June 2, 201412 yr just completed my monthly parity check, and it came up with 0 errors, BUT i noticed on the main unraid page, beside drive 6 it shows 360 errors! not sure how long this has been here, haven't noticed it before. also thought it was weird that the month check report completed clean with 0 errors also out of the interested, the unmenu main page shows 0 errors beside drive 6... is this some kind of reporting glitch, or should i check something out? edit here's the smart report, doesn't look too good smartm.txt
June 2, 201412 yr Try running a SMART report on that drive... Its probably nothing...a data error that the drive self corrected. It would be good to know that its happened, and if the 360 starts growing, then trouble is coming.
June 2, 201412 yr Errors in the error column mean that the drive has reported a read error back to the OS. When unRaid gets a read error it reconstructs the sector from the remaining disks and then writes the sector back to the original disk. In theory this should make the disk remap the sector. During a parity check you might say this behavior sort of defeats the purpose as the one bad sector is reconstructed before the parity is checked - so it can't be wrong. Your smart report is showing 32 pending sectors. This is not a healthy condition. Your array is protecting you. I would continue to run parity checks (uncorrecting) and running smart reports. I am not sure if unRaid is logging the locations of the errors but if so I would track them in a spreadsheet. There are some tools that will read a disk and force sector remapping. I have used one called spinrite but it is old and I've had problems with it on larger drives (and larger means 200G drives!). I doubt it would even try to work on anything above 2T. But there may be other newer tools. The manufacturer diagnostics may also be able to repair the damage. The bigger question may be is this a specific problem with these few sectors or the beginning of the drive failing. Often my experience is that once sectors start to fail its like a small pothole that gets bigger and bigger as cars go by. Before long the pothole gets so big as to make the road impassable. Only with further testing and tracking of the number of errant sectors would you be able to tell. Good luck!
June 3, 201412 yr Author as a general question, do folks run smart reports regularly and check for pending counts? i guess i have only really done that when unraid shows an error on the main page, or when the monthly parity check shows an error i'm now thinking that if two drives start going bad, and their pending sector counts are growing, unraid doesn't really communicate this. i also seem to recall reading that unraid may not survive rebuilding a dead drive IF one of the array drives has pending counts or errors on it. so i'm thinking that if 2 or more drives are in the process of going bad, does that mean i don't really have parity, because when one of them dies, it can't be rebuilt because of the failing health of other drives? sorry if its a dumb question!
June 3, 201412 yr so i'm thinking that if 2 or more drives are in the process of going bad, does that mean i don't really have parity, because when one of them dies, it can't be rebuilt because of the failing health of other drives? sorry if its a dumb question! Not at dumb question at all. In my early days of unraid, I had the mistaken notion that because unraid could rebuild a failed drive, it was ok to leave a questionable drive in use. WRONG. It bit me big time, when another drive decided to fail, and I couldn't rebuild it because the questionable drive decided the stress of the rebuild was too much for it, and it quit. Monthly parity checks are a good way to help uncover a bad drive before it totally fails, but if you don't proactively monitor the situation and quickly replace a drive that looks iffy, you are asking for trouble. I always pull smart reports after every parity check, and if a drive is starting to show problems, it's replaced. I'll generally run several preclear cycles on the drive I pulled, and if it gets better, it can live on in a stack of spares.
June 4, 201412 yr People often think that parity can rebuild one failed disk. This is not true. Parity in combination with every other disk in the array being healthy and not altered outside of the array is necessary for a rebuild. People that get into troublesome recovery situation often say they never understood that - so yours is a very good question!
July 4, 201412 yr Author sorry to resurrect a slightly old thread, but i have been watching this drive and i now see that the pending sectors count has risen from 32, to 125. although the drive is out of warranty, i'm thinking its time to replace it. the drive is still readable....can anyone suggest the safest way to replace it? i ran a parity check and everything seems fine. its been a long time since i replaced a drive... should i replace the drive into the same slot as the dying drive, and let parity rebuild? or should i copy data off the drive to my windows machine, then install new drive, then copy back? the parity rebuild seems easier thoughts?
