They are on this list somewhere. I've sent several over time. This has
been going on for a long while, and is very reproduceable. I'd love for
someone work with me on this to get this fixed, I'm sure I could aid
since I have a setup which does it consistently, no matter the XFS
version, but the log-replay is always affected.
On Mon, 2003-10-27 at 18:38, Glen Overby wrote:
> On October 27, Austin Gonyou wrote:
> > The issue is that I don't see any "ERROR" statements in this output,
> but
> > trying to replay the log or mounting -oro will oops the kernel and
> that
> > one device is then unavailable and the system must be rebooted.
>
> Can you send the exact oops message.. where it occured, etc?
>
> > I would like to extract the log so I can put this server back into our
> > beta environment. If someone could aid me in this I'd much appreciate
> > it.
>
> Here's an example:
>
> [root@HOSTNAME root]# /a23/overby/bin.l/mkfs.xfs -f -b size=4096 -l
> size=8192b /dev/hda6
>
> meta-data=/dev/hda6 isize=256 agcount=8, agsize=140317
> blks
> = sectsz=512
> data = bsize=4096 blocks=1122534, imaxpct=25
> = sunit=0 swidth=0 blks, unwritten=1
> naming =version 2 bsize=4096
> log =internal log bsize=4096 blocks=8192, version=1
> = sectsz=512 sunit=0 blks
> realtime =none extsz=65536 blocks=0, rtextents=0
>
> [root@HOSTNAME root]# /a23/overby/bin.l/xfs_db /dev/hda6
> xfs_db> sb 0
> xfs_db> print logstart
> logstart = 1048580
> xfs_db> print logblocks
> logblocks = 8192
> xfs_db> convert fsb 1048580 daddr
> 0x4483c0 (4490176)
> xfs_db> convert fsb 8192 daddr
> 0x10000 (65536)
>
> [root@HOSTNAME root]# dd if=/dev/hda6 of=/tmp/hda6log bs=512
> skip=4490176 count=65536
> 65536+0 records in
> 65536+0 records out
>
> Glen Overby
--
Austin Gonyou <austin@xxxxxxxxxxxxxxx>
Coremetrics, Inc.
|