reiserfsck failing with external journal

reiserfsck failing with external journal

[Ronny T. Lampert]

Hi,

I've made a fs with the newest reiserfs-progs.

#> mkreiserfs -j /dev/hda5 /dev/hda6

Mounting is fine.
But reiserfsck fails every time with:

-o-o-

###########
reiserfsck --check started at Thu Sep 30 15:42:47 2004
###########

reiserfs_open_journal: journal parameters from the super block does not match
to journal parameters from the journal. You should run  reiserfsck with 
--rebuild-sb to check your superblock consistency.

-o-o-

If I do as adviced (reiserfsck --rebuild-sb), it says the superblock is fine.
Next time it fails with the same message.

Only supplying -j /dev/hda5 to every reiserfs-prog will help :(
Doing an

#> reiserfstune --new-journal-dev /dev/hda5

doesn't cure it, too.

I think, that is not as it's supposed to work because it'll _break all system 
boot-up scripts_ !

Is it some subtle bug in the reiserfs-progs, because they try to extract the 
journal_dev from the SB (as a quick look at the source reveals) but don't use 
it properly?

Kind regards,
Ronny

Re: reiserfsck failing with external journal

[Edward Shishkin]

Ronny T. Lampert wrote:

> Hi,
>
> I've made a fs with the newest reiserfs-progs.
>
> #> mkreiserfs -j /dev/hda5 /dev/hda6
>
> Mounting is fine.
> But reiserfsck fails every time with:
>
> -o-o-
>
> ###########
> reiserfsck --check started at Thu Sep 30 15:42:47 2004
> ###########
>
> reiserfs_open_journal: journal parameters from the super block does 
> not match
> to journal parameters from the journal. You should run  reiserfsck 
> with --rebuild-sb to check your superblock consistency.


Hello.
It is essential to specify right journal device for fsck, although I am 
not sure
 if this warning is correct. As I remember, fsck should ask for a 
correct journal
device or suggest to use the option --no-journal-available (then you'll 
be invited
to assign new journal device by reiserfstune). Vitaly?..

>
> -o-o-
>
> If I do as adviced (reiserfsck --rebuild-sb), it says the superblock 
> is fine.
> Next time it fails with the same message.
>
> Only supplying -j /dev/hda5 to every reiserfs-prog will help :(
> Doing an
>
> #> reiserfstune --new-journal-dev /dev/hda5
>
> doesn't cure it, too.
>
> I think, that is not as it's supposed to work because it'll _break all 
> system boot-up scripts_ !


Let's first take a look at the scripts that will be broken ;)
Edward.

>
> Is it some subtle bug in the reiserfs-progs, because they try to 
> extract the journal_dev from the SB (as a quick look at the source 
> reveals) but don't use it properly?
>
> Kind regards,
> Ronny
>
>

Re: reiserfsck failing with external journal

[Ronny T. Lampert]

Hi,

thanks for your fast answer.

> It is essential to specify right journal device for fsck, although I am 
I am not 100% sure it should be essential.

Eg. jfs_fsck (and all jfs-utils) extracts the journal-dev from the SB (if 
external) and go on checking without any manual intervention, no matter where 
the journal is.
Only if the external journal is broken or somesuch, it will yell.

> Let's first take a look at the scripts that will be broken ;)
My RH9 scripts simply call "fsck -T -R -A -a $fsckoptions" (in 
/etc/rc.d/rc.sysinit).

YOS 2.1.X (which happens to be my test-system) also fails at boot-up dropping 
me into

*** An error occurred during the file system check.
*** Dropping you to a shell; the system will reboot
*** when you leave the shell.

when entering that volume into fstab.

Kind regards,
Ronny

Re: reiserfsck failing with external journal

[Vitaly Fertman]

On Thursday 30 September 2004 18:52, Edward Shishkin wrote:
> Ronny T. Lampert wrote:
> > Hi,
> >
> > I've made a fs with the newest reiserfs-progs.
> >
> > #> mkreiserfs -j /dev/hda5 /dev/hda6
> >
> > Mounting is fine.
> > But reiserfsck fails every time with:
> >
> > -o-o-
> >
> > ###########
> > reiserfsck --check started at Thu Sep 30 15:42:47 2004
> > ###########
> >
> > reiserfs_open_journal: journal parameters from the super block does
> > not match
> > to journal parameters from the journal. You should run  reiserfsck
> > with --rebuild-sb to check your superblock consistency.
>
> Hello.
> It is essential to specify right journal device for fsck, although I am
> not sure  if this warning is correct. As I remember, fsck should ask for a
> correct journal device or suggest to use the option --no-journal-available 
> (then you'll be invited to assign new journal device by reiserfstune). Vitaly?..

yes, these messages should be corrected.

> > -o-o-
> >
> > If I do as adviced (reiserfsck --rebuild-sb), it says the superblock
> > is fine.
> > Next time it fails with the same message.
> >
> > Only supplying -j /dev/hda5 to every reiserfs-prog will help :(
> > Doing an
> >
> > #> reiserfstune --new-journal-dev /dev/hda5
> >
> > doesn't cure it, too.
> >
> > I think, that is not as it's supposed to work because it'll _break all
> > system boot-up scripts_ !
>
> Let's first take a look at the scripts that will be broken ;)
> Edward.
>
> > Is it some subtle bug in the reiserfs-progs, because they try to
> > extract the journal_dev from the SB (as a quick look at the source
> > reveals) but don't use it properly?

Edward is right that currently relocated journal needs to be specified 
explicitly for reiserfsck. would be not bad to not require this though.

-- 
Thanks,
Vitaly Fertman