add "Reply-To" headers to any
messages, I have ended up talking only to Steve, which was not my
original objective.
I'm posting this here so that it gets archived and that people know that
my problem is solved. (Yeah, I want the Internet to remember that I hate
PCI SCSI RAID controllers. :-D)
I'll probably also have a chat with the list server admin about
Reply-To. Lack thereof is, IMHO, a Bad Thing(tm), and this would be the
first ML without. (Yup, just about all the other lists I'm on are served
by either Majordomo or Mailman...)
~~ Ondra
-------- Original Message --------
From: Ondřej Hošek <[EMAIL PROTECTED]>
Subject: 4.0u0 and testing crash during install on Lynx
Steve Langasek wrote:
If you can get some readable output with the text interface, I'll see what I
can do to help fix it, or at least direct the bug report in the right
direction
With a slightly esoteric-looking setup (RDP from machine A to machine B
because A didn't have a serial port, serial connection from B to Alpha
for SRM console, Debian installation on Alpha's VGA console), I found
out that it was a crash in the PALcode. The machine worked after
removing the PCI SCSI RAID controller (Mylex DAC960, as far as I
remember correctly), getting as far as the hard drive detection and then
aborting because I didn't have any disks installed. :-D
I'll have to toy with the integrated disk stack. As far as I can tell,
you can connect it to the on-board SCSI controller either as one
probably-RAID-0-array (with up to eight disks) or two such arrays (four
disks each, probably a RAID 1/0). The current mode (as the box came from
the General Hospital in Vienna/AUT) is two-arrays where the first
would-be array had no disks and the second array was connected to the
Mylex. I'll have to replace two elements on the disk stack's PCB (a
terminator and a jumper) to switch it to one-array mode, which I'll
probably do tomorrow (if time and toolset allow).
I'll get rid of all PCI RAIDs since I absolutely don't feel like having
to shove them up an x86, reconfigure crap, put it back into the Lynx and
wait for it to fail for some esoteric reason. I wonder how all the
patients survived with THAT piece of hardware governing the storage. :-D
Thanks again for your patience. Once the system is up and running, I'll
drop a line so that everybody here knows Debian Testing works with Lynx.
~~ Ondra
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
