headers
Eric Sandeen | 1 Apr 2003 01:01
Picon
Favicon

Re: Moving XFS between sparc64 and ia32

On Mon, 2003-03-31 at 16:48, Hendrik Visage wrote:
> Hi there,
> 
> I've created an XFS on an Intel 32bit machine, and are now trying
> to mount it on a sparc64 machine. I've noticed issues with the
> pagebuff/blocksize, but how do I circumvent it?
> 
> The messages in dmesg output I'm getting is:
> 
> XFS mounting filesystem ide0(3,66)
> Starting XFS recovery on filesystem: ide0(3,66) (dev: 3/66)
> XFS: dirty log written in incompatible format - can't recover
> XFS: log mount/recovery failed
> XFS: log mount failed

From this message, the best thing to do would be to put it back in the
intel machine, and cleanly unmount it before you try to move it.

-Eric

--

-- 
Eric Sandeen      XFS for Linux     http://oss.sgi.com/projects/xfs
sandeen <at> sgi.com   SGI, Inc.         651-683-3102
Permalink | Reply | 0 comments |
headers
Austin Gonyou | 1 Apr 2003 01:13
Favicon

Re: xfs_repair of root filesystem

On Mon, 2003-03-31 at 16:59, Eric Sandeen wrote:
> hi Jeremy - It's in the faq,
> http://oss.sgi.com/projects/xfs/faq.html#xfschecks
> 
> Q: My filesystem is ok - but xfs_check or xfs_repair shows
> errors/won't
> run - whats wrong here?

I'm kinda curious about a thought I had around this. Would it be at all
possible to perform xfs_freeze on the root FS then check? Where
xfs_freeze would potentially have a special "repair" sesion or something
so you didn't have to bring the box down to do repair filesystems? 

--

-- 
Austin Gonyou <austin <at> coremetrics.com>
Coremetrics, Inc.
Permalink | Reply | 1 comment |
headers
Hendrik Visage | 1 Apr 2003 01:24
Picon

Re: Moving XFS between sparc64 and ia32

On Mon, Mar 31, 2003 at 04:58:02PM -0600, Steve Lord wrote:
> 
> You are moving between little endian and big endian machines. While
> metadata is always big endian in xfs, the log contents are not. 
> If you are sure you did a clean unmount on ia32, try running
> xfs_repair -L on the filesystem. But if you can I would
> mount and unmount it on ia32 just in case before this.

Thanx.
Will try that tomorow.
Permalink | Reply | 0 comments |
headers
Steve Lord | 1 Apr 2003 01:24
Picon
Favicon

Re: xfs_repair of root filesystem

On Mon, 2003-03-31 at 17:13, Austin Gonyou wrote:
> On Mon, 2003-03-31 at 16:59, Eric Sandeen wrote:
> > hi Jeremy - It's in the faq,
> > http://oss.sgi.com/projects/xfs/faq.html#xfschecks
> > 
> > Q: My filesystem is ok - but xfs_check or xfs_repair shows
> > errors/won't
> > run - whats wrong here?
> 
> I'm kinda curious about a thought I had around this. Would it be at all
> possible to perform xfs_freeze on the root FS then check? Where
> xfs_freeze would potentially have a special "repair" sesion or something
> so you didn't have to bring the box down to do repair filesystems? 

The real answer is to fix up the user space libraries to allow
repair to run on a mounted fs and issue a big fat warning to
reboot NOW at the end.

Steve

--

-- 

Steve Lord                                      voice: +1-651-683-3511
Principal Engineer, Filesystem Software         email: lord <at> sgi.com
Permalink | Reply | 0 comments |
headers
kishor bhagwat | 1 Apr 2003 06:22
Picon
Favicon

Re: LVM + XFS + Quotas


--- Nathan Scott <nathans <at> sgi.com> wrote:
> On Fri, Mar 28, 2003 at 07:05:36PM -0800, kishor bhagwat
> wrote:
> > 
> > --- Chris Wedgwood <cw <at> f00f.org> wrote:
> > > On Fri, Mar 28, 2003 at 03:27:51AM -0800, kishor
> bhagwat
> > > wrote:
> > > 
> > > > I'm using RH 8.0 with kernel upgraded to
> kernel.org's
> > > 2.4.19, and
> > > > XFS 1.2 patches.
> > > 
> > > > I'm able to create quotas etc..but there is an
> issue
> > > with the actual
> > > > usable space.
> > > 
> > > what does 'quota' report?
> > 
> > quota reports.....
> >  Filesystem  /dev/HOME/DataVol
> > blocks quota   limit   grace   files   quota   limit 
> grace
> >   32      64      80              12       0       0   
>     
> > 
>
(Continue reading)

