[plug] Boot up problems RH6.2

[Mark Bailey]

Thanx Matt.

Thought I would check after reading the bitwizard article for reasons
for SIGNAL 11. The following prompted me to take the lid off ...

"CPU temperature. A high speed processor might overheat 
 without the correct heat sink. This can also be caused 
 by a failing fan" and 
"Dust buildup. Some dust might conduct a bit and create 
 a weak short. It might increase capacitances somewhere, 
 and degrade timing characteristics".

Sure enough. Fan was full of dust. Gave the box a good clean
out and so far, presto. Has rebooted a dozen times with no problem.

It seems to still be unmounting /hda1 improperly on shutdown but
goes throught the check on boot up without a problem.

Hey, really appreciate the help.

Cheers,
Mark Bailey.

-----Original Message-----
From: Matt Kemner [mailto:zombie at wasp.net.au]
Sent: Saturday, 27 January 2001 9:31 PM
To: plug at plug.linux.org.au
Subject: Re: [plug] Boot up problems RH6.2


On Sat, 27 Jan 2001, Mark Bailey wrote:

> Warning... fsck.ext2 for device /dev/hda1 exited with signal 11. [FAILED]

Signal 11 (SEGFAULT) usually means a hardware problem.

It can also signify a software problem (bug) but programs like fsck which
is one of the most widely used linux programs can be assumed to be fairly
bug-free.

See http://www.bitwizard.nl/sig11/ for more info.
 
> Give root password for maintenance
> (or type Control-D for normal startup):
> 
> Obviously I have tried the Control-D bit to no avail.

Have you tried the "Give root password for maintenance" bit?

It should drop you to a shell where you can type
"e2fsck /dev/hda1" which will try to check the drive again.
If you still get segfaults you will need to try to isolate which of your
hardware is causing problems - see the Sig11 FAQ I linked above.

> Is there anything I can run to do a check on the filesystems.

Yes, "e2fsck /dev/hda1" but you will need to remount the drive read-only
first, which will probably mean you will need to go to single-user mode,
so you are better off rebooting and running the fsck command from the
maintenance prompt, as above.
 
> Also something I noticed in the dmesg is the following which
> does not sound correct. Surely the VFS line should not be
> read only. Any ideas where I can fix this.

No fix needed, this is the way Linux boots - with the root system mounted
read-only so fsck can scan the disk without any other process interfering
by writing to the disk at the same time fsck is.
 
After fsck is finished, the root filesystem will be remounted read-write.

 - Matt