reiserfs - problems mounting after power outage

reiserfs - problems mounting after power outage

[Jeff McWilliams]

I'm having difficulty mounting a reiserfs partition after a power outage.

This is 2.4.0-test9 compiled with reiserfs as a module, and 
the ide.2.4.0-t9-6.task.0923.path IDE patch - mostly to get updated support for
the 3WARE IDE RAID controller.  

/dev/sda is the 3ware escalade raid mirror - two Maxtor 20 gig drives.

reiserfs is compiled as a module, the distribution is Debian Linux 
"Potato"

uname -a shows:
Linux ns2 2.4.0-test9 #1 Wed Dec 6 16:28:45 EST 2000 i586 unknown

/etc/fstab shows:

# /etc/fstab: static file system information.
#
# <file system> <mount point>   <type>  <options>                       <dump> <pass>
/dev/sda1       /               ext2    defaults,errors=remount-ro      0      1
/dev/sda5       none            swap    sw                      0       0
proc            /proc           proc    defaults                        0      0
/dev/sda6       /var/www        reiserfs defaults 0 0
/dev/sda7       /home           reiserfs defaults,noauto 0 0
/dev/fd0        /floppy         auto    defaults,user,noauto            0      0
/dev/cdrom      /cdrom          iso9660 defaults,ro,user,noauto         0      0


The partition I'm having trouble with is /dev/sda7.  /dev/sda6 recovered okay.
/dev/sda7 doesn't have a lot of data on it, and I CAN deal with the lost data,
but it doesn't leave me with a very good feeling of confidence in reiserfs
if I can't successfully recover from a power failure.


What happens when I type mount /dev/sda7 is this:

reiserfs: using 3.5.x disk format
reiserfs: checking transaction log (device 08:07) ...

and then it hangs.  I then have to hit the reset button to reboot.

I've tried running reiserfsck with --check, --correct-bitmap, 
--rebuild-sb, and --rebuild-tree.  NO luck there.

debugreiserfs /dev/sda7 shows:

<-------------debugreiserfs, 2000------------->
reiserfsprogs 3.x
Super block of format 3.5 found on the 0x3 in block 16
Block count 2405720
Blocksize 4096
Free blocks 2397152
Busy blocks (skipped 16, bitmaps - 74, journal blocks - 8193
1 super blocks, 284 data blocks
Root block 8212
Journal block (first) 18
Journal dev 0
Journal orig size 8192
Filesystem state ERROR
Tree height 2
Hash function used to sort names: tea hash

Any help?  Please cc Jeff.McWilliams@acetechnologies.net

Thanks,

Jeff McWilliams
 
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: reiserfs - problems mounting after power outage

[Chris Mason]

On Wednesday, February 07, 2001 12:31:43 PM -0500 Jeff McWilliams <Jeff.McWilliams@acetechnologies.net> wrote:

> I'm having difficulty mounting a reiserfs partition after a power outage.
> 
> This is 2.4.0-test9 compiled with reiserfs as a module, and 

Which reiserfs version is this?  Upgrading to the reiserfs included in 2.4.1 would be a good plan, there have been a few bug fixes since test9 times (none specfically related to this).

> the ide.2.4.0-t9-6.task.0923.path IDE patch - mostly to get updated support for
> the 3WARE IDE RAID controller.  
> 

If this raid controller has writeback cache, make sure you either have a battery backup for the cache, or have writeback turned off.

> /dev/sda is the 3ware escalade raid mirror - two Maxtor 20 gig drives.
> 
> reiserfs is compiled as a module, the distribution is Debian Linux 
> "Potato"
> 
[ ... ]

> The partition I'm having trouble with is /dev/sda7.  /dev/sda6 recovered okay.
> /dev/sda7 doesn't have a lot of data on it, and I CAN deal with the lost data,
> but it doesn't leave me with a very good feeling of confidence in reiserfs
> if I can't successfully recover from a power failure.
> 
> 
> What happens when I type mount /dev/sda7 is this:
> 
> reiserfs: using 3.5.x disk format
> reiserfs: checking transaction log (device 08:07) ...
> 
> and then it hangs.  I then have to hit the reset button to reboot.
> 

How hung is it?  Does the whole system stop, or does this mount just block forever?

> I've tried running reiserfsck with --check, --correct-bitmap, 
> --rebuild-sb, and --rebuild-tree.  NO luck there.
> 

These programs also hang?  Does reiserfsck (which version BTW) give any output at all?

-chris


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: reiserfs - problems mounting after power outage

[Jeff McWilliams]

Chris,

Thanks for the reply.

> Which reiserfs version is this?  Upgrading to the reiserfs included in
> 2.4.1 would be a good plan, there have been a few bug fixes since test9
> times (none specfically related to this).
>

/var/log/messages shows ReiserFS reporting version 3.6.18.

> If this raid controller has writeback cache, make sure you either have a
> battery backup for the cache, or have writeback turned off.
>

It has no battery backup for the cache.  Writeback cacheing is disabled.

> How hung is it?  Does the whole system stop, or does this mount just
> block forever?
>

When the mount hangs the rest of the system is responsive.  I can to go other
virtual consoles and do other things and it works fine. If I attempt to shut
down the system (shutdown -h now) the kernel blocks trying to unmount local
filesystems just before powering down.


