From: Paul LaMadeleine (plamadeleine@lightbridge.com)
Date: Tue May 14 2002 - 09:56:37 EDT
Well,
I found the problem. It turned out to be I had the console in the
wrong serial console port. Another Sys Admin (a vms one) found this out
after I asked him to take a look at the box. Here is basicly what I did
when I found the bad disk:
1) shut down the server
2) removed the server from the cab - disconnecting all cables
3) opened the box up
4) unplugged the bad hard disk
5) put the box back together
6) put the box back in the cab
7) booted to the boot prompt
8) did a show dev (seen below) - things looked good
9) booted from cdrom and as soon as the system started to boot
unix - it stopped responding to console
I then spent day's working on getting this thing to boot.
I then plugged the console into serial port 1 and instantly got my
console back.
The vms guy said this was done this way do DEC could dial in on a
modem on serial port 2 and do hardware diags and not be on a priviaged port.
thanks to all who replied. Most of the responses were -
check the console var (make sure it's set to serial),
check scsi term as it's possible to have three terminators
and you are only allowed two,
completely remove the failed drive not just disconnect
it's power,
check scsi id's to make sure you don't have an issue,
bad boot media (I did try a 5.0, 5.1 and a 5.1A boot cdrom
- I was starting to think I lost the cdrom drive),
Paul
At 09:00 AM 5/14/02 -0400, Paul LaMadeleine wrote:
>Greetings,
>
> I've got an Alphaserver 400 that was running t64 5.1a just
> fine. It lost an internal disk so I disconnected it's power cable and
> attached and external disk cab with two disks in it. However, now I
> can't seem to boot from CDROM. I can do a "show dev" from the the boot
> prompt fine. When I issue a "boot dka400" ( the cdrom drive ) this is
> what I get:
>
> >>>boot dka400
>(boot dka400.4.0.6.0)
>block 0 of dka400.4.0.6.0 is a valid boot block
>reading 15 blocks from dka400.4.0.6.0
>bootstrap code read in
>base = 1f2000, image_start = 0, image_bytes = 1e00
>initializing HWRPB at 2000
>initializing page table at 1e4000
>initializing machine state
>setting affinity to the primary CPU
>jumping to bootstrap code
>
> I see a lot of cdrom disk activity, but no console activity. I
> let it sit like this for about 1/2 hour and no results.
>
> any ideas?
>
> thanks,
>
> Paul
>
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:48:41 EDT