JorgeB Posted March 23, 2016 Share Posted March 23, 2016 Thank you Johnnie! So I interpret it as the drive came with the attributes this way, close to the threshold, from Seagate and just barely passed internal tests for e.g. end-to-end error. Yes, but that those values are the same on all Seagates (or at least most recent models), it's not because of internal testing. Quote Link to comment
skylark Posted March 23, 2016 Share Posted March 23, 2016 Wow .... 195 hours for a preclear. Kinda rules out 3 cycles i'd guess ... Quote Link to comment
itimpi Posted March 23, 2016 Share Posted March 23, 2016 Wow .... 195 hours for a preclear. Kinda rules out 3 cycles i'd guess ... Looking at the time shown for the individual phases the 195 total must be for more than one cycle . The individual phase times are about what I see on my system with the 8TB drives so seem perfectly normal. With larger drives the elapsed time gain from the 'faster' preclear are quite significant. Quote Link to comment
motorskalle Posted March 23, 2016 Share Posted March 23, 2016 Thank you Johnnie! So I interpret it as the drive came with the attributes this way, close to the threshold, from Seagate and just barely passed internal tests for e.g. end-to-end error. Yes, but that those values are the same on all Seagates (or at least most recent models), it's not because of internal testing. Thanks, that was the answer I was looking for! @itimpi and @skylark: It was a 3-cycle preclear. Quote Link to comment
skylark Posted March 23, 2016 Share Posted March 23, 2016 Good to hear it was 3 passes and not just one!! Quote Link to comment
skylark Posted March 24, 2016 Share Posted March 24, 2016 My Seagate 8TB preclear has just finished and i think there MAY be a problem. Although passing SMART there is an attribute 'Reported_Uncorrect 0x0032 083 083 000 Old_age Always - 17' which is being highlighted in orange? I've attached the reports, this is the second independant preclear cycle of the disk which started with a threshold of '100' now down to '083'. Busy running an extended SMART test as i write but what do you think?? This disk was earmarked for data, i have already precleared and added a second disk to the array as parity, its SMART was clean. I have three days left to RMA but given that it's going to be pressed into use as a data disk maybe i shouldnt be worrying (??), if its use was as a second parity disk i would be worried! preclear_reports.zip Quote Link to comment
trurl Posted March 24, 2016 Share Posted March 24, 2016 ...given that it's going to be pressed into use as a data disk maybe i shouldnt be worrying (??), if its use was as a second parity disk i would be worried! Sounds like you may have some misconceptions about parity. Parity by itself won't allow you to recover any of the data from a failed drive. Parity plus all the other data drives are needed to rebuild a failed data disk. And parity doesn't actually contain any of your data, but a data disk does. So actually parity is slightly less important than a data disk. Quote Link to comment
skylark Posted March 24, 2016 Share Posted March 24, 2016 So ... understood. Any comment to make on the overall health of the disk? Quote Link to comment
JorgeB Posted March 24, 2016 Share Posted March 24, 2016 I believe that attribute does not mean the disk is bad for sure, and I've seen many disks here on the forum, especially Seagates, with much larger values and still ok, but it's certainly not good, if you can get a new one go for it. Quote Link to comment
skylark Posted March 24, 2016 Share Posted March 24, 2016 Thanks. Looks as though respected authorities say anything > 0 is bad. Drive RMA'd. Another 140hrs wasted. Quote Link to comment
