I upgraded from 4.7 to 5.0-rc6-r8168-test a couple of months ago.  Did parity checks before and after and all was well.  Last night I started a parity check and this morning discovered a red ball with a whole bunch of errors on one disk.  Tried (via the web interface) to get a smart report and do a reiser fsck, neither one would run.  Fsck said it couldn't find a file system, which was interesting since I could see at least a portion of files via the command prompt (though an 'ls -lR' eventually hung) and spot checking the files I could see leads me to believe they're good.

 

I'm assuming I will have to replace the drive.  I'm concerned that since the failure seems to have happened in the middle of a check/correct, that parity might be corrupted. I'm hoping someone can reassure me that that can't be the case.  Is there anything I should do differently or look out for in replacing this drive than I would normally do?

 

BTW, found some possible interesting info in the syslog (attached).  Some is likely to be more related to the upgrade in versions than it is to the disk failure, but attached the syslog just in case.  See Sep 16 17:50:01 for example.  The problems in the parity check occur starting at Nov 4 05:01:21.

 

Thanks for any help and advice you might offer.

 

syslog-2012-11-04.zip

Power down and check the cableing to disk 3. Restart and get a SMART report. If you can't get a SMART report the disk must be replaced. The files are available because unRAID is simulating the disabled drive. All operations on disk 3 will take a long time dependent on your hardware. The ls command may have eventually finished. If another disk fails you will lose the data on both drives.