November 13, 201510 yr I upgraded from unraid 5.04 to 6.1.3...everything went well (a few hiccups but nothing too serious)...there is one issue I'm having though. I can't confirm it, but I believe my drives were fine before the upgrade..after the upgrade it's saying one of my drives is bad. Is there any way I can have it check teh drives? How can I do a parity check? (I don't see anything in the GUI)
November 13, 201510 yr I believe that its saying that the drive is missing. (You can tell for sure when you stop the array). Is the drive still recognized by the BIOS?
November 13, 201510 yr Community Expert If still having trouble, go to Tools - Diagnostics and post the complete diagnostics zip file.
November 13, 201510 yr ... How can I do a parity check? You absolutely can NOT do a parity check -- 'nor do you want to. At this point the parity drive plus disk #1 have all the information needed to be able to emulate disk #2 and to rebuild it onto a new drive. Parity is shown as good ... hopefully that's accurate (when did you do the last check??). As long as it is indeed good, you can rebuild disk #2 and all of the data will still be fine. Hopefully you just have a loose cable or connection ... if that's the case after you fix it you'll be able to rebuild the contents back onto the same drive. If not, you'll need to replace the drive and then do a rebuild onto the new drive.
December 3, 201510 yr Author Sorry for the long pause before a reply... I've attached the diagnostics...the BIOS sees all 3 drives installed tower-diagnostics-20151203-1453.zip
December 3, 201510 yr Community Expert The syslog in the diagnostic is too recent to have captured the event. Has anything changed since the screenshot? Probably the disk has been disabled and is being emulated. It has very bad SMART and should be replaced by rebuilding to another disk.
December 3, 201510 yr Author The only things that have changed is that I've purchased the key, and rebooted the server.... Why would the drive the works have bad SMART and the drive that is dead has good SMART? So I'd need 2 new drives--what's the safest way to rebuild everything?
December 3, 201510 yr Community Expert The only things that have changed is that I've purchased the key, and rebooted the server.... Why would the drive the works have bad SMART and the drive that is dead has good SMART? So I'd need 2 new drives--what's the safest way to rebuild everything? Don't understand what you mean. Your screenshot only shows parity, disk1, and disk2. None of them are dead. Disk2 has been disabled by unRAID because it is unreliable. Disk2 is the one that should be replaced. Can you read the disk2 data? If so that is a good sign. It means the array can recover the disk2 data so it can be rebuilt to a new drive. What has happened is disk2 is failing as indicated by its SMART report in the diagnostics. When unRAID attempted to write to disk2 the write failed. When that happens, unRAID disables the drive and WILL NOT USE that drive slot again until it is rebuilt. Instead of actually using that disk, unRAID emulates it. When it needs to read the drive, unRAID reads all of the other data drives plus parity and calculates the data for that disk instead of actually reading the disabled drive. When it needs to write the drive, unRAID reads all of the other data drives plus parity, calculates what change would happen to parity as a result of the data change, and writes parity instead of actually writing the disabled drive. See this wiki
December 3, 201510 yr Author Right--I understand that..but when I look at the 'dashboard', it shows the SMART status for the parity drive and data 2 to be good...but data 2 is faulty... Should I replace data 2...then rebuild...then replace data 1?
December 3, 201510 yr Community Expert In Disk Settings you can see what attributes are monitored. If you click on a disk from Main it will let you look at its attributes. From your diagnostics: Disk1 is not perfect Serial Number: W3005LLG 5 Reallocated_Sector_Ct 0x0033 100 100 010 Pre-fail Always - 8 187 Reported_Uncorrect 0x0032 098 098 000 Old_age Always - 2 197 Current_Pending_Sector 0x0012 100 100 000 Old_age Always - 0 198 Offline_Uncorrectable 0x0010 100 100 000 Old_age Offline - 0 But not nearly as bad as disk2 Serial Number: W3006GJ1 5 Reallocated_Sector_Ct 0x0033 093 093 010 Pre-fail Always - 8752 187 Reported_Uncorrect 0x0032 001 001 000 Old_age Always - 145 197 Current_Pending_Sector 0x0012 001 001 000 Old_age Always - 55976 198 Offline_Uncorrectable 0x0010 001 001 000 Old_age Offline - 55976 Probably the dashboard should be giving a thumbs down on disk2 also but maybe it isn't really checking it anymore since it's disabled.
December 3, 201510 yr The "thumbs down" symbol for disk 1 doesn't mean it's bad ... just that there are some attributes that bear watching. In this case, the reallocated sectors. Your real issue -- as noted above -- is that disk #2 is disabled, so THAT needs to be resolved (by replacing the disk and letting the data rebuild). AFTER that is completed, if you want to also replace disk #1 that's fine ... but the reality is that 8 reallocated sectors isn't a bad count with modern drives ... the key thing is to watch that and see if the number changes (increases) => if so, I'd replace the drive before it becomes a problem.
Archived
This topic is now archived and is closed to further replies.