SATA woesJuly 17, 2006It's about three months since the last hard disk crash, so another one is due. My computer never fails to oblige. It's now saying:
Jul 17 14:36:51 kernel: ata1: command 0x25 timeout, stat 0xd0 host_stat 0x61
Jul 17 14:36:51 kernel: ata1: status=0xd0 { Busy }
Jul 17 14:36:51 kernel: sd 0:0:0:0: SCSI error: return code = 0x8000002
Jul 17 14:36:51 kernel: sda: Current: sense key: Aborted Command
Jul 17 14:36:51 kernel: Additional sense: Scsi parity error
Jul 17 14:36:51 kernel: end_request: I/O error, dev sda, sector 83227991
Jul 17 14:36:51 kernel: ATA: abnormal status 0xD0 on port 0xF8832C87
Jul 17 14:36:51 last message repeated 2 times
Jul 17 14:36:51 kernel: EXT3-fs error (device sda5): ext3_find_entry: reading directory #297570 offset 0
Jul 17 14:37:51 kernel: ata1: command 0x25 timeout, stat 0x50 host_stat 0x61
This is on a EXT-3 partion (OS is Fedora Core 5). There is another partiion with Reiserfs (I am not religious) and that fairs no better. Jul 17 14:23:42 kernel: Buffer I/O error on device sda3, logical block 5851 Jul 17 14:23:42 kernel: lost page write due to I/O error on sda3 Jul 17 14:23:42 kernel: ATA: abnormal status 0xD0 on port 0xF8832C87 Jul 17 14:23:42 last message repeated 2 times Jul 17 14:23:42 kernel: REISERFS: abort (device sda3): Journal write error in flush_commit_list Jul 17 14:23:42 kernel: REISERFS: Aborting journal for filesystem on sda3 Jul 17 14:24:42 kernel: ata1: command 0x25 timeout, stat 0x50 host_stat 0x61 Jul 17 14:27:30 kernel: ata1: command 0x25 timeout, stat 0x50 host_stat 0x61 One of the partitions in question is the '/usr' and when these errors occur the system even freezes (curiously enough it recorvers if you give it enuogh time). Rebooting gives you a jolt, the hard disk is not detected at all, reboot again and it's picked up fine. If you search the linux archives there are all kinds of explainations ranging from faulty hard drives through faulty SATA controllers to buggy kernel drivers, what's disapoointing is that there isn't a definite conclusion. Apparently all linux distros seem to be effected by this. My approach was to try booting up with an Ubuntu live CD and then fsck. Unfortunatley fsck suffered the same fate, it got struck halfway through. Then I opened up the covers and liberally doused the connectors and the card with contact cleaner and restarted. The errors still show up in the logs. So as a last ditch attempt I am going to upgrade the kernel, I hate doing that! Posted by raditha at July 17, 2006 10:59 AM
|
|