July 4, 201412 yr sorry to resurrect a slightly old thread, but i have been watching this drive and i now see that the pending sectors count has risen from 32, to 125. although the drive is out of warranty, i'm thinking its time to replace it. the drive is still readable....can anyone suggest the safest way to replace it? i ran a parity check and everything seems fine. its been a long time since i replaced a drive... should i replace the drive into the same slot as the dying drive, and let parity rebuild? or should i copy data off the drive to my windows machine, then install new drive, then copy back? the parity rebuild seems easier thoughts? If you have a drive with Pending Sectors, then if another drive fails it cannot be rebuilt with complete confidence as a Pending sector implies there was a read problem so the data returned could be unreliable for that sector and thus affect the value on another drive being rebuilt. It is possible that the Pending sectors will be cleared successfully the next time they are written to (or re-allocated) but it is still an indication of potential loss of data integrity. The safest thing to do is to simply replace the drive with a replacement and let unRAID rebuild onto it. That way you still have the original drive available if for any reason the rebuild fails. If the rebuild succeeds it can be worth putting the replaced drive through an extensive pre-clear cycle (assuming you have a SATA slot to put it into) to see if the drive really has issues. It is possible that after such an exercise the drive appears to be OK so that it can be kept around as a spare or as a drive suitable for offline backups.
July 4, 201412 yr Author EDIT: sorry for the below post, took about 20 minutes to start array and mount all disks. has never taken that long, seems weird. has now started and appears to be rebuilding. i will leave the below post in case any one enjoys reading my panic attack! ok seem to be having a problem... did backup of disk.cfg, and super.dat then stopped array, and powered down. then removed old drive, and put new precleared drive in same slot booted, gui showed up but for the 'new drive' it had a red dot and said 'no device'. i selected drop down box, and chose the new drive (this was the only item in the drop box). this was a bit different than the instructions i read, but seemed to be logical then i chose 'Start will bring the array on-line, start Data-Rebuild, and then expand the file system' so i clicked yes i want to do this at the bottom on the main page, i see spinning up all drives, starting array, mounting disks.... but that's it. is that normal? gui seems frozen, and i see none of my apps are running ie sab, plex etc. i thought the array would be usable during the rebuild there are a couple of flashing hd activity lights on the front of the server, but i thought they would all be flashing... lastly, i can reach the unmenu gui, and it shows disk 1 to 5 appearing normal, but then disk 6 says 'disk invalid (this is the new disk) then disk 7 to 20 all seem to be showing full green bars as if they were empty (which is not the case). if i had to guess it seems like the array started, tried to mount all disks but stopped at disk 6 any advice?
July 4, 201412 yr EDIT: sorry for the below post, took about 20 minutes to start array and mount all disks. has never taken that long, seems weird. has now started and appears to be rebuilding. i will leave the below post in case any one enjoys reading my panic attack! ok seem to be having a problem... did backup of disk.cfg, and super.dat then stopped array, and powered down. then removed old drive, and put new precleared drive in same slot booted, gui showed up but for the 'new drive' it had a red dot and said 'no device'. i selected drop down box, and chose the new drive (this was the only item in the drop box). this was a bit different than the instructions i read, but seemed to be logical then i chose 'Start will bring the array on-line, start Data-Rebuild, and then expand the file system' so i clicked yes i want to do this at the bottom on the main page, i see spinning up all drives, starting array, mounting disks.... but that's it. is that normal? gui seems frozen, and i see none of my apps are running ie sab, plex etc. i thought the array would be usable during the rebuild there are a couple of flashing hd activity lights on the front of the server, but i thought they would all be flashing... lastly, i can reach the unmenu gui, and it shows disk 1 to 5 appearing normal, but then disk 6 says 'disk invalid (this is the new disk) then disk 7 to 20 all seem to be showing full green bars as if they were empty (which is not the case). if i had to guess it seems like the array started, tried to mount all disks but stopped at disk 6 any advice? Do you have removable / hotswap cages, or did you open the case and physically replace the disk? It is very easy to knock something loose when replacing a failed disk, and this leads to headaches doing drive recoveries. Don't do anything crazy trying to recover. You are not in a bad spot. Give it some time - I have seen some drive operations take over 1/2 hour to complete before returning control to the Web GUI. I'd likely let it go an hour before considering shutting down. If you feel you may have knocked something loose, you can shutdown (if GUI is frozen the next best thing is to run the PowerDown script). Then carefully check your cabling and try again. The backup disk.cfg and super.dat have a very limited life, and I consider it very unlikely you would need to use them. If you feel you should, please post back to confirm. After the rebuild make sure to delete them as they can cause a lot of trouble if you tried to restore them after replacing the failing disk. Concerning the pending sectors going from 32 to 125, it is not unusual. Once a disk starts failing, it sometimes acts like a pothole and more and more sectors fall into that pothole. But the other thing I have seen is a drive builds up a fair number of pending sectors with none reallocated and suddenly the pending sectors go away with no ill effects. I saw this replay itself a few times, but as I recall it was with new disks. I agree it is smart to replace it, but with the process itimpi documents you can try to salvage it.
