JDW Posted June 7, 2017 Share Posted June 7, 2017 I noticed that I can no longer add files to my array. Went to web GUI and found that my parity drive is marked "disabled". When I restart it says my drive is (blue ball) "new drive" What do I do now? At this point I would be happy to resync and start over. I still on 5.0.5, if that helps Quote Link to comment
JorgeB Posted June 8, 2017 Share Posted June 8, 2017 Check SMART and if it looks good check cables and re-sync parity. Quote Link to comment
itimpi Posted June 8, 2017 Share Posted June 8, 2017 9 hours ago, JDW said: I noticed that I can no longer add files to my array. Went to web GUI and found that my parity drive is marked "disabled". When I restart it says my drive is (blue ball) "new drive" What do I do now? At this point I would be happy to resync and start over. I still on 5.0.5, if that helps The parity drive dropping offline should not be sufficient to stop files being written to the array. It suggests there might be file system corruption on one (or more) of the drives. You should put the array into Maintenance mode and carry out a file system (reiserfsck) check on each of your drives. Quote Link to comment
JDW Posted June 11, 2017 Author Share Posted June 11, 2017 Thanks for tip to run reiserfsck. This led me to unraid wiki articles talking about disk checks [1]. I ran reiserfsck --check that resulted in below screenshot. Then I tried reiserfsck --rebuild-tree and that logged a bunch of statements about corrupted blocks and got hung without making further progress. Should I just preclear_disk at this point? [1] https://wiki.lime-technology.com/Check_Disk_Filesystems#Drives_formatted_with_ReiserFS_using_unRAID_v5_or_later Quote Link to comment
itimpi Posted June 11, 2017 Share Posted June 11, 2017 The command line you used looks incorrect! When using the raw device names you need to include the partition number so you should have been using '/dev/hdb1'. if the array is in maintenance mode then you can use /dev/mdX where X refers to the disk number in the unRAID GUI as the mdX devices already take the partition into account. Quote Link to comment
SSD Posted June 11, 2017 Share Posted June 11, 2017 Could be an issue with the USB disk. The USB disk contains the absolute record of what drives are assigned to what slots. If a slot is blue, that often means the USB was not updated properly. I would remove it and insert into workstation to run a chkdsk to check for and correct corruption. Quote Link to comment
JDW Posted June 24, 2017 Author Share Posted June 24, 2017 So I have decided to basically start over since this is just a parity disk. I took parity out of array, ran mkraiserfs, then tried to preclear_disk The first time array was in maintenance mode, preclear only made it to ~40% before becoming unresponsive (and no longer writing to disk). Then I restarted & stopped array completely, this time I made it to 90%. Restarted again and made sure array was stopped, only made it to 40%. I am kind of confused why it would do that? I initialized all 4 disks on this box without issue. Quote Link to comment
JorgeB Posted June 24, 2017 Share Posted June 24, 2017 Not following, parity doesn't have a file system. Preclear in maintenance mode? Quote Link to comment
trurl Posted June 24, 2017 Share Posted June 24, 2017 4 hours ago, johnnie.black said: Not following, parity doesn't have a file system. Preclear in maintenance mode? JDW, it sounds like you went completely offtrack on your 2nd post. itimpi was suggesting checking filesystem on your data disks. As johnnie.black says, parity doesn't have a filesystem to check. And there is no need to preclear a parity disk. You should post a syslog, then we may want SMART reports from your disks. Quote Link to comment
JDW Posted June 25, 2017 Author Share Posted June 25, 2017 Yes, it sounds like I am way off track. Here's the syslog syslog.bak Quote Link to comment
JorgeB Posted June 25, 2017 Share Posted June 25, 2017 Post a SMART report of your disks, or at least from your parity disk: https://wiki.lime-technology.com/Troubleshooting#Obtaining_a_SMART_report Quote Link to comment
JDW Posted June 25, 2017 Author Share Posted June 25, 2017 Here's smart report for parity disk (this should be the result of long test I ran shortly after it stopped working) sdb.smart Quote Link to comment
JorgeB Posted June 26, 2017 Share Posted June 26, 2017 SMART looks fine but there are signs of a bad SATA cable, it could be from before, but replace it and re-sync parity, if it fails post the syslog before rebooting. Quote Link to comment
JDW Posted June 27, 2017 Author Share Posted June 27, 2017 Thanks, I will replaces cables and post here in a few days. Just out of curiosity, how do you tell bad cables? Quote Link to comment
JorgeB Posted June 27, 2017 Share Posted June 27, 2017 199 UDMA_CRC_Error_Count 0x003e 200 198 000 Old_age Always - 26055 26055 CRC errors, but can't tell if these are new or old errors, unless you have an earlier SMART report. Quote Link to comment
JDW Posted July 1, 2017 Author Share Posted July 1, 2017 I picked up a new set of cables from monoprice [1] and started rebuilding array. [1] https://www.monoprice.com/product?p_id=8782 Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.