Ext3 users - Oct 2002 - How come fsck still kicks in and reports major errors with Ext3?

Greetings.


  Total newbie with Ext3.  I selected when I upgraded my system to 
RH7.2.  I upgraded my system again to RH8.0.  Here is some info:

[root@world root]#  cat /proc/mounts
rootfs / rootfs rw 0 0
/dev/root / ext3 rw 0 0
/proc /proc proc rw 0 0
usbdevfs /proc/bus/usb usbdevfs rw 0 0
none /dev/shm tmpfs rw 0 0
none /dev/pts devpts rw 0 0
/dev/hdd1 /mnt/genericdata1 ext2 rw 0 0

  Ok, so it seams that my / is Ext3, as specified in /etc/fstab:

LABEL=/                 /                       ext3    defaults        1 1

  Now, why is that when the power went off and later came back, I had to 
go for an hour with fsck trying to fix the HD?  Isn't journaling 
supposed to prevent this?  I had no feeling that the journal was 
sollicitated at boot time when the system detected errors.  Is my Ext3 
properly setup?  Is this normal?  Where resides the journal BTW?  Are 
there config files for Ext3?

  This is probably a FAQ, but since this group does not have an archive 
(I have not seen the link on RH's page if there is one), I could not 
find the info I am looking for.  No luck with Google either.


Sincerely,
Hans Deragon

Hans Deragon (imap) wrote:
>
> /dev/hdd1 /mnt/genericdata1 ext2 rw 0 0

It seems that you weren't actually waiting for your root filesystem to 
fsck, but rather this one.  If partitions were mounted when power was 
lost, then all of them get checked out, not just the root -- and fsck'ed 
if they need it.  Since this partition was mounted (and formatted) ext2, 
it needed it.

At least, that seems a likely (to me) cause of this.

HTH

On Sat, Oct 26, 2002 at 04:41:42PM -0400, Hans Deragon (imap)
wrote:
> 
>  Now, why is that when the power went off and later came back, I had to 
> go for an hour with fsck trying to fix the HD?  Isn't journaling 
> supposed to prevent this?  I had no feeling that the journal was 
> sollicitated at boot time when the system detected errors.  Is my Ext3 
> properly setup?  Is this normal?  Where resides the journal BTW?  Are 
> there config files for Ext3?
For safety reasons, an ext3 filesystem has to be fsck'd every so many
mounts, or every so many days. You can adjust the times with tune2fs, but
it is recommended to fsck them one or two times a year just to be sure
nothing has been corrupted.

-- 
-- Skylar Thompson (skylar@attglobal.net)
-- http://lizw090-016.resnet.wisc.edu/~skylar/, http://www.earlham.edu/~thompsk/

And also in /usr/src/2.4.18-14 there is the configs directory with all the
possible config files, depending on processor types and numbers (smp vs
uniproc), debug and memory amount available.
With them you can recreate the kernel suitable for what you have and want.
Bye,

Gianluca 



-----Original Message-----
From: Juri Haberland [mailto:juri@koschikode.com]
Sent: domenica 27 ottobre 2002 23.07
To: Hans Deragon
Cc: ext3-users@redhat.com
Subject: Re: How come fsck still kicks in and reports major errors with
Ext3?


Hans Deragon wrote:
> Juri Haberland wrote:
> 
>>Hans Deragon wrote:
>>
> Wait a sec... I search for this and look what I found in my logs:
[---SNIP---]
> Oct 27 13:09:09 world kernel: EXT3-fs: ide0(3,1): orphan cleanup on
readonly fs
> Oct 27 13:09:09 world kernel: EXT3-fs: ide0(3,1): 5 orphan inodes deleted
> Oct 27 13:09:09 world kernel: EXT3-fs: recovery complete.
[---SNIP---]
> So it is working after all.  Its just that yesterday, fsck kicked in and 
> I really had to fix the partition manually.  I have not noticed any ext3 
> related text on the screen when my kernel refused to continue the 
> bootstrap because it was asking me to manually fix the partition.  So my 
> guess is that journaling works 98% of the time, but once in a while 
> after an uncontrolled shutdown, my filesystem goes seriously kaput 
> anyway, right?
I highly doubt that. As Ralf wrote, there are two other possibilities:
Every ext2/ext3 filesystem is per default checked on every 20th mount or
if more than 6 months have passed since the last check. This happens
whether the filesystem was cleanly unmounted or not. So it looks like
that this is what happened to you.
>>What distribution are you running?
>>
> RedHat 8.0.  The kernel running on my system is the one installed from 
> the CDROM.  I have the source code of the kernel, but not the .config 
> file which corresponds to my running kernel.  Anybody knows if it is 
> possible to get the installed kernel .config file so I can check out the 
> configuration?  This is off-topic regarding this mailing list, so I 
> suggest anyone responding to this to reply to me personnaly.
Look in /boot/, there is a file called
'config-${VERION_OF_YOUR_KERNEL}'.

Regards,
Juri



_______________________________________________
Ext3-users mailing list
Ext3-users@redhat.com
https://listman.redhat.com/mailman/listinfo/ext3-users