July 4, 201412 yr Author cool thanks for the advice. i did let things play out and the array and rebuild started after maybe 20min, which is longer than normal for my rig. i think part of the additional delay might have been the 2 or 3 minutes where the precleared drive is formatted prior to rebuild? in any case its in the process of rebuilding. once completed (if successful) i will embark on preclearing a new backup drive (i like to keep a precleared new drive on the shelf for cases like these) and then ill run a few preclears on the 'old' drive with the 125 pending counts
July 4, 201412 yr cool thanks for the advice. i did let things play out and the array and rebuild started after maybe 20min, which is longer than normal for my rig. i think part of the additional delay might have been the 2 or 3 minutes where the precleared drive is formatted prior to rebuild? in any case its in the process of rebuilding. once completed (if successful) i will embark on preclearing a new backup drive (i like to keep a precleared new drive on the shelf for cases like these) and then ill run a few preclears on the 'old' drive with the 125 pending counts Great!
July 7, 201412 yr Author hi all! so i have the above drive which had grown in its 'pending sectors'... i replaced it with a new drive, and rebuild the data yesterday i put it back in my system and ran a preclear cycle. here is the output: looks like the pending sectors dropped to only 2? any opinions on what to do with this drive? ========================================================================1.13 == invoked as: ./preclear_disk.sh -M 4 /dev/sdp == WDC WD20EARS-00MVWB0 WD-WMAZA0125200 == Disk /dev/sdp has been successfully precleared == with a starting sector of 64 == Ran 1 cycle == == Using :Read block size = 8225280 Bytes == Last Cycle's Pre Read Time : 8:02:17 (69 MB/s) == Last Cycle's Zeroing time : 6:07:04 (90 MB/s) == Last Cycle's Post Read Time : 14:10:12 (39 MB/s) == Last Cycle's Total Time : 28:20:37 == == Total Elapsed Time 28:20:37 == == Disk Start Temperature: 31C == == Current Disk Temperature: 30C, == ============================================================================ ** Changed attributes in files: /tmp/smart_start_sdp /tmp/smart_finish_sdp ATTRIBUTE NEW_VAL OLD_VAL FAILURE_THRESHOLD STATUS RAW_VALUE Seek_Error_Rate = 100 200 0 ok 0 Temperature_Celsius = 120 119 0 ok 30 No SMART attributes are FAILING_NOW 125 sectors were pending re-allocation before the start of the preclear. 125 sectors were pending re-allocation after pre-read in cycle 1 of 1. 0 sectors were pending re-allocation after zero of disk in cycle 1 of 1. 2 sectors are pending re-allocation at the end of the preclear, a change of -123 in the number of sectors pending re-allocation. 0 sectors had been re-allocated before the start of the preclear. 0 sectors are re-allocated at the end of the preclear, the number of sectors re-allocated did not change. ============================================================================ ============================================================================ == == S.M.A.R.T Initial Report for /dev/sdp == Disk: /dev/sdp smartctl 5.40 2010-10-16 r3189 [i486-slackware-linux-gnu] (local build) Copyright © 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net === START OF INFORMATION SECTION === Model Family: Western Digital Caviar Green (Adv. Format) family Device Model: WDC WD20EARS-00MVWB0 Serial Number: WD-WMAZA0125200 Firmware Version: 50.0AB50 User Capacity: 2,000,398,934,016 bytes Device is: In smartctl database [for details use: -P show] ATA Version is: 8 ATA Standard