I've taken the liberty to remove debian-boot and Adam DiCarlo from this
thread. Adam Di Carlo because as if he doesn't get enough email
already, and debian-boot because I was informed by Mr. Di Carlo it
doesn't handle this package--also referencing the debian-boot list
description. Please forward this to them if I am mistaken.
To reply to Ben, I'm really not that slick with Debian, so I have to
say, I just answer package questions with the best of my ability and the
little research I do into the man pages of packages and programs. I
could have sworn all I did was upgrade. Why it didn't work is beyond
me, but forcing a write of the boot block helped (a lot!).
There are some peculiar things though:
When I first booted with the Potato drive I ran fsck on the broken drive
and it cleaned up some zero dtime inodes (not really amazing, but I
thought I'd mention it).
Also, here is my drive map with my 18.2GB SCA drive
/dev/sda1 35MB /boot
/dev/sda2 16GB /[everything_else]
[and empty space for the rest here]
I was thinking there was a problem with silo trying to look at
/etc/silo.conf since it's not contained in the /boot drive partition.
Is this possible? I witnessed a silo.conf in the /boot directory
partition, but I'm not sure how it got there. The silo execution that
fixed my drive was almost eighty characters long cause I had to tell it
where to get its information and where to write the boot block etc, etc.
I'm glad to hear of your past success. It's always good to know I'm the
only one this happened to (sort of). At least that means Debian > Woody
upgrades should go flawless for the rest. It is possible I did miss
something.
Mario
Ben Collins wrote:
>On Fri, Nov 09, 2001 at 12:48:24AM -0500, Mario Lombardo wrote:
>
>>1. Installed old Potato 2.2r2 hard disk in SCSI bus under target zero
>>2. Did a STOP-A, set the automatic boot under printenv (setenv) to false
>>3. 'reset' the machine
>>4. at the PROM prompt I typed 'boot disk0'
>>5. When I logged into Linux under root, I mounted the non-bootable disk
>>under /mnt/bigdisk
>>6. I ran 'silo -r /mnt/bigdisk' with some other forgotten options to
>>force a write of the boot block (see the SILO manual
>>http://silo.sourceforge.net/)
>>7. rebooted and all was ok
>>
>
>If that's all you did, then it sounds like one of two things happened.
>Either you didn't allow silo to rerun itself when it upgraded, or
>something in your boot block got corrupted.
>
>I've never heard of this problem occuring, and quite honestly, as the
>current SILO upstream maintainer, I can tell you I've done a lot of
>upgrades of the SILO package from 0.9.x to current 1.2.x versions.
>
>
>Ben
>
This archive was generated by hypermail 2.1.3 : Fri Aug 01 2014 - 17:26:37 EDT