I'm getting scsi parity errors which cause disk failures with my raid. The obvious thing is to fix the hardware which I'm working on. If I make a file system on just one disk (no raid), I get the exact same parity errors, but the disk now keeps working. I don't know if the scsi driver does a retry or what, but it keeps working. So my question is why is the disk immediately faulted from the raid array? Is there some way of having it retry, or be more tolerant of certain errors? This is running 2.4.4. Below are the error messages. Thanks. Andy Jun 3 14:46:06 rad kernel: SCSI disk error : host 0 channel 0 id 0 lun 0 return code = 8000002 Jun 3 14:46:06 rad kernel: [valid=0] Info fld=0x0, Current sd08:02: sense key Aborted Command Jun 3 14:46:06 rad kernel: Additional sense indicates Scsi parity error Jun 3 14:46:06 rad kernel: I/O error: dev 08:02, sector 344512 Jun 3 14:46:06 rad kernel: interrupting MD-thread pid 832 Jun 3 14:46:06 rad kernel: raid5: parity resync was not fully finished, restarting next time. Jun 3 14:46:06 rad kernel: raid5: Disk failure on sda2, disabling device. Operation continuing on 4 devices - To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to [EMAIL PROTECTED]