is: Exact ATA specification draft version not indicated Local Time is: Sat Jul 5 15:21:19 2014 CDT SMART support is: Available - device has SMART capability. SMART support is: Enabled === START OF READ SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED General SMART Values: Offline data collection status: (0x84) Offline data collection activity was suspended by an interrupting command from host. Auto Offline Data Collection: Enabled. Self-test execution status: ( 0) The previous self-test routine completed without error or no self-test has ever been run. Total time to complete Offline data collection: (36600) seconds. Offline data collection capabilities: (0x7b) SMART execute Offline immediate. Auto Offline data collection on/off support. Suspend Offline collection upon new command. Offline surface scan supported. Self-test supported. Conveyance Self-test supported. Selective Self-test supported. SMART capabilities: (0x0003) Saves SMART data before entering power-saving mode. Supports SMART auto save timer. Error logging capability: (0x01) Error logging supported. General Purpose Logging supported. Short self-test routine recommended polling time: ( 2) minutes. Extended self-test routine recommended polling time: ( 255) minutes. Conveyance self-test routine recommended polling time: ( 5) minutes. SCT capabilities: (0x3035) SCT Status supported. SCT Feature Control supported. SCT Data Table supported. SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 200 200 051 Pre-fail Always - 112 3 Spin_Up_Time 0x0027 174 165 021 Pre-fail Always - 6300 4 Start_Stop_Count 0x0032 094 094 000 Old_age Always - 6704 5 Reallocated_Sector_Ct 0x0033 200 200 140 Pre-fail Always - 0 7 Seek_Error_Rate 0x002e 200 200 000 Old_age Always - 0 9 Power_On_Hours 0x0032 064 064 000 Old_age Always - 26448 10 Spin_Retry_Count 0x0032 100 100 000 Old_age Always - 0 11 Calibration_Retry_Count 0x0032 100 100 000 Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 852 192 Power-Off_Retract_Count 0x0032 199 199 000 Old_age Always - 800 193 Load_Cycle_Count 0x0032 163 163 000 Old_age Always - 111536 194 Temperature_Celsius 0x0022 119 115 000 Old_age Always - 31 196 Reallocated_Event_Count 0x0032 200 200 000 Old_age Always - 0 197 Current_Pending_Sector 0x0032 200 200 000 Old_age Always - 125 198 Offline_Uncorrectable 0x0030 200 200 000 Old_age Offline - 1 199 UDMA_CRC_Error_Count 0x0032 200 200 000 Old_age Always - 26 200 Multi_Zone_Error_Rate 0x0008 200 199 000 Old_age Offline - 34 SMART Error Log Version: 1 No Errors Logged SMART Self-test log structure revision number 1 No self-tests have been logged. [To run self-tests, use: smartctl -t] SMART Selective self-test log data structure revision number 1 SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS 1 0 0 Not_testing 2 0 0 Not_testing 3 0 0 Not_testing 4 0 0 Not_testing 5 0 0 Not_testing Selective self-test flags (0x0): After scanning selected spans, do NOT read-scan remainder of disk. If Selective self-test is pending on power-up, resume after 0 minute delay. == ============================================================================ ============================================================================ == == S.M.A.R.T Final Report for /dev/sdp == Disk: /dev/sdp smartctl 5.40 2010-10-16 r3189 [i486-slackware-linux-gnu] (local build) Copyright © 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net === START OF INFORMATION SECTION === Model Family: Western Digital Caviar Green (Adv. Format) family Device Model: WDC WD20EARS-00MVWB0 Serial Number: WD-WMAZA0125200 Firmware Version: 50.0AB50 User Capacity: 