March 27, 20206 yr Hi guys, got a disk failure and I decided to replace it with a new one. I pre-cleared it and than I begun the rebuilt process that started normally and with no problems. I went to sleep and the estimated finish was around 15 hrs but this morning I woke up and the estimated finish is now almost 300 days... On top I can see now some errors on a different disk that I am not sure were there yesterday night. Attached a diagnostics file that I just download if it is of any help. monstruo-diagnostics-20200327-0915.zip
March 27, 20206 yr Community Expert Disk8 dropped offline, looks more like a connection problem, cancel rebuild, replace cables on disk8 and try again. P.S. there were also a few read errors on disk19, and that looks like an actual disk problem, if it happens again the rebuilt disk can have some corruption.
March 27, 20206 yr Author Ok Done... Now that I restarted the rebuilt the disk 8 say: Unmountable: no file system. Normal?
March 27, 20206 yr Community Expert Unfortunately disk8 also appears to be failing, stop the rebuild and run an extended SMART test, take the opportunity and do the same for disks 10, 14 and 19.
March 28, 20206 yr Author Ok... after almost 20 hrs the Extended self-test routine seems to be stucked at 90% (since several hours now...). According the Extended self-test routine recommended polling time it was supposed to be done in around 527minutes. What next? Sorry if I am useless but I already lost several disks in the past without being able to recover the data and now I would like to do things in the correct way.
March 28, 20206 yr Community Expert You should wait a few more hours, if the disk as slow sectors, and very likely there are some due to the very high number of raw_read_errors, it can take much longer than a healthy disk, SMART test should never get stuck, it either completes or aborts with an error.
March 28, 20206 yr Community Expert Longer than normal time means disk might be marginal, but if there's no error it's OK for now, all other disks I mentioned above also finished without error?
March 28, 20206 yr Author Did not know I could have done all together.... Sorry. Started all of them now. Shall I start again the Data Rebuilt? Or shall I wait all the disks to be checked?
March 29, 20206 yr Author Finally all checks are done (I attach the results of the three disks). They all return a Completed without error message even (I do not know how important it is) all three disks have Current Pending Sectors in yellow... I am waiting more instructions to follow! Thanks a lot for the help and the patience monstruo-smart-Disk8.zip monstruo-smart-Disk14.zip monstruo-smart-Disk19.zip
March 30, 20206 yr Author Still same problem.... The estimated finish is oscillating between 295 days and 24 hours. Normal?
March 30, 20206 yr Community Expert As long as there are no read errors let it go on, some of your disks appear to be marginal (but OK for now), and when reading those slow sectors zones speeds can fluctuate a lot.
March 30, 20206 yr Author Ok. So for the time I let it go till is done. Just wondering how long shall I set as maximum waiting time (2/3 days?). Shall I plan to change disks 8,14,19? Thanks
March 30, 20206 yr Author Second thing... Can I copy files into the server even if the Data rebuilt is running or shall I wait when is completed?
March 30, 20206 yr Community Expert 7 minutes ago, gideva said: Just wondering how long shall I set as maximum waiting time (2/3 days?). I would estimate max a couple of days, any more you'll likely get errors first. 8 minutes ago, gideva said: Shall I plan to change disks 8,14,19? I would, IMHO it's just a matter of time until they fail again. 5 minutes ago, gideva said: Second thing... Can I copy files into the server even if the Data rebuilt is running or shall I wait when is completed? You can, but it will slow down even more, so avoid as much as you can.
March 30, 20206 yr Community Expert Regarding the disks when this is over you can run the diskspeed docker, it might help identify which ones are performing worse.
Archived
This topic is now archived and is closed to further replies.