Permalink | Reply | 1 comment |
headers
Seth Mos | 1 Apr 2003 10:33
Picon
Picon
Favicon

Re: Moving XFS between sparc64 and ia32

At 00:48 1-4-2003 +0200, Hendrik Visage wrote:
>Hi there,
>
>I've created an XFS on an Intel 32bit machine, and are now trying
>to mount it on a sparc64 machine. I've noticed issues with the
>pagebuff/blocksize, but how do I circumvent it?

What pagesize does the sparc64 have? IIRC you can specify the blocksize to 
use on mount time.

Recovery might be even more tricky then mounting a clean filesystem.

>The messages in dmesg output I'm getting is:
>
>XFS mounting filesystem ide0(3,66)
>Starting XFS recovery on filesystem: ide0(3,66) (dev: 3/66)
>XFS: dirty log written in incompatible format - can't recover
>XFS: log mount/recovery failed
>XFS: log mount failed

Cheers

--
Seth
It might just be your lucky day, if you only knew.
Permalink | Reply | 1 comment |
headers
bugzilla-daemon | 1 Apr 2003 11:51
Picon
Favicon

[Bug 230] umount hangs after high disk load

http://oss.sgi.com/bugzilla/show_bug.cgi?id=230

------- Additional Comments From atu <at> dmeti.dp.ua  2003-04-01 01:51 -------
Sorry, but no functions like *get_req*  in my trace. 
What else can I do? 

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
Permalink | Reply | 0 comments |
headers
Christian Guggenberger | 1 Apr 2003 13:19
Picon
Favicon

kdb and rqueue - how to do it ? (Bug 230)

Hi all!

Russel suggested in Bug #230 to dump the rqueue of get_request_wait in kdb.
But how can this be done?

[0]kdb> rqueue
Unknown kdb command: 'rqueue
'

thanks for any advice.

Christian
Permalink | Reply | 1 comment |
headers
Michael Sinz | 1 Apr 2003 13:48

Re: xfs_repair of root filesystem

Jeremy Jackson wrote:
> Hi,
> 
> I'm wonder what's the official word about xfs_repair on a read-only
> mounted fs.  The utilities complain for me, so I have to boot from a
> repair partition to fix XFS (a while back when the shutdown files in use
> bug was still a problem).  Ext2 has no problem with this.  I'd just to
> know for future reference, so I know if I have to have a spare root fs 
> or not.

While I too would like to have a way to repair XFS read-only mounts,
there are other reasons to have a special recovery partition for
this.

What I have done is to make /boot its own partition at the start of
the disk.  This is where the kernel lives, along with lilo stuff (or
grub if you use that)

Anyway, I have made a script that will build a mini-boot system in
that partition and that will then run as the init process to fix up
any other filesystems.  To help reduce the chance that /boot is
corrupted, I mount it read-only and, since nothing normally runs from
it, if it ever needs fixing, I can just unmount it and fix it after
a regular boot.

The install script (see attached) builds everything needed and even
tells you how much space it used in /boot to do its work.  You may need
to add lilo/grub entries that match your environment plus, as currently
written, it assumes a devfs kernel (someone want to make a different
version?)
(Continue reading)

Permalink | Reply | 2 comments |
headers
Christian Guggenberger | 1 Apr 2003 14:27
Picon
Favicon

Re: kdb and rqueue - how to do it ? (Bug 230)

On 01.04.2003   13:19 Christian Guggenberger wrote:
> Hi all!
> 
> Russel suggested in Bug #230 to dump the rqueue of get_request_wait in kdb.
> But how can this be done?
> 
> [0]kdb> rqueue
> Unknown kdb command: 'rqueue'

arrgh... CONFIG_KDB_MODULES was not set...
rqueue dump is on the way.

btw. the help on that option is a little bit misleading.
Permalink | Reply | 0 comments |

Gmane