2,000,398,934,016 bytes Device is: In smartctl database [for details use: -P show] ATA Version is: 8 ATA Standard is: Exact ATA specification draft version not indicated Local Time is: Sun Jul 6 19:41:55 2014 CDT SMART support is: Available - device has SMART capability. SMART support is: Enabled === START OF READ SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED General SMART Values: Offline data collection status: (0x84) Offline data collection activity was suspended by an interrupting command from host. Auto Offline Data Collection: Enabled. Self-test execution status: ( 0) The previous self-test routine completed without error or no self-test has ever been run. Total time to complete Offline data collection: (36600) seconds. Offline data collection capabilities: (0x7b) SMART execute Offline immediate. Auto Offline data collection on/off support. Suspend Offline collection upon new command. Offline surface scan supported. Self-test supported. Conveyance Self-test supported. Selective Self-test supported. SMART capabilities: (0x0003) Saves SMART data before entering power-saving mode. Supports SMART auto save timer. Error logging capability: (0x01) Error logging supported. General Purpose Logging supported. Short self-test routine recommended polling time: ( 2) minutes. Extended self-test routine recommended polling time: ( 255) minutes. Conveyance self-test routine recommended polling time: ( 5) minutes. SCT capabilities: (0x3035) SCT Status supported. SCT Feature Control supported. SCT Data Table supported. SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 200 200 051 Pre-fail Always - 285 3 Spin_Up_Time 0x0027 174 165 021 Pre-fail Always - 6300 4 Start_Stop_Count 0x0032 094 094 000 Old_age Always - 6704 5 Reallocated_Sector_Ct 0x0033 200 200 140 Pre-fail Always - 0 7 Seek_Error_Rate 0x002e 100 253 000 Old_age Always - 0 9 Power_On_Hours 0x0032 064 064 000 Old_age Always - 26475 10 Spin_Retry_Count 0x0032 100 100 000 Old_age Always - 0 11 Calibration_Retry_Count 0x0032 100 100 000 Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 852 192 Power-Off_Retract_Count 0x0032 199 199 000 Old_age Always - 800 193 Load_Cycle_Count 0x0032 163 163 000 Old_age Always - 111536 194 Temperature_Celsius 0x0022 120 115 000 Old_age Always - 30 196 Reallocated_Event_Count 0x0032 200 200 000 Old_age Always - 0 197 Current_Pending_Sector 0x0032 200 200 000 Old_age Always - 2 198 Offline_Uncorrectable 0x0030 200 200 000 Old_age Offline - 1 199 UDMA_CRC_Error_Count 0x0032 200 200 000 Old_age Always - 26 200 Multi_Zone_Error_Rate 0x0008 200 199 000 Old_age Offline - 34 SMART Error Log Version: 1 No Errors Logged SMART Self-test log structure revision number 1 No self-tests have been logged. [To run self-tests, use: smartctl -t] SMART Selective self-test log data structure revision number 1 SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS 1 0 0 Not_testing 2 0 0 Not_testing 3 0 0 Not_testing 4 0 0 Not_testing 5 0 0 Not_testing Selective self-test flags (0x0): After scanning selected spans, do NOT read-scan remainder of disk. If Selective self-test is pending on power-up, resume after 0 minute delay. == ============================================================================
July 7, 201412 yr hi all! so i have the above drive which had grown in its 'pending sectors'... i replaced it with a new drive, and rebuild the data yesterday i put it back in my system and ran a preclear cycle. here is the output: looks like the pending sectors dropped to only 2? any opinions on what to do with this drive? Run another preclear and see if it gets better or worse.
