vs-5150: search_by_key: invalid format found in block 30690. Fsck?

vs-5150: search_by_key: invalid format found in block 30690. Fsck?

[Vadim]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I just noticed quite a lot of those:

vs-5150: search_by_key: invalid format found in block 9170. Fsck?
vs-13070: reiserfs_read_inode2: i/o failure occurred trying to 
find stat data of [3614 3618 0x0 SD]
is_tree_node: node level 65448 does not match to the expected one 
1
vs-5150: search_by_key: invalid format found in block 14568. 
Fsck?
vs-13070: reiserfs_read_inode2: i/o failure occurred trying to 
find stat data of [8839 12110 0x0 SD]
is_tree_node: node level 17741 does not match to the expected one 
1
vs-5150: search_by_key: invalid format found in block 26324. 
Fsck?
vs-13070: reiserfs_read_inode2: i/o failure occurred trying to 
find stat data of [32635 32751 0x0 SD]
is_tree_node: node level 0 does not match to the expected one 1
vs-5150: search_by_key: invalid format found in block 30690. 
Fsck?
vs-13070: reiserfs_read_inode2: i/o failure occurred trying to 
find stat data of [39269 39617 0x0 SD]
is_tree_node: node level 0 does not match to the expected one 1
vs-5150: search_by_key: invalid format found in block 30690. 
Fsck?
vs-13070: reiserfs_read_inode2: i/o failure occurred trying to 
find stat data of [39269 39617 0x0 SD]
is_tree_node: node level 0 does not match to the expected one 1
vs-5150: search_by_key: invalid format found in block 30690. 
Fsck?
vs-13070: reiserfs_read_inode2: i/o failure occurred trying to 
find stat data of [39269 39617 0x0 SD]
is_tree_node: node level 0 does not match to the expected one 1
vs-5150: search_by_key: invalid format found in block 30690. 
Fsck?
vs-13070: reiserfs_read_inode2: i/o failure occurred trying to 
find stat data of [39269 39617 0x0 SD]

The disk seems to be good, no errors, kernel is 2.4.21 with LVM 
1.0.7.

Why does this error happen? Is it a bug fixed in a later version?
BTW, it'd be nice if it told me what partition it happened on.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)

iD8DBQE/PRq2vCkUtBccqkoRAnCPAJ9+BXblT9ML5ljIgfKf3Bp1mZ9E9wCgohN+
3NMCsV5AMbNKnuU8S9HAgL8=
=Gk0C
-----END PGP SIGNATURE-----

Re: vs-5150: search_by_key: invalid format found in block 30690. Fsck?

[Oleg Drokin]

Hello!

On Fri, Aug 15, 2003 at 07:39:02PM +0200, Vadim wrote:
> I just noticed quite a lot of those:
> vs-5150: search_by_key: invalid format found in block 9170. Fsck?
> vs-13070: reiserfs_read_inode2: i/o failure occurred trying to 
> find stat data of [3614 3618 0x0 SD]
> is_tree_node: node level 65448 does not match to the expected one 
> 1
> vs-5150: search_by_key: invalid format found in block 14568. 
> Fsck?

Some of the blocks are got filled with garbage for some (unknown) reason.

> The disk seems to be good, no errors, kernel is 2.4.21 with LVM 
> 1.0.7.
> Why does this error happen? Is it a bug fixed in a later version?

Well, I am not sure there is a bug.
What was happenning before you started to get those messsages? Were there any
unexpected shutdowns and you do not have write cache disabled in your harddrives
or something like that?

> BTW, it'd be nice if it told me what partition it happened on.

We merged a patch that does just this in 2.4.22-pre3

Bye,
    Oleg

Re: vs-5150: search_by_key: invalid format found in block 30690. Fsck?

