Reiserfs partition keeps getting checked after lockup

BenjiF · 11-12-2003, 07:07 PM

I have an old Pentium II running as an Apache web server under Slackware 9.0 (with the 2.6.0-test9 kernel). From time to time, I do something strange enough to cause a lockup. When the system reboots, fsck is often asked to check the root partition for errors - which it never finds, because the file system is Reiserfs. The line from /etc/fstab is:

/dev/hda5 / reiserfs defaults 0 0

It wastes a *long* time checking this partition. What can I do to convince Linux not to do this?

|)
|)enji

Tinkster · 11-12-2003, 08:09 PM

Don't use experimental kernels on
production machines? :)

Cheers,
Tink

h3pc4t · 11-12-2003, 08:27 PM

I do know that your / partition should be:

/dev/hda5 / reiserfs defaults 1 1

Read man fstab.

BenjiF · 11-12-2003, 08:59 PM

Quote:

Originally posted by Tinkster
Don't use experimental kernels on
production machines?

Might this be a kernel issue? I'd assumed it was related to /etc/fstab, fsck, or some of the startup scripts about which I know so little.

I'll try rebooting the machine in the middle of some heavy disk activity, and boot into a 2.4 kernel. That should resolve the question.

Quote:

Originally posted by h3pc4t
I do know that your / partition should be:

/dev/hda5 / reiserfs defaults 1 1

Read man fstab.

Already had. From man fstab:

Code:

The  fifth field, (fs_freq), is used for these filesystems
by the dump(8) command to determine which filesystems need
to  be dumped.  If the fifth field is not present, a value
of zero is returned and dump will assume that the filesys-
tem does not need to be dumped.

The  sixth field, (fs_passno), is used by the fsck(8) pro-
gram to determine the order in which filesystem checks are
done at reboot time.  The root filesystem should be speci-
fied with a fs_passno of 1, and other  filesystems  should
have a fs_passno of 2.  Filesystems within a drive will be
checked sequentially, but filesystems on different  drives
will  be  checked  at the same time to utilize parallelism
available in the hardware.  If the sixth field is not pre-
sent  or  zero,  a value of zero is returned and fsck will
assume that the filesystem does not need to be checked.

My interpretation of this information would be that values of 0 for both of these fields should prevent the partition from being checked. The values used to be 1. I changed them to 0 thinking that it would stop this annoying behavior - but it had no effect.

|)
|)enji

BenjiF · 11-13-2003, 04:15 PM

The root partition still gets checked when I boot the 2.4 kernel which ships with Slackware 9.0. I'm inclined, therefore, to think that this is not a kernel bug. Thanks, nonetheless, for pointing out such an obvious path of investigation, which I'd managed to overlook completely

I'd welcome any additional suggestions/recommendations.

(Hmmm...I just ran this message through the board's built-in spell-checker. First "mis-spelled" word: Slackware.)

|)
|)enji