July 8, 201411 yr Author okay so here's my second pre-clear cycle... looks like the remaining two pending sectors have now been reallocated! in fact i think the final diagnostics look pretty good! except i realize that the power on hours is fairly high (although i have higher ones in my server) not sure what to do about this... i guess options are adding back array and watching, or else putting in my pc, copying over some stuff and using it for offline storage i guess i thought this second preclear would show its weaknesses, but it looks a bit like its 'fixed'? any thoughts? ========================================================================1.13 == invoked as: ./preclear_disk.sh -M 4 /dev/sdp == WDC WD20EARS-00MVWB0 WD-WMAZA0125200 == Disk /dev/sdp has been successfully precleared == with a starting sector of 64 == Ran 1 cycle == == Using :Read block size = 8225280 Bytes == Last Cycle's Pre Read Time : 6:42:24 (82 MB/s) == Last Cycle's Zeroing time : 6:06:23 (90 MB/s) == Last Cycle's Post Read Time : 13:57:26 (39 MB/s) == Last Cycle's Total Time : 26:47:15 == == Total Elapsed Time 26:47:15 == == Disk Start Temperature: 31C == == Current Disk Temperature: 32C, == ============================================================================ ** Changed attributes in files: /tmp/smart_start_sdp /tmp/smart_finish_sdp ATTRIBUTE NEW_VAL OLD_VAL FAILURE_THRESHOLD STATUS RAW_VALUE Temperature_Celsius = 118 119 0 ok 32 No SMART attributes are FAILING_NOW 2 sectors were pending re-allocation before the start of the preclear. 3 sectors were pending re-allocation after pre-read in cycle 1 of 1. 0 sectors were pending re-allocation after zero of disk in cycle 1 of 1. 0 sectors are pending re-allocation at the end of the preclear, a change of -2 in the number of sectors pending re-allocation. 0 sectors had been re-allocated before the start of the preclear. 0 sectors are re-allocated at the end of the preclear, the number of sectors re-allocated did not change. ============================================================================ ============================================================================ == == S.M.A.R.T Initial Report for /dev/sdp == Disk: /dev/sdp smartctl 5.40 2010-10-16 r3189 [i486-slackware-linux-gnu] (local build) Copyright © 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net === START OF INFORMATION SECTION === Model Family: Western Digital Caviar Green (Adv. Format) family Device Model: WDC WD20EARS-00MVWB0 Serial Number: WD-WMAZA0125200 Firmware Version: 50.0AB50 User Capacity: 2,000,398,934,016 bytes Device is: In smartctl database [for details use: -P show] ATA Version is: 8 ATA Standard is: Exact ATA specification draft version not indicated Local Time is: Sun Jul 6 20:26:32 2014 CDT SMART support is: Available - device has SMART capability. SMART support is: Enabled === START OF READ SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED General SMART Values: Offline data collection status: (0x84) Offline data collection activity was suspended by an interrupting command from host. Auto Offline Data Collection: Enabled. Self-test execution status: ( 0) The previous self-test routine completed without error or no self-test has ever been run. Total time to complete Offline data collection: (36600) seconds. Offline data collection capabilities: (0x7b) SMART execute Offline immediate. Auto Offline data collection on/off support. Suspend Offline collection upon new command. Offline surface scan supported. Self-test supported. Conveyance Self-test supported. Selective Self-test supported. SMART capabilities: (0x0003) Saves SMART data before entering power-saving mode. Supports SMART auto save timer. Error logging capability: (0x01) Error logging supported. General Purpose Logging supported. Short self-test routine recommended polling time: ( 2) minutes. Extended self-test routine recommended polling time: ( 255) minutes. Conveyance self-test routine recommended polling time: ( 5) minutes. SCT capabilities: (0x3035) SCT Status supported. SCT Feature Control supported. SCT Data Table supported. SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 200 200 051 Pre-fail Always - 285 3 Spin_Up_Time 0x0027 174 165 021 Pre-fail Always - 6300 4 Start_Stop_Count 0x0032 094 094 000 Old_age Always - 6704 5 Reallocated_Sector_Ct 0x0033 200 200 140 Pre-fail Always - 0 7 Seek_Error_Rate 0x002e 100 