Skip to contentSkip to conversation

software raid 5 broken on 2.6.14.4 sparc

DISCUSSION Robert Hancock 2 participants 2 messages Dec 29, 2005 1 message not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw
Quoted 3 lines · Gerhard Mack

Hello,

I had one of 5 drives fail in my raid 5 setup and now many files are unreadable.. Isn't raid 5 supposed to compensate for exactly this happening?

It looks like you have multiple drives with problems here:

Here's sde unhappy:

Quoted 2 lines
Dec 25 18:36:27 localhost kernel: sde: Current: sense key: Recovered Error
Dec 25 18:36:27 localhost kernel:     Additional sense: Recovered data - recommend reassignment

sdb seems to be overheating:

Quoted 2 lines
Dec 26 11:25:56 localhost kernel: sdb: Current: sense key: Recovered Error
Dec 26 11:25:56 localhost kernel:     Additional sense: Warning - specified temperature exceeded

Now sdf is unhappy..

Quoted 18 lines
Dec 29 06:26:07 localhost kernel: sdf: Current: sense key: Hardware Error
Dec 29 06:26:07 localhost kernel:     Additional sense: Mechanical positioning error
Dec 29 06:26:07 localhost kernel: end_request: I/O error, dev sdf, sector 9664
Dec 29 06:26:07 localhost kernel: raid5: Disk failure on sdf, disabling device. Operation continuing on 4 devices
Dec 29 06:26:07 localhost kernel: RAID5 conf printout:
Dec 29 06:26:07 localhost kernel:  --- rd:6 wd:4 fd:2
Dec 29 06:26:07 localhost kernel:  disk 0, o:1, dev:sdc
Dec 29 06:26:07 localhost kernel:  disk 1, o:1, dev:sdd
Dec 29 06:26:07 localhost kernel:  disk 2, o:1, dev:sde
Dec 29 06:26:07 localhost kernel:  disk 3, o:0, dev:sdf
Dec 29 06:26:07 localhost kernel:  disk 4, o:1, dev:sdg
Dec 29 06:26:07 localhost kernel: RAID5 conf printout:
Dec 29 06:26:07 localhost kernel:  --- rd:6 wd:4 fd:2
Dec 29 06:26:07 localhost kernel:  disk 0, o:1, dev:sdc
Dec 29 06:26:07 localhost kernel:  disk 1, o:1, dev:sdd
Dec 29 06:26:07 localhost kernel:  disk 2, o:1, dev:sde
Dec 29 06:26:07 localhost kernel:  disk 4, o:1, dev:sdg
Quoted 8 lines
Dec 29 06:26:07 localhost kernel: Buffer I/O error on device md0, logical block 6040
Dec 29 06:26:07 localhost kernel: lost page write due to I/O error on md0
Dec 29 06:26:07 localhost kernel: REISERFS: abort (device md0): Journal write error in flush_commit_list
Dec 29 06:26:07 localhost kernel: REISERFS: Aborting journal for filesystem on md0
Dec 29 07:16:56 localhost kernel: ReiserFS: md0: warning: clm-6006: writing inode 2996 on readonly FS
Dec 29 07:16:56 localhost kernel: ReiserFS: md0: warning: clm-6006: writing inode 2996 on readonly FS
Dec 29 07:17:28 localhost kernel: Buffer I/O error on device md0, logical block 5052984
Dec 29 07:17:28 localhost kernel: lost page write due to I/O error on md0

I don't know if this completely explains the failure, but it seems you have bigger problems than one bad drive, and RAID5 cannot handle multiple drive failures.

