Freezes and data loss (sata_via)
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| linux-source-2.6.20 (Ubuntu) |
Invalid
|
Undecided
|
Unassigned | ||
Bug Description
Binary package hint: linux-source-2.6.20
I get random freezes using 2.6.20-13-generic and 2.6.20-14-generic on my AMD64-machine with Asus A8V-E Deluxe (VIA K8T890). Everything is working fine with 2.6.20-12.
There seems to be a problem with my SATA-disk (Hitachi HDS722525VLSA80). These freezes also caused data loss once (nothing serious, only alsa-config, everything else could be repaired by fsck). I'm pretty sure that this was the reason, because I was doing nothing unusual, but I consider a whole bunch of errors on a ext3-fs that appear all of sudden _very_ unusual.
I get the following messages several times. I'm not able to get the messages before the freezes, because the only thing I can still do is to reboot using SysRQ-keys.
I'll attach a more complete log later (need to reboot using the newer kernel).
[ 61.130261] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2
[ 61.130322] ata1.00: (BMDMA stat 0x5)
[ 61.130388] ata1.00: cmd 35/00:00:
[ 61.130389] res 51/84:00:
[ 61.130461] ata1: soft resetting port
[ 61.296870] ATA: abnormal status 0x7F on port 0x000000000001c007
[ 61.307523] ATA: abnormal status 0x7F on port 0x000000000001c007
[ 61.326366] ata1.00: configured for UDMA/100
[ 61.326375] ata1: EH complete
[ 61.335213] SCSI device sda: 488397168 512-byte hdwr sectors (250059 MB)
[ 61.335483] sda: Write Protect is off
[ 61.335486] sda: Mode Sense: 00 3a 00 00
[ 61.337419] SCSI device sda: write cache: enabled, read cache: enabled, doesn't support DPO or FUA
| Andreas Gnau (rondom) wrote | #1 |
| Andreas Gnau (rondom) wrote | #2 |
| Changed in linux-source-2.6.20: | |
| status: | Unconfirmed → Rejected |
| Andreas Gnau (rondom) wrote | #3 |
| Andreas Gnau (rondom) wrote | #4 |
I've just found out that this problem also occurs on 2.6.20-12. I'm writing this using a feisty i386-install on the same HD (the filesystem of the root-partition of my amd64-install is too damaged now i.e. it doesn't boot anymore))
Such errors and freezes only occur if I access (mount from the i386-install) the root partition of the amd64-feisty-
I ran all kind of tests using Hitachi's Drive Fitness Test and it tells me that everything is ok (I still don't believe it, as of now).
As this problem doesn't seem to be directly related to any kernel upgrade, I'll close this bug.
I'll re-open it if the problem still exists after having bought a new HD.