253 000 Old_age Always - 0 9 Power_On_Hours 0x0032 064 064 000 Old_age Always - 26476 10 Spin_Retry_Count 0x0032 100 100 000 Old_age Always - 0 11 Calibration_Retry_Count 0x0032 100 100 000 Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 852 192 Power-Off_Retract_Count 0x0032 199 199 000 Old_age Always - 800 193 Load_Cycle_Count 0x0032 163 163 000 Old_age Always - 111544 194 Temperature_Celsius 0x0022 119 115 000 Old_age Always - 31 196 Reallocated_Event_Count 0x0032 200 200 000 Old_age Always - 0 197 Current_Pending_Sector 0x0032 200 200 000 Old_age Always - 2 198 Offline_Uncorrectable 0x0030 200 200 000 Old_age Offline - 1 199 UDMA_CRC_Error_Count 0x0032 200 200 000 Old_age Always - 26 200 Multi_Zone_Error_Rate 0x0008 200 199 000 Old_age Offline - 34 SMART Error Log Version: 1 No Errors Logged SMART Self-test log structure revision number 1 No self-tests have been logged. [To run self-tests, use: smartctl -t] SMART Selective self-test log data structure revision number 1 SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS 1 0 0 Not_testing 2 0 0 Not_testing 3 0 0 Not_testing 4 0 0 Not_testing 5 0 0 Not_testing Selective self-test flags (0x0): After scanning selected spans, do NOT read-scan remainder of disk. If Selective self-test is pending on power-up, resume after 0 minute delay. == ============================================================================ ============================================================================ == == S.M.A.R.T Final Report for /dev/sdp == Disk: /dev/sdp smartctl 5.40 2010-10-16 r3189 [i486-slackware-linux-gnu] (local build) Copyright © 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net === START OF INFORMATION SECTION === Model Family: Western Digital Caviar Green (Adv. Format) family Device Model: WDC WD20EARS-00MVWB0 Serial Number: WD-WMAZA0125200 Firmware Version: 50.0AB50 User Capacity: 2,000,398,934,016 bytes Device is: In smartctl database [for details use: -P show] ATA Version is: 8 ATA Standard is: Exact ATA specification draft version not indicated Local Time is: Mon Jul 7 23:13:46 2014 CDT SMART support is: Available - device has SMART capability. SMART support is: Enabled === START OF READ SMART DATA SECTION === SMART overall-health self-assessment test result: PASSED General SMART Values: Offline data collection status: (0x84) Offline data collection activity was suspended by an interrupting command from host. Auto Offline Data Collection: Enabled. Self-test execution status: ( 0) The previous self-test routine completed without error or no self-test has ever been run. Total time to complete Offline data collection: (36600) seconds. Offline data collection capabilities: (0x7b) SMART execute Offline immediate. Auto Offline data collection on/off support. Suspend Offline collection upon new command. Offline surface scan supported. Self-test supported. Conveyance Self-test supported. Selective Self-test supported. SMART capabilities: (0x0003) Saves SMART data before entering power-saving mode. Supports SMART auto save timer. Error logging capability: (0x01) Error logging supported. General Purpose Logging supported. Short self-test routine recommended polling time: ( 2) minutes. Extended self-test routine recommended polling time: ( 255) minutes. Conveyance self-test routine recommended polling time: ( 5) minutes. SCT capabilities: (0x3035) SCT Status supported. SCT Feature Control supported. SCT Data Table supported. SMART Attributes Data Structure revision number: 16 Vendor Specific SMART Attributes with Thresholds: ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 1 Raw_Read_Error_Rate 0x002f 200 200 051 Pre-fail Always - 342 3 Spin_Up_Time 0x0027 174 165 021 Pre-fail Always - 6300 4 Start_Stop_Count 0x0032 094 094 000 Old_age Always - 6704 5 Reallocated_Sector_Ct 0x0033 200 200 140 Pre-fail Always - 0 7 Seek_Error_Rate 0x002e 200 200 000 Old_age Always - 0 9 Power_On_Hours 0x0032 064 064 000 Old_age Always - 26502 10 Spin_Retry_Count 0x0032 100 100 000 Old_age Always - 0 11 Calibration_Retry_Count 0x0032 100 100 000 Old_age Always - 0 12 Power_Cycle_Count 0x0032 100 100 000 Old_age Always - 852 192 Power-Off_Retract_Count 0x0032 199 199 000 Old_age Always - 800 193 Load_Cycle_Count 0x0032 163 163 000 Old_age Always - 111545 194 Temperature_Celsius 0x0022 118 115 000 Old_age Always - 32 196 Reallocated_Event_Count 0x0032 200 200 000 Old_age