July 26, 20178 yr I'm just wondering is this drive safe to add this drive to my array? It passed two cycles of preclear but it has a SMART warning for offline uncorrectable. The drive in question is an old (4-5years) WD green 2TB drive ############################################################################################################################ # # # unRAID Server Preclear of disk WD-WMAZ20281452 # # Cycle 2 of 2, partition start on sector 64. # # # # # # Step 1 of 5 - Pre-read verification: [7:12:58 @ 0 MB/s] SUCCESS # # Step 2 of 5 - Zeroing the disk: [6:19:23 @ 87 MB/s] SUCCESS # # Step 3 of 5 - Writing unRAID's Preclear signature: SUCCESS # # Step 4 of 5 - Verifying unRAID's Preclear signature: SUCCESS # # Step 5 of 5 - Post-Read verification: [6:15:04 @ 88 MB/s] SUCCESS # # # # # # # # # # # # # # # ############################################################################################################################ # Cycle elapsed time: 19:49:05 | Total elapsed time: 39:35:45 # ############################################################################################################################ ############################################################################################################################ # # # S.M.A.R.T. Status default # # # # # # ATTRIBUTE INITIAL CYCLE 2 STATUS # # 5-Reallocated_Sector_Ct 0 0 - # # 9-Power_On_Hours 27335 27374 Up 39 # # 194-Temperature_Celsius 30 34 Up 4 # # 196-Reallocated_Event_Count 0 0 - # # 197-Current_Pending_Sector 1 0 Down 1 # # 198-Offline_Uncorrectable 1 1 - # # 199-UDMA_CRC_Error_Count 0 0 - # # # # # # # # # # # ############################################################################################################################ # SMART overall-health self-assessment test result: PASSED # ############################################################################################################################ --> ATTENTION: Please take a look into the SMART report above for drive health issues. --> RESULT: Preclear Finished Successfully!.
July 26, 20178 yr It's all up to your risk tolerance. Keep in mind that all other drives are needed to recover a failed drive, and must be read perfectly to accurately reconstruct the missing drive. That means any dodgy drive, even empty, can jeopardize a successful recovery. That said, I've personally run with drives sketchier than what yours is reporting, but I have decent backups. Your call.
July 27, 20178 yr Did this just happen or has it been that way for a long time? If is often the movement in values, rather that the values themselves, that give the most information. Part of me says watch it and see if it gets worse. The other part of me says that a 5 year old 2T is too old and too small to be worried about. Buy a couple bigger disks and clean out the small disks. Use the small disks as backups.
July 27, 20178 yr Author 10 hours ago, bjp999 said: Did this just happen or has it been that way for a long time? If is often the movement in values, rather that the values themselves, that give the most information. Part of me says watch it and see if it gets worse. The other part of me says that a 5 year old 2T is too old and too small to be worried about. Buy a couple bigger disks and clean out the small disks. Use the small disks as backups. I'm not sure when it has been that way. I recently took the 2TB out of my main PC to put it in my unraid server and that's when I noticed the SMART report. I guess i'll leave it as part of my array and if it fails I can always buy another HDD to replace it. All the other HDD are newly bought 4TB which have passed preclear so I don't think the other HDD are prone to fail anytime soon
July 27, 20178 yr "Fails" is a strong term. Drives rarely fail. They are not light light bulbs and other electronics, which will just refuse to turn on when they break. Drives tend to (but not always) show some SMART attribute anomalies. If left untreated to they tend to do nastier things like cause read errors, and/or occasionally start reporting invalid data. It is VERY rare that a drive actually "breaks" to the point it won't power up or be recognized by a controller. As an aside - in some ways unRAID would work much better if drives DID fail like light bulbs. Poof - drive is dead. Replace / rebuild it. It is the act and diagnosis of failing and that causes so many problems! If you are attentive (and I think plugins like find common problems help with this), you see the SMART problems and act on them, preemptively addressing problems before the drive starts spewing garbage which can will mess up parity and cause even single disk recovery to be imperfect. When I get a problem like yours, I will run a few non-correcting parity checks. If I can get through 3 of them in a row and SMART problems get no worse, I'll trust the drive. But if I run 4-5 and they keep coming, or worse I start getting parity errors, it is time to replace the drive.