[Vadim]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Monday 18 August 2003 08:08, Oleg Drokin wrote:
> Hello!
>
> On Fri, Aug 15, 2003 at 07:39:02PM +0200, Vadim wrote:
> > I just noticed quite a lot of those:
> > vs-5150: search_by_key: invalid format found in block 9170.
> > Fsck? vs-13070: reiserfs_read_inode2: i/o failure occurred
> > trying to find stat data of [3614 3618 0x0 SD]
> > is_tree_node: node level 65448 does not match to the
> > expected one 1
> > vs-5150: search_by_key: invalid format found in block 14568.
> > Fsck?
>
> Some of the blocks are got filled with garbage for some
> (unknown) reason.
>
> > The disk seems to be good, no errors, kernel is 2.4.21 with
> > LVM 1.0.7.
> > Why does this error happen? Is it a bug fixed in a later
> > version?
>
> Well, I am not sure there is a bug.
> What was happenning before you started to get those messsages?
> Were there any unexpected shutdowns and you do not have write
> cache disabled in your harddrives or something like that?
Yeah, there was a quite bad crash once. I tried using a Gentoo 
kernel which seemed to have too many patches. Crashed quite 
horribly, with lots of oopses scrolling by the screen. I ran 
reiserfsck 3.6.11, and it found no errors, though.

Now, write cache sounds like a possibility. hdparm says:
        Enabled Supported:
           *    NOP cmd
           *    READ BUFFER cmd
           *    WRITE BUFFER cmd
           *    Host Protected Area feature set
           *    Look-ahead
           *    Write cache
           *    Power Management feature set
                Security Mode feature set
           *    SMART feature set
           *    Mandatory FLUSH CACHE command
           *    Automatic Acoustic Management feature set
                SET MAX security extension
           *    DOWNLOAD MICROCODE cmd

Excuse me if this is a stupid question, but hdparm doesn't seem 
to have any arguments to disable write cache. How do I do that?

I did know ReiserFS is supposed to be used without write cache, 
but for some reason I thought that my old disk probably wouldn't 
have one. Perhaps ReiserFS should automatically disable write 
cache on the disks it's used on?

> We merged a patch that does just this in 2.4.22-pre3
Ok, thanks

>
> Bye,
>     Oleg

Thanks for the help :-)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)

iD8DBQE/QIeWvCkUtBccqkoRAvFZAJ4s7lNIjVXDnxdUeUdIjH3IE+FoVACgtdTe
nGYed21uf5zXCO309xa/0lU=
=EWmq
-----END PGP SIGNATURE-----

Re: vs-5150: search_by_key: invalid format found in block 30690. Fsck?

[Oleg Drokin]

Hello!

On Mon, Aug 18, 2003 at 10:00:16AM +0200, Vadim wrote:

> Excuse me if this is a stupid question, but hdparm doesn't seem 
> to have any arguments to disable write cache. How do I do that?

hdparm -W0 will do that.

> I did know ReiserFS is supposed to be used without write cache, 
> but for some reason I thought that my old disk probably wouldn't 
> have one. Perhaps ReiserFS should automatically disable write 
> cache on the disks it's used on?

This is not reiserfs' business.
But the solution called IDE barriers is going to be implemented soon (at least
in 2.6, there is standalone patch that does this in SuSE kernels, too)

Bye,
    Oleg

Re: vs-5150: search_by_key: invalid format found in block 30690. Fsck?

[Vadim]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Monday 18 August 2003 15:35, Oleg Drokin wrote:
> Hello!
>
> On Mon, Aug 18, 2003 at 10:00:16AM +0200, Vadim wrote:
> > Excuse me if this is a stupid question, but hdparm doesn't
> > seem to have any arguments to disable write cache. How do I
> > do that?
>
> hdparm -W0 will do that.
>
*bangs head* It is indeed there... sorry.

> This is not reiserfs' business.
> But the solution called IDE barriers is going to be
> implemented soon (at least in 2.6, there is standalone patch
> that does this in SuSE kernels, too)
Interesting, I'll take a look at that

>
> Bye,
>     Oleg

Thanks a lot for the help, and sorry for the silly questions
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2 (GNU/Linux)

iD8DBQE/QQyvvCkUtBccqkoRAjTKAJ4zs0eSmIHPkJwxy9WOrh9BrHEJgACeL3ZX
9jRl/8cn1dfEsjmTE2k66HI=
=DHis
-----END PGP SIGNATURE-----