RobJ Posted March 24, 2016 Share Posted March 24, 2016 *Something* happened at hour 158, but I don't know what. At least 5 of the 17 (probably all 17) occurred at 158, and they were UNC's but the message is the same for each, and the sector is unspecified (perhaps unknown?). It shows only as 0xfffffff, which is almost certainly not a valid sector number. If you did not see any other evidence of bad sectors, then I would ignore it. However, because it happened with the drive so new, it's certainly important, and worth monitoring closely. Just a comment about the BackBlaze work - they have been great to release so much data, and their findings and recommendations about drive quality are invaluable, absolutely worth relying on. Their SMART data though is not as reliable, and shows some issues. While I'm sure they have people that know what they are doing, the one(s) that provided the SMART charts and graphs did not fully understand actual SMART usage and made some embarrassing decisions. That has made it hard for me to fully trust *some* of their work, and the interpretations derived from it. Quote Link to comment
skylark Posted March 24, 2016 Share Posted March 24, 2016 Thanks for taking a look. That might tie in with its 'history'. I started preclearing it and overnight it just ended. The array hadnt restarted, just the preclear had disappeared the next morning, no reports, nothing. Its had two preclear cycles since then with the same results. Think i mentioned it on gfjardim's preclear plugin thread. I'm busy running an extended smart test now so we'll see what comes of that but if i am upfront with the vendor about the nature of the fault then i might get to rma it. Shouldnt be in that state. Its a brand new drive. In warramty and outside of initial 30 day warranty i might have lived with it for a while but i want it to be good from the get go. Quote Link to comment
RobJ Posted March 24, 2016 Share Posted March 24, 2016 One small comment about that 8TB drive - it's not yet in the SMART database (drivedb.h), so it didn't interpret Command Timeout correctly. It shows a RAW value of 65537, up from 0, which *looks* like a huge increase. But correctly interpreted, 65537 is 1,1,0 (or possibly 0,1,1) - a single command timeout, an increase of one. Quote Link to comment
skylark Posted March 24, 2016 Share Posted March 24, 2016 Thanks again but that one passed me by .... Quote Link to comment
wgstarks Posted March 24, 2016 Share Posted March 24, 2016 Another 140hrs wasted. Not wasted. Saved you from getting stuck with a possibly defective drive. Things could have turned out much worse. Quote Link to comment
skylark Posted March 24, 2016 Share Posted March 24, 2016 I guess so ... let's stay positive Quote Link to comment
BobPhoenix Posted March 24, 2016 Share Posted March 24, 2016 I guess so ... let's stay positive That's always hard to do. During the drive shortage because of the floods in Asia a couple of years ago I had to return 6 drives in a row to the shipper before I just decided to settle on a Refurbished drive from the manufacturer. And by 6 drives I mean I RMA'd some drives and had to return the RMA'd drives as well. It didn't help that the shipper thought all they had to do was drop OEM drives in a HUGE box without packing material. Quote Link to comment
skylark Posted March 24, 2016 Share Posted March 24, 2016 One small comment about that 8TB drive - it's not yet in the SMART database (drivedb.h), so it didn't interpret Command Timeout correctly. It shows a RAW value of 65537, up from 0, which *looks* like a huge increase. But correctly interpreted, 65537 is 1,1,0 (or possibly 0,1,1) - a single command timeout, an increase of one. RobJ ... can you expand on that one for me please? Update ... the exetended SMART test passed successfully with no increment to the figures so there's still 17 at attribute 187. I dont understand what 188 is trying to tell me at 65537 and why is that attribute not a notifiable one in the SMART settings of unRAID? Quote Link to comment
