OpenVZ Forum


Home » General » Support » Ploop problems with md/RAID10
Ploop problems with md/RAID10 [message #49945] Wed, 10 July 2013 09:29 Go to next message
miskano is currently offline  miskano
Messages: 3
Registered: July 2013
Junior Member
Hi,

after converting some containers to ploop, I noticed the following errors in dmesg:

....
[10416544.911982] JBD2: Detected IO errors while flushing file data on ploop34745p1-8
[10416545.501406] JBD2: Detected IO errors while flushing file data on ploop55626p1-8
[10416545.672716] JBD2: Detected IO errors while flushing file data on ploop55626p1-8
[10416545.749478] JBD2: Detected IO errors while flushing file data on ploop55626p1-8
[10416545.875440] JBD2: Detected IO errors while flushing file data on ploop55626p1-8
[10416545.959033] JBD2: Detected IO errors while flushing file data on ploop55626p1-8
[10416546.087430] JBD2: Detected IO errors while flushing file data on ploop55626p1-8
[10416550.840862] JBD2: Detected IO errors while flushing file data on ploop34745p1-8
[10416551.834908] JBD2: Detected IO errors while flushing file data on ploop55626p1-8
[10416576.164470] JBD2: Detected IO errors while flushing file data on ploop34745p1-8
[10416603.904074] JBD2: Detected IO errors while flushing file data on ploop34745p1-8
[10416604.180471] JBD2: Detected IO errors while flushing file data on ploop34745p1-8
[10416604.885114] md/raid10:md1: make_request bug: can't convert block across chunks or bigger than 256k 2493927408 16
[10416608.887892] md/raid10:md1: make_request bug: can't convert block across chunks or bigger than 256k 2490152912 32

...

Other containers work w/o any issues. I tried upgrading kernel to latest stable 2.6.32-042stab078.27, which didn't help at all. Centos 6.4 64bit.

If I mount the ploop file and fsck it, no errors are found.

Disks seem fine nad no problems otherwise with them.

Re: Ploop problems with md/RAID10 [message #49946 is a reply to message #49945] Wed, 10 July 2013 17:17 Go to previous messageGo to next message
miskano is currently offline  miskano
Messages: 3
Registered: July 2013
Junior Member
More different errors, all related to ploop:

[26595.153880] EXT4-fs warning (device ploop25117p1): ext4_end_bio: I/O error writing to inode 1842090 (size 8192 starting block 5046991)
[26595.154033] Buffer I/O error on device ploop25117p1, logical block 5046735
[26595.154107] lost page write due to I/O error on ploop25117p1
[26595.154180] Buffer I/O error on device ploop25117p1, logical block 5046736
[26595.154255] lost page write due to I/O error on ploop25117p1

Re: Ploop problems with md/RAID10 [message #49947 is a reply to message #49945] Wed, 10 July 2013 17:39 Go to previous messageGo to next message
miskano is currently offline  miskano
Messages: 3
Registered: July 2013
Junior Member
Only thing peculiar about the containers in question is they contain a folder with 250k+ small files (mail files from cron). Don't know if this could be an issue.

I'm now converting them back to simfs.
Re: Ploop problems with md/RAID10 [message #49952 is a reply to message #49945] Thu, 11 July 2013 07:11 Go to previous message
poiuty is currently offline  poiuty
Messages: 16
Registered: September 2011
Junior Member

Please, report in https://bugzilla.openvz.org/
Previous Topic: Process start time incorrect inside VE
Next Topic: Flushing iptables does not work.
Goto Forum:
  


Current Time: Fri Nov 08 22:52:34 GMT 2024

Total time taken to generate the page: 0.03303 seconds