October 2, 201411 yr My monthly parity check found 1 error. Is there any way to narrow down the location of the error? No drives show any error on main status screen. All drives are green. In looking through my log file, scanning for the words "error" or "incorrect" the only entries I find are: [glow=red,2,300](Right off-the-bat at the start of the scan)[/glow] Oct 1 00:00:01 tower kernel: mdcmd (115): check NOCORRECT Oct 1 00:00:01 tower kernel: Oct 1 00:00:01 tower kernel: md: recovery thread woken up ... Oct 1 00:00:01 tower kernel: md: recovery thread checking parity... Oct 1 00:00:01 tower kernel: md: using 1536k window, over a total of 1953514552 blocks. Oct 1 00:00:06 tower kernel: md: parity incorrect, sector=128 Oct 1 00:47:26 tower sSMTP[4764]: Creating SSL connection to host Oct 1 00:47:26 tower sSMTP[4764]: SSL connection using RC4-SHA [glow=red,2,300]...the completion of things...[/glow] Oct 1 19:18:07 tower kernel: md: sync done. time=69485sec Oct 1 19:18:07 tower kernel: md: recovery thread sync completion status: 0 Even if the error can't be narrowed down further than 'sector 128 on one of the drives' I was hoping there'd be some way to determine which files are sitting on that sector on each of my drives. (A lot easier to verify 15 files than 150,000)
October 2, 201411 yr There is actually no practical way to relate a particular sector to a particular file. Having said that, sector 128 will almost certainly be in the control area of one of the drives, not in a file. Were you running a correcting parity check? If so I expect everything is fine and this was a genuine correction of parity. If you want to be more confident then you might want to consider putting the array into maintenance mode and running a reiserfsck check against your drives to ensure there is no file system corruption on any of the drives.
Archived
This topic is now archived and is closed to further replies.