On Mon, Sep 28, 2009 at 9:49 AM, Dave Howorth
Greg Freemyer wrote:
You may need a new disk. It's not conclusive in my opinion. If it was my drive, I'd likely replace it just to be sure.
Thanks for your help, Greg
Media error is a real error. Bad cables, controller, PSU, wont cause a false media error report. You have at least one sector of the actual platter which when read does not get a checksum agree. Writing to that sector _may_ trigger a relocate and all is well. Or if it is a "soft" error, writing to the sector may just update the data and the checksum/crc thus making all well without a relocate.
If the drive was in the middle of writing that specific sector when power failed, it may have only been partially written and thus the checksum / CRC or whatever it is fails. That would be an example of a soft error.
So a single bad sector may not be a big deal at all in that case.
Sep 28 13:50:28 suse1 kernel: sd 2:0:0:0: [sda] Add. Sense: Unrecovered read error - auto reallocate failed
I'm not familiar with that error. I did not think reallocate was even tried on a read, so I have no idea what the drive is trying to tell you with that.
Sep 28 13:50:28 suse1 kernel: end_request: I/O error, dev sda, sector 57917603
This is the sector you could try to force a relocate on:
dd if=/dev/zero of=/dev/sda seek=57917603 bs=512 count=1
I tried that but it appears to have failed. It took a long time and top showed a high percentage io wait time while it was active:
# dd if=/dev/zero of=/dev/sda seek=57917603 bs=512 count=1 dd: writing `/dev/sda': Input/output error 1+0 records in 0+0 records out 0 bytes (0 B) copied, 25.471 s, 0.0 kB/s
and the log is still filling with the same error messages
What do the new logs say about reallocate? If reallocate on write failed, toss the drive. Did you try the dd command twice. Maybe it took 25 seconds to do the first realloc, and now all is well. Greg -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org