On 07/27/2011 01:24 AM, Gregory Farnum wrote:
Hmm, I'm afraid I can't see anything useful in these logs right now --
just confusion!
The only thing I can think of that might provide useful information
now is to turn off osd5, update its config file to have heavy
debugging information (debug osd = 20, at a minimum), and then start
the OSD back up again. Then we can see if pg 0.256 is still having any
issues (it is in the logs you sent me).
If not I'll have to mark this as "Can't reproduce" and hope we see it
again from somebody with higher log levels on. :(
-Greg
I started manual scrubbing on osd5 and got next message:
2011-07-27 11:38:30.360812 log 2011-07-27 11:38:28.540559 osd5
10.5.51.145:6803/19737 2842 : [ERR] 0.256 scrub stat mismatch, got
2196/2196 objects, 0/0 clones, 9046200695/9046200817 bytes,
8834197/8834197 kb.
2011-07-27 11:38:30.360812 log 2011-07-27 11:38:28.540571 osd5
10.5.51.145:6803/19737 2843 : [ERR] 0.256 scrub 1 errors
"Hmmm" I said to myself.
stop osd5
fsck -fy /dev/sdb1
And got:
Inode 230558734, i_blocks is 8, should be 16. Fix? yes
Inode 230559279, i_blocks is 8208, should be 8200. Fix? yes
Inode 230559410, i_blocks is 40, should be 32. Fix? yes
Inode 230559527, i_blocks is 24, should be 32. Fix? yes
Inode 231342174, i_blocks is 40, should be 32. Fix? yes
Inode 231342209, i_blocks is 24, should be 32. Fix? yes
Ups. Familiar message. :(
I login to different node and got the same.
Greg, I think to catch any errors no sense. 2.6.39 have broken ext4. I
will try 3.0.0 (or 3.1.0) and tell about result.
WBR,
Fyodor.
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