cyberstyx Posted March 25, 2016 Share Posted March 25, 2016 Started a Preclear for 3 HDDs simultaneously, on a WD 2TB RE4 + a WD 2TB Caviar Green on a DIGITUS 30104-1 and on another WD 2TB RE4 on a secondary same controller. All disks were bought a few years ago and have had some operation on a VMWare ESXi host at home, which are now been migrated to unRAID. One WD 2TB RE passed preclear, the one that was the only HDD connected on a controller. The other two failed as following: - WD 2TB RE4 (SMART says it has 4 months of operation) == Disk /dev/sde has NOT been precleared successfully == skip=461200 count=200 bs=1000448 returned 16433 instead of 00000 - WD 2TB Caviar Green (SMART says it has 5 years of operation - oh oh...) == Disk /dev/sdf has NOT been precleared successfully == skip=1478200 count=200 bs=1000448 returned 00065 instead of 00000 skip=1603800 count=200 bs=1000448 returned 00005 instead of 00000 skip=1604400 count=200 bs=1000448 returned 00065 instead of 00000 I then found out about this post and some preclear general thoughts, so I am now running again a single Preclear on the WD 2TB RE4 that failed, as preclearing many drives simultaneously may give strange results when reading data back from disks during testing. I 'm doing single passes btw. Will report back tomorrow when preclearing has finished. (Damn that Caviar Green one is sooooo slow with 5400rpms ) Preclear_Finish_WDCWD2003FYYS-02W0B1_WD-WMAY04480522_2016-03-24.txt Preclear_Finish_WDCWD20EADS-00S2B0_WD-WCAVY0562655_2016-03-24.txt Quote Link to comment
cyberstyx Posted March 26, 2016 Share Posted March 26, 2016 Started a Preclear for 3 HDDs simultaneously, on a WD 2TB RE4 + a WD 2TB Caviar Green on a DIGITUS 30104-1 and on another WD 2TB RE4 on a secondary same controller. All disks were bought a few years ago and have had some operation on a VMWare ESXi host at home, which are now been migrated to unRAID. One WD 2TB RE passed preclear, the one that was the only HDD connected on a controller. The other two failed as following: - WD 2TB RE4 (SMART says it has 4 months of operation) == Disk /dev/sde has NOT been precleared successfully == skip=461200 count=200 bs=1000448 returned 16433 instead of 00000 - WD 2TB Caviar Green (SMART says it has 5 years of operation - oh oh...) == Disk /dev/sdf has NOT been precleared successfully == skip=1478200 count=200 bs=1000448 returned 00065 instead of 00000 skip=1603800 count=200 bs=1000448 returned 00005 instead of 00000 skip=1604400 count=200 bs=1000448 returned 00065 instead of 00000 I then found out about this post and some preclear general thoughts, so I am now running again a single Preclear on the WD 2TB RE4 that failed, as preclearing many drives simultaneously may give strange results when reading data back from disks during testing. I 'm doing single passes btw. Will report back tomorrow when preclearing has finished. (Damn that Caviar Green one is sooooo slow with 5400rpms ) Beh... == WDCWD2003FYYS-02W0B1 WD-WMAY04480522 == Disk /dev/sde has NOT been precleared successfully == skip=1131600 count=200 bs=1000448 returned 00049 instead of 00000 skip=1131800 count=200 bs=1000448 returned 00003 instead of 00000 skip= 1363800 count=200 bs=1000448 returned 00101 instead of 00000 skip=1829600 count=200 bs=1000448 returned 16489 instead of 00000 So, like, this disk will never be added in the array and I can use it only in Windows? Quote Link to comment
JorgeB Posted March 26, 2016 Share Posted March 26, 2016 Beh... == WDCWD2003FYYS-02W0B1 WD-WMAY04480522 == Disk /dev/sde has NOT been precleared successfully == skip=1131600 count=200 bs=1000448 returned 00049 instead of 00000 skip=1131800 count=200 bs=1000448 returned 00003 instead of 00000 skip= 1363800 count=200 bs=1000448 returned 00101 instead of 00000 skip=1829600 count=200 bs=1000448 returned 16489 instead of 00000 So, like, this disk will never be added in the array and I can use it only in Windows? Don't remember seeing that error before, you can try the new beta plugin: http://lime-technology.com/forum/index.php?topic=39985.msg453938#msg453938 Quote Link to comment
trurl Posted March 26, 2016 Share Posted March 26, 2016 I had preclear fail on post-read once due to bad memory. You might try a memtest just to eliminate that possibility. Quote Link to comment
Parigot Posted April 7, 2016 Share Posted April 7, 2016 I just received a WD RED 3TB drive that I will use to replace my 1TB parity drive. I ran a