Signature · 3 lines
Robert Hancock      Saskatoon, SK, Canada
To email, remove "nospam" from [email protected]
Home Page: http://www.roberthancock.com/
#2 lore raw

[solved][Re: software raid 5 broken on 2.6.14.4 sparc

Quoted 65 lines · Robert Hancock

Date: Thu, 29 Dec 2005 12:51:12 -0600
From: Robert Hancock <[email protected]>
To: linux-kernel <[email protected]>
Cc: [email protected]
Subject: Re: software raid 5 broken on 2.6.14.4 sparc

Quoted 5 lines · Gerhard Mack

Hello,

I had one of 5 drives fail in my raid 5 setup and now many files are unreadable.. Isn't raid 5 supposed to compensate for exactly this happening?

It looks like you have multiple drives with problems here:

Here's sde unhappy:

Quoted 3 lines
Dec 25 18:36:27 localhost kernel: sde: Current: sense key: Recovered Error
Dec 25 18:36:27 localhost kernel:     Additional sense: Recovered data -

recommend reassignment

sdb seems to be overheating:

Quoted 3 lines
Dec 26 11:25:56 localhost kernel: sdb: Current: sense key: Recovered Error
Dec 26 11:25:56 localhost kernel:     Additional sense: Warning - specified

temperature exceeded

Now sdf is unhappy..

Quoted 27 lines
Dec 29 06:26:07 localhost kernel: sdf: Current: sense key: Hardware Error
Dec 29 06:26:07 localhost kernel:     Additional sense: Mechanical
positioning error
Dec 29 06:26:07 localhost kernel: end_request: I/O error, dev sdf, sector
9664
Dec 29 06:26:07 localhost kernel: raid5: Disk failure on sdf, disabling

device. Operation continuing on 4 devices

Dec 29 06:26:07 localhost kernel: RAID5 conf printout:
Dec 29 06:26:07 localhost kernel:  --- rd:6 wd:4 fd:2
Dec 29 06:26:07 localhost kernel:  disk 0, o:1, dev:sdc
Dec 29 06:26:07 localhost kernel:  disk 1, o:1, dev:sdd
Dec 29 06:26:07 localhost kernel:  disk 2, o:1, dev:sde
Dec 29 06:26:07 localhost kernel:  disk 3, o:0, dev:sdf
Dec 29 06:26:07 localhost kernel:  disk 4, o:1, dev:sdg
Dec 29 06:26:07 localhost kernel: RAID5 conf printout:
Dec 29 06:26:07 localhost kernel:  --- rd:6 wd:4 fd:2
Dec 29 06:26:07 localhost kernel:  disk 0, o:1, dev:sdc
Dec 29 06:26:07 localhost kernel:  disk 1, o:1, dev:sdd
Dec 29 06:26:07 localhost kernel:  disk 2, o:1, dev:sde
Dec 29 06:26:07 localhost kernel:  disk 4, o:1, dev:sdg
Dec 29 06:26:07 localhost kernel: Buffer I/O error on device md0, logical
block 6040
Dec 29 06:26:07 localhost kernel: lost page write due to I/O error on md0
Dec 29 06:26:07 localhost kernel: REISERFS: abort (device md0): Journal
write error in flush_commit_list
Dec 29 06:26:07 localhost kernel: REISERFS: Aborting journal for filesystem
Quoted 8 lines

on md0

Dec 29 07:16:56 localhost kernel: ReiserFS: md0: warning: clm-6006: writing
inode 2996 on readonly FS
Dec 29 07:16:56 localhost kernel: ReiserFS: md0: warning: clm-6006: writing
inode 2996 on readonly FS
Dec 29 07:17:28 localhost kernel: Buffer I/O error on device md0, logical
block 5052984
Dec 29 07:17:28 localhost kernel: lost page write due to I/O error on md0
Quoted 4 lines

I don't know if this completely explains the failure, but it seems you have bigger problems than one bad drive, and RAID5 cannot handle multiple drive failures.

Ahah.. I knew about sdb.. overheating (it's not part of the array) I missed sde .. that one explains it.

Thanks,
Gerhard

--
Gerhard Mack

[email protected]

<>< As a computer I find your faith in technology amusing.