I'm pretty sure that this is my fault as some things I was tweaking around on the hardware itself caused the server to lock up, but I'm at a loss as to how to fix it.

 

I have monthly automatic parity checks scheduled that run every month on the first of the month. For three months in a row (only have a syslog for the last two), I have been getting the same 4 parity errors discovered near the beginning of the check and the rest of the check proceeds smoothly. I think that this might be a file system corruption that needs to be corrected but I'm not sure if it is or how to go about fixing it.

 

I have included the full syslog below as an attachment and I have the two months of parity errors copied and pasted below.

 

Thanks in advance!

 

Jul  1 00:00:02 Tower kernel: md: parity incorrect: 42904 (Errors)

Jul  1 00:00:02 Tower kernel: md: parity incorrect: 42928 (Errors)

Jul  1 00:00:02 Tower kernel: md: parity incorrect: 42936 (Errors)

Jul  1 00:00:02 Tower kernel: md: parity incorrect: 42944 (Errors)

 

Aug  1 00:00:02 Tower kernel: md: parity incorrect: 42904 (Errors)

Aug  1 00:00:02 Tower kernel: md: parity incorrect: 42928 (Errors)

Aug  1 00:00:02 Tower kernel: md: parity incorrect: 42936 (Errors)

Aug  1 00:00:02 Tower kernel: md: parity incorrect: 42944 (Errors)

 

syslog.zip

Your error sounds rather like the issue mentioned in these two topics:

 

http://lime-technology.com/forum/index.php?topic=13603.msg128868

 

http://lime-technology.com/forum/index.php?topic=13560.msg128495#msg128495

 

Regards,

 

Stephen

Looks like it, only I haven't installed any new hard disks to the array or done any resizing. So far I can't find anything wrong with the data stored on the array but obviously parity errors make me nervous.

 

You were doing a non-correcting check, right?

The data should be fine. Run a correcting parity check.