My qnap 8tbx4 raid 5 has a drive that popped up with am alert and some bad blocks. A couple days later the bad block count has gone from to 30 to over 1,0000.
I’ve ordered another drive, but is there a way to repair it and stop the bleeding? Is my onlychoice to chunk iy and hope the rebuild finishes?
Thanks!
To save the disk from a landfill, not the data on it, I would do a full badblocks -wsv write test on it.
This will give the drive plenty of opportunities to discover issues (and let it do its thing internally to work around them if possible) as it writes and reads multiple patterns across the whole drive. Each round of testing will be a total 64TB job for an 8TB drive, so set it and forget it and check back many days later. If this isn’t worth your time, then just give it away or chuck it.
Log your SMART values before running the test.
If your drive does not die during testing, check the terminal for errors. If there are no errors there that is a good sign but check SMART values next. If any values increased since before you ran the test, I would run another round.
If the drive can pass with no errors and the SMART values stay the same I would consider it to be proven stable and working. At this point I wouldn’t use it for any RAID and would keep it strictly as a single-disk, mostly as an extra backup.