Always - 0 197 Current_Pending_Sector 0x0032 200 200 000 Old_age Always - 0 198 Offline_Uncorrectable 0x0030 200 200 000 Old_age Offline - 1 199 UDMA_CRC_Error_Count 0x0032 200 200 000 Old_age Always - 26 200 Multi_Zone_Error_Rate 0x0008 200 199 000 Old_age Offline - 34 SMART Error Log Version: 1 No Errors Logged SMART Self-test log structure revision number 1 No self-tests have been logged. [To run self-tests, use: smartctl -t] SMART Selective self-test log data structure revision number 1 SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS 1 0 0 Not_testing 2 0 0 Not_testing 3 0 0 Not_testing 4 0 0 Not_testing 5 0 0 Not_testing Selective self-test flags (0x0): After scanning selected spans, do NOT read-scan remainder of disk. If Selective self-test is pending on power-up, resume after 0 minute delay. == ============================================================================
July 8, 201411 yr okay so here's my second pre-clear cycle... looks like the remaining two pending sectors have now been reallocated! in fact i think the final diagnostics look pretty good! except i realize that the power on hours is fairly high (although i have higher ones in my server) not sure what to do about this... i guess options are adding back array and watching, or else putting in my pc, copying over some stuff and using it for offline storage i guess i thought this second preclear would show its weaknesses, but it looks a bit like its 'fixed'? any thoughts? Sounds like you have a pretty solid grasp on what happened and what it means, so I'll just agree with you. Your call on how much you trust it, and by extension, whether it goes back in the array. Only known good drives should live in the array, that way you have the best shot of recovering from the inevitable surprise failure. Whether or not this drive is array worthy with a now clean smart report is up to you.
July 8, 201411 yr I have been monitoring this thread and the information is accurate relative to how unRaid handles read and write errors and indirectly forces remapping of sectors. As to when to trust the drive it is a personal decision. My rule of thumb is to run 3 preclears and if the allocated / reallocated sector picture doesn't change, and nothing new goes sideways, I'd trust it back in my array, although I'd continue to monitor it. It would actually be more dangerous to use it in an unprotected fashion!
July 8, 201411 yr It would actually be more dangerous to use it in an unprotected fashion!Here is where we differ. If a drive has shown a propensity to generate errors, I'd rather it be an offline backup, that way if it fails I've still got my files elsewhere. If it fails while you are trying to rebuild another failed drive, you are hosed on the contents of both drives, and have to restore from backup. Now, we can argue whether the drive you have already seen problems with and recovered with multiple preclear passes is more reliable than one that hasn't failed yet. All drives fail, it's just a matter of when. With unraid we just hope the odds are in our favor as to the timing.
July 8, 201411 yr It would actually be more dangerous to use it in an unprotected fashion!Here is where we differ. If a drive has shown a propensity to generate errors, I'd rather it be an offline backup, that way if it fails I've still got my files elsewhere. If it fails while you are trying to rebuild another failed drive, you are hosed on the contents of both drives, and have to restore from backup. Now, we can argue whether the drive you have already seen problems with and recovered with multiple preclear passes is more reliable than one that hasn't failed yet. All drives fail, it's just a matter of when. With unraid we just hope the odds are in our favor as to the timing. Sorry I wasn't tracking with your scenario exactly. My point is if the drive were to fail outside of redundancy you would lose data on that drive. If it were in the array, unRaid could force remappings or reconstruct it. But you're right, if a second drive were to fail you would be in trouble. You could use even a questionable drive to store an extra backup. If the alternative is not to have the extra backup then it is better than nothing. But if you're aft this level of distrust on a drive it would be better to RMA it if that is an option.
Archived
This topic is now archived and is closed to further replies.