July 27, 20178 yr Author 15 minutes ago, bjp999 said: "Fails" is a strong term. Drives rarely fail. They are not light light bulbs and other electronics, which will just refuse to turn on when they break. Drives tend to (but not always) show some SMART attribute anomalies. If left untreated to they tend to do nastier things like cause read errors, and/or occasionally start reporting invalid data. It is VERY rare that a drive actually "breaks" to the point it won't power up or be recognized by a controller. As an aside - in some ways unRAID would work much better if drives DID fail like light bulbs. Poof - drive is dead. Replace / rebuild it. It is the act and diagnosis of failing and that causes so many problems! If you are attentive (and I think plugins like find common problems help with this), you see the SMART problems and act on them, preemptively addressing problems before the drive starts spewing garbage which can will mess up parity and cause even single disk recovery to be imperfect. When I get a problem like yours, I will run a few non-correcting parity checks. If I can get through 3 of them in a row and SMART problems get no worse, I'll trust the drive. But if I run 4-5 and they keep coming, or worse I start getting parity errors, it is time to replace the drive. Something interesting happened which I did not know could happen. I got a notification for that same drive today which showed the error disappeared in the SMART report.
July 27, 20178 yr It can happen. Some of the attributes can reset/reduce under certain circumstances. Doesn't mean you don't have a problem. I'd prefer that the attributes had stayed where they were consistently. The fact that they are changing may still be a bad sign. I'd still go with the 3 parity check approach I laid out earlier. I'm not sure what the "undefined undefined ..." overlay means. Are you sure the drive didn't drop offline and this is not the smart report for a different disk? What exactly did the notification say?
July 27, 20178 yr Author It isn't a smart report for a different disk. If you compare the values from the post with the one from the first post all the values are the same except the offline uncorrectable. The notification message I got was that the offline uncorrectable value changed in the SMART status. Not sure what that "undefined undefined " overlay is but its a notification I just noticed and cleared it...Maybe because I was browsing using mobile when I took the screenshot? Not sure. I haven't added this drive to the array yet. So please correct me if I'm wrong. I should add the device to the array first, let it do a parity check 3 times without corrections to the parity. Is the only way to do that is clicking on the check button 3 times (while leaving the write corrections to parity unchecked) or can I run it three times consecutively. Each parity check takes quite a while to do so it might take a couple days to complete.
July 27, 20178 yr Since it is not in the array yet - do a preclear on it. Let's see what the attributes show afterwards.
July 27, 20178 yr Author 10 minutes ago, bjp999 said: Since it is not in the array yet - do a preclear on it. Let's see what the attributes show afterwards. It's passed two cycles of preclear (results are in the first post) That's why I was wondering if it was safe to add to my array. I think it should be fine now that the SMART status has changed with no errors
July 27, 20178 yr Community Expert Do you really need its capacity right now? If not, I would just leave it out and go for newer bigger later on as bjp999 says. 16 hours ago, bjp999 said: a 5 year old 2T is too old and too small to be worried about. Buy a couple bigger disks and clean out the small disks. Use the small disks as backups. Adding a disk just because you have it isn't really a good idea. More disks means more opportunities for problems. If you don't need the space don't add a disk.
July 27, 20178 yr Author 20 minutes ago, trurl said: Do you really need its capacity right now? If not, I would just leave it out and go for newer bigger later on as bjp999 says. Adding a disk just because you have it isn't really a good idea. More disks means more opportunities for problems. If you don't need the space don't add a disk. In that case should I just leave it as an unassigned disk not on the array and use it to backup important appdata and documents? Would I be able to create a usershare that is on the unassigned disk or is the only way to access the disk directly through /mnt/disk/
July 27, 20178 yr Community Expert 3 minutes ago, bobokun said: In that case should I just leave it as an unassigned disk not on the array and use it to backup important appdata and documents? Would I be able to create a usershare that is on the unassigned disk or is the only way to access the disk directly through /mnt/disk/ If you install the Unassigned devices plugin then it helps you manage drives that are not part of the array. Although you cannot set such a drive as part of a User Share you can give the drive a share name to make it visible across the network. The UD plugin also allows you to use file system formats (e.g. NTFS) that are not supported for the array data disks which can be useful if you want to later use the disk on another system.
July 27, 20178 yr I'd run it through another preclear while it is still empty - if the attributes are still good - you can use it as an unassigned disk or as an array disk with greater confidence.
July 28, 20178 yr Author Passed another round of preclear and values didn't change thanks for everyone's support!
Archived
This topic is now archived and is closed to further replies.