The reiserfs tools I used are from Debian's reiserfsprogs_3.0.20001019-3.deb

> These programs also hang?  Does reiserfsck (which version BTW) give any
> output at all?
>
>

If I do a reiserfsck /dev/sda7 I get the following:

ns2:/usr/share/doc/reiserfsprogs# reiserfsck --rebuild-tree /dev/sda7

<-------------reiserfsck, 2000------------->
reiserfsprogs 3.x

This is an experimental version of reiserfsck, MAKE A BACKUP FIRST!
Don't run this program unless something is broken.  You may want
to backup first.  Some types of random FS damage can be recovered
from by this program, which basically throws away the internal nodes
of the tree and then reconstructs them.  This program is for use only
by the desperate, and is of only beta quality.  Email
reiserfs@devlinux.com with bug reports.
Will replay just like mounting would
Do you want to run this program?[N/Yes] (note need to type Yes):Yes
Replaying log..
Looking for the oldest transaction to start with  ok
1 valid trans found. Will replay from 50 to 50
Replaying transaction..   0 left..


How long should I wait for this to run?  It's an 8 gig partition with perhaps
a few megs of data on it.  I've let this run for 20 minutes or so (the system is
a Pentium 200MMX with 64M of RAM, console mode only, no XFree86 stuff), and
it just runs, consuming 97% CPU according to top. /proc/loadavg reports
1.00 0.97 0.71 2/29 378

I'm interested in building 2.4.1 but I haven't seen a released one yet and
haven't done an Alan Cox patch series kernel build before.  Maybe it's time to
try.  :-)

That might have to wait until this weekend though.  I'm a developer by day,
and a sys admin by night or on the weekends.  I can also grab and build the
latest toolset instead of using the Debian pre-packaged ones if you think that
this will help.

Jeff


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: reiserfs - problems mounting after power outage

[Andre Hedrick]

Dream on if you think "Writeback cacheing is disabled" on all systems bye
default.  If you did not wack it.  It is not off.

On Wed, 7 Feb 2001, Jeff McWilliams wrote:

> Chris,
> 
> Thanks for the reply.
> 
> > Which reiserfs version is this?  Upgrading to the reiserfs included in
> > 2.4.1 would be a good plan, there have been a few bug fixes since test9
> > times (none specfically related to this).
> >
> 
> /var/log/messages shows ReiserFS reporting version 3.6.18.
> 
> > If this raid controller has writeback cache, make sure you either have a
> > battery backup for the cache, or have writeback turned off.
> >
> 
> It has no battery backup for the cache.  Writeback cacheing is disabled.
> 
> > How hung is it?  Does the whole system stop, or does this mount just
> > block forever?
> >
> 
> When the mount hangs the rest of the system is responsive.  I can to go other
> virtual consoles and do other things and it works fine. If I attempt to shut
> down the system (shutdown -h now) the kernel blocks trying to unmount local
> filesystems just before powering down.
> 
> 
> The reiserfs tools I used are from Debian's reiserfsprogs_3.0.20001019-3.deb
> 
> > These programs also hang?  Does reiserfsck (which version BTW) give any
> > output at all?
> >
> >
> 
> If I do a reiserfsck /dev/sda7 I get the following:
> 
> ns2:/usr/share/doc/reiserfsprogs# reiserfsck --rebuild-tree /dev/sda7
> 
> <-------------reiserfsck, 2000------------->
> reiserfsprogs 3.x
> 
> This is an experimental version of reiserfsck, MAKE A BACKUP FIRST!
> Don't run this program unless something is broken.  You may want
> to backup first.  Some types of random FS damage can be recovered
> from by this program, which basically throws away the internal nodes
> of the tree and then reconstructs them.  This program is for use only
> by the desperate, and is of only beta quality.  Email
> reiserfs@devlinux.com with bug reports.
> Will replay just like mounting would
> Do you want to run this program?[N/Yes] (note need to type Yes):Yes
> Replaying log..
> Looking for the oldest transaction to start with  ok
> 1 valid trans found. Will replay from 50 to 50
> Replaying transaction..   0 left..
> 
> 
> How long should I wait for this to run?  It's an 8 gig partition with perhaps
> a few megs of data on it.  I've let this run for 20 minutes or so (the system is
> a Pentium 200MMX with 64M of RAM, console mode only, no XFree86 stuff), and
> it just runs, consuming 97% CPU according to top. /proc/loadavg reports
> 1.00 0.97 0.71 2/29 378
> 
> I'm interested in building 2.4.1 but I haven't seen a released one yet and
> haven't done an Alan Cox patch series kernel build before.  Maybe it's time to
> try.  :-)
> 
> That might have to wait until this weekend though.  I'm a developer by day,
> and a sys admin by night or on the weekends.  I can also grab and build the
> latest toolset instead of using the Debian pre-packaged ones if you think that
> this will help.
> 
> Jeff
> 
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> Please read the FAQ at http://www.tux.org/lkml/
> 

Andre Hedrick
Linux ATA Development
ASL Kernel Development
-----------------------------------------------------------------------------
ASL, Inc.                                     Toll free: 1-877-ASL-3535
1757 Houret Court                             Fax: 1-408-941-2071
Milpitas, CA 95035                            Web: www.aslab.com

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: reiserfs - problems mounting after power outage

[Jeff McWilliams]

Quoting Andre Hedrick <andre@linux-ide.org>:

>
> Dream on if you think "Writeback cacheing is disabled" on all systems
> bye
> default.  If you did not wack it.  It is not off.
>
Yeah, I know, Andre.  I turned it off in the 3ware card's BIOS setup.
I'm not completely idiotic, just dumb enough to not have ordered a UPS for this
box sooner.  One is on the way now though.  :-)

Jeff

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

Re: reiserfs - problems mounting after power outage (FIXED)

[Jeff McWilliams]

Problem solved.  I downloaded the latest reiserfs_utils 3.6.25 from
namesys.com and built and ran reiserfsck against the partition with the
--rebuild-tree option.  It completed successfully and I was able to mount
the partition without any further problems.

I'll build and install a 2.4.1 kernel sometime tomorrow and get it running on
that machine to address any further issues.

many thanks for the tips & suggestions

Jeff

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/