smdion Posted February 20, 2014 Share Posted February 20, 2014 ran reiserfsck after unRAID put a drive in readonly ########### reiserfsck --check started at Thu Feb 20 14:22:24 2014 ########### Replaying journal: Done. Reiserfs journal '/dev/md2' in blocks [18..8211]: 0 transactions replayed Checking internal tree.. \/ 1 (of 21|/ 48 (of 99//141 (of 170-block 227639328: The level of the node (61551) is not correct, (1) expected the problem in the internal node occured (227639328), whole subtree is skipped finished Comparing bitmaps..vpf-10640: The on-disk and the correct bitmaps differs. Bad nodes were found, Semantic pass skipped 1 found corruptions can be fixed only when running with --rebuild-tree ########### reiserfsck finished at Thu Feb 20 15:11:16 2014 I pulled the drive and I'm doing a parity rebuild to a new one. What can I do to make sure this drive is safe to put back in. Link to comment
JonathanM Posted February 21, 2014 Share Posted February 21, 2014 I pulled the drive and I'm doing a parity rebuild to a new one.The parity rebuilt drive will have exactly the same corruptions as the drive you pulled. Parity has no concept of files, it only puts back the bits exactly as they were when the drive was pulled. Good news is you will have two exact copies of the corrupt drive, so you can try different recovery strategies if one fails. Unraid is not a backup. It's a safeguard against a single drive catastrophic failure. Link to comment
smdion Posted February 21, 2014 Author Share Posted February 21, 2014 Thats why I have a 2nd box with backup So Parity would take the filesystem error? I assume it would try to make parity for that file and wouldn't be able to. So it would have corrupted files for the files that were on those bad nodes and it would restore the corrupted files, but not the bad node as that is "below" the file. As it is backing up the files and not the file system. If I'm wrong please correct me My question remains what is the best thing I can to do get back that disk working or should I RMA it? If the parity does not work, I will pull from my backup array. Thanks! Link to comment
JonathanM Posted February 21, 2014 Share Posted February 21, 2014 Unraid parity is partition level, files aren't considered at all. Start reading here. http://lime-technology.com/wiki/index.php/FAQ#How_does_parity_work.3F As far as the drive goes, how does the smart report look? Link to comment
smdion Posted February 21, 2014 Author Share Posted February 21, 2014 Unraid parity is partition level, files aren't considered at all. Start reading here. http://lime-technology.com/wiki/index.php/FAQ#How_does_parity_work.3F As far as the drive goes, how does the smart report look? Short/Long pass fine. Thanks for that info. That makes a lot more sense! Link to comment
JonathanM Posted February 21, 2014 Share Posted February 21, 2014 As far as the drive goes, how does the smart report look? Short/Long pass fine. In that case, if you are planning to add that drive back to the array in another slot, you should probably run a preclear pass on it, so the array won't be down for so long while the drive is cleared. Preclear will irretrievably erase the drive. I'm not sure what you meant by What can I do to make sure this drive is safe to put back in. so I'm assuming since you rebuilt that drive slot, you want to add the drive to the array as a blank drive to increase your free space. Link to comment
smdion Posted February 21, 2014 Author Share Posted February 21, 2014 As far as the drive goes, how does the smart report look? Short/Long pass fine. In that case, if you are planning to add that drive back to the array in another slot, you should probably run a preclear pass on it, so the array won't be down for so long while the drive is cleared. Preclear will irretrievably erase the drive. I'm not sure what you meant by What can I do to make sure this drive is safe to put back in. so I'm assuming since you rebuilt that drive slot, you want to add the drive to the array as a blank drive to increase your free space. I've read a few posts in here where the output from reiserfsck was a certain way where they recommended RMA'ing the drive and not putting back into service. Just wanted to make sure this wasn't that case. I've now made a new config and I'm backing up from my other unRAID box. Once that completes I will pre clear the drive and put it back into service. I appreciate your support and sharing knowledge with me. Link to comment
JonathanM Posted February 21, 2014 Share Posted February 21, 2014 I've read a few posts in here where the output from reiserfsck was a certain way where they recommended RMA'ing the drive and not putting back into service. Just wanted to make sure this wasn't that case.My recommendation to run Joe L.'s preclear script on the drive has two purposes, the first is to check the health of the drive. If you are at all concerned that the drive is not healthy, I'd run multiple passes of the script. The preclear script is excellent for weeding out bad drives, as it reads, writes, and verifies every single sector that unraid will use. The second is to speed the addition of the drive into an already parity protected array. If you have an already healthy parity protected array, any new drive must be set to all zeroes so as not to effect the parity calculations. If the drive is not precleared, unraid will take many hours to write zeroes to the drive, during that time the array will be offline. I've now made a new config and I'm backing up from my other unRAID box. Once that completes I will pre clear the drive and put it back into service. Setting a new configuration immediately invalidates parity, and forces a rebuild of the parity disk based on the current contents of the data disks. After parity has been built, you should immediately do a parity check, to make sure the parity disk can be read error free. Link to comment
Recommended Posts
Archived
This topic is now archived and is closed to further replies.