"Justin Hamilton" writes:
=> Here is the xfs_logprint:
=>
=> [root@Voyager /]# xfs_logprint /dev/vg04/lv_movies
=> xfs_logprint:
=> data device: 0x3a07
=> log device: 0x3a07 daddr: 10485792 length: 9600
=>
=> Header 0xc wanted 0xfeedbabe
=> **********************************************************************
=> * ERROR: header cycle=12 block=1463 *
=> **********************************************************************
=> Bad log record header
=>
=> So somehow, it looks like the log is getting clobbered.
=>
=> Before I run the xfs_repair, is there any other info that will be useful
=> to ensuring this get fixed?
Well actually xfs_logprint without '-t' will sometimes do this because
the first block of the log doesn't always look like a valid log record
(it can be the tail end of a record that started right at the end
of the log - hence no 0xfeedbabe). So this doesn't really prove anything.
I'd have to hand this off to someone else on the list at this point -
I don't have time to follow though. But I can refer you to a comment from
Steve:
lord@xxxxxxx said:
=> There is definitely a problem with the log on raid5 right now, we did
=> not change anything on our end, so it looks like the underlying code
=> may have broken things. We are looking into this. As for performance,
=> the
Are you using raid5? if so, this might explain it. If not, it might anyway.
I've never seen this kind of symptom before.
-----------------------------------------------------
Daniel Moore dxm@xxxxxxx
R&D Software Engineer Phone: +61-3-98348209
SGI Performance Tools Group Fax: +61-3-98132378
-----------------------------------------------------
|