Index: [Article Count Order] [Thread]

Date:  Sat, 28 Oct 2006 11:38:30 -0700
From:  "Ken Marcus - Precision Web Hosting, Inc." <kenmarcus (at mark) precisionweb.net>
Subject:  [coba-e:07765] Re: SCSI
To:  <coba-e (at mark) bluequartz.org>
Message-Id:  <0e0501c6fac0$4464a830$6700a8c0@OfficeKen>
References:  <001101c6fa46$e9f637c0$1702a8c0 (at mark) homepc1> <45430E74.2080107 (at mark) mixfans.org>
X-Mail-Count: 07765


----- Original Message ----- >

> BCooper wrote:
>> Have you reboot again since you steps below?
>>
>> Blair
>>
>> -----Original Message-----
>> From: Ken Marcus - Precision Web Hosting, Inc.
>> [mailto:kenmarcus (at mark) precisionweb.net] Sent: Friday, October 27, 2006 2:52 
>> PM
>> To: coba-e (at mark) bluequartz.org
>> Subject: [coba-e:07747] Re: SCSI
>>
>>
>>> On Fri, 27 Oct 2006, Colin Jack wrote:
>>>
>>>
>>>> Hi Blues,
>>>>
>>>> We have had a very uncomfortable day and would appreciate some
>>>> help/ideas from those more knowledgeable than me!
>>>>
>>>> We had a server with Brian's 4.3 ISO installed and running happily. We
>>>> have rebooted the server since install a number of times without
>>>> incident and yum updated regularly.
>>>>
>>>> Then we recently had a yum update stall on us and
>>>>
>>>> killall yum_install
>>>> killall yum
>>>>
>>>> This sent the server (which was working up until then) into a sulk.
>>>> We rebooted and it wouldn't go past the LILO
>>>>
>>>> Booted a Knoppix disk and tried to mount but would only see /boot
>>>>
>>>> We have an identical (hardwarewise) back up server without an OS, so
>>>> installed 4.5, which appeared to install cleanly but on reboot did the
>>>> same as the sick server. Tried the 4.6 ISO and the 4.3 ISO .... with 
>>>> the
>>>> same result. Stopped at the LILO screen.
>>>>
>>>> The guy working on it is pretty knowledgeable (more so than me) but
>>>> couldn't resolve the problem.
>>>>
>>>> We have restored to another lower spec machine (Dell) and that is
>>>> running - albeit very slowly.
>>>>
>>>> I need to get the main servers running, but am short on ideas.
>>>>
>>>> Basic spec.
>>>>
>>>> Intel mainboard
>>>> 1 x Xeon
>>>> 2Gb RAM
>>>> Intel SCSI
>>>> 2 x 72Gb HDD (software mirror)
>>>>
>>>> Anybody any ideas?
>>>>
>>>> many thanks
>>>>
>>>> Colin
>>>>
>>>>
>>>>
>>
>>
>>
>>> Hi,
>>>
>>> There was a time when my lilo failed on my redhat 7.x server
>>> after a reboot.
>>>
>>> So I overwrote it with grub and it works.
>>>
>>> so .. 2 cents try grub.
>>>
>>>
>>> Cheers
>>> patrick
>>>
>>>
>>
>> I had that on a P4 with software Raid Sata drives. It got stuck on Grub, 
>> wnated to boot from the CD. I asume it did not see the drives.
>>
>> The way I fixed it was:
>> 1. booted up using Brian install disk,
>> 2. then *carefully* selected the rescue mode,
>> 3. followed the chroot instructions that appear
>> 4. typed /sbin/lilo -v
>> 5. logged out and took out the CD
>>
>> Then it booted normally.
>>
>>
>> ----
>> Ken Marcus



From: "Dennis"


> 24 days I had the same and also reported here. Nobody could help nor any 
> answer
> So there is something in the installation which is causing this issue (I'm 
> using one of the first ISO images of Brian)
> it's 100% yummed.
>
> However I managed to have my solution, although it's already 24 days the 
> server did not go down and I have no time to do a re-test
>
> It's true that during the rescue option you can fix LILO again to have the 
> system booting again, until, you reboot again
>
> My system was able to boot / reboot again after multiple times using the 
> LILO command
>
> So machine dead: run rescue, mount the image, run lilo
> reboot and see the system running again
> than again: run LILO as root
> than reboot
> see what's happening is it rebooting?
> reboot again ..
> rebooting normally? problem fixed
>
> Why all this is happening to us: No idea, but if it's something causing to 
> people with GRUB and LILO: it has to do with an update somewhere.
> or kernel or something else.
>
> Yes I am using software raid also and just before all this stuff my 
> machine was dying under me: failing services, wierd hdd stuff ..
> I WAS using latest kernel, but removed that one due to wierd things and I 
> only use kernels from where I know: that thing is working for me.
> now I am using 2.6.9-42.0.2.EL kernel and it works.
>
> Please check: kernel versions, Lilo, rescue, normal working mode running 
> Lilo and Kernel change to older version ..
>
> Dennis

My thought was that it was due to a yum update. So to be safe,  and make 
sure I would not forget on that server, I have a cron that runs the 
/sbin/lilo -v  daily.

Also, I don't recall if this was one of the servers that we had manually 
tried the GRSEC kernel ( http://www.grsecurity.net/ ) on. The ASP did not 
work with that kernel.  So, it may have been something we did.


----
Ken Marcus
Precision Web Hosting, Inc.
http://www.precisionweb.net