Kernel loses USB-drive

Bug #98706 reported by Fridtjof Busse
2
Affects Status Importance Assigned to Milestone
linux-meta (Ubuntu)
Fix Released
Undecided
Unassigned

Bug Description

Binary package hint: linux-image-generic

I have a serious problem with my backup-drive (using dar): The kernel reproducable loses contact to the drive on feisty. Works just fine on edgy, and I tested it with two different drives, it happens every time.
I have to reboot the system to be able to use the drive again (though no fsck was necessary so far after the hang).

[ 285.173952] EXT3 FS on sdf1, internal journal
[ 285.173956] EXT3-fs: mounted filesystem with ordered data mode.
[ 959.704026] kjournald starting. Commit interval 5 seconds
[ 959.704758] EXT3 FS on sdf1, internal journal
[ 959.704861] EXT3-fs: mounted filesystem with ordered data mode.
[ 2573.352524] usb 7-6: reset high speed USB device using ehci_hcd and address 4
[ 2573.464394] usb 7-6: device descriptor read/64, error -71
[ 2573.680144] usb 7-6: device descriptor read/64, error -71
[ 2573.895890] usb 7-6: reset high speed USB device using ehci_hcd and address 4
[ 2574.007758] usb 7-6: device descriptor read/64, error -71
[ 2574.223509] usb 7-6: device descriptor read/64, error -71
[ 2574.439264] usb 7-6: reset high speed USB device using ehci_hcd and address 4
[ 2574.846785] usb 7-6: device not accepting address 4, error -71
[ 2574.958656] usb 7-6: reset high speed USB device using ehci_hcd and address 4
[ 2575.366182] usb 7-6: device not accepting address 4, error -71
[ 2575.366211] usb 7-6: USB disconnect, address 4
[ 2575.366219] sd 5:0:0:0: scsi: Device offlined - not ready after error recovery
[ 2575.366232] sd 5:0:0:0: SCSI error: return code = 0x00050000
[ 2575.366235] end_request: I/O error, dev sdf, sector 259151583
[ 2575.366240] Buffer I/O error on device sdf1, logical block 32393940
[ 2575.366242] lost page write due to I/O error on sdf1
[ 2575.366247] Buffer I/O error on device sdf1, logical block 32393941
[ 2575.366249] lost page write due to I/O error on sdf1
[ 2575.366252] Buffer I/O error on device sdf1, logical block 32393942
[ 2575.366255] lost page write due to I/O error on sdf1
[ 2575.366258] Buffer I/O error on device sdf1, logical block 32393943
[ 2575.366260] lost page write due to I/O error on sdf1
[ 2575.366263] Buffer I/O error on device sdf1, logical block 32393944
[ 2575.366265] lost page write due to I/O error on sdf1
[ 2575.366268] Buffer I/O error on device sdf1, logical block 32393945
[ 2575.366271] lost page write due to I/O error on sdf1
[ 2575.366274] Buffer I/O error on device sdf1, logical block 32393946
[ 2575.366276] lost page write due to I/O error on sdf1
[ 2575.366279] Buffer I/O error on device sdf1, logical block 32393947
[ 2575.366281] lost page write due to I/O error on sdf1
[ 2575.366285] Buffer I/O error on device sdf1, logical block 32393948
[ 2575.366288] lost page write due to I/O error on sdf1
[ 2575.366291] Buffer I/O error on device sdf1, logical block 32393949
[ 2575.366293] lost page write due to I/O error on sdf1
[ 2575.370186] sd 5:0:0:0: SCSI error: return code = 0x00010000
[ 2575.370190] end_request: I/O error, dev sdf, sector 259151823
[ 2575.373121] EXT3-fs error (device sdf1): read_block_bitmap: Cannot read block bitmap - block_group = 990, block_bitmap = 32440320
[ 2575.373523] EXT3-fs error (device sdf1): read_block_bitmap: Cannot read block bitmap - block_group = 990, block_bitmap = 32440320
[ 2575.389043] Aborting journal on device sdf1.
[ 2575.389070] EXT3-fs error (device sdf1) in ext3_ordered_writepage: IO failure
[ 2575.389397] ext3_abort called.
[ 2575.389399] EXT3-fs error (device sdf1): ext3_journal_start_sb: Detected aborted journal

Revision history for this message
Fridtjof Busse (fbusse-deactivatedaccount-deactivatedaccount) wrote :

After the last bunch of updates, the problem vanished, i.e. I can't reproduce it anymore. Don't know what fixed it, but it seems to work now :)

Changed in linux-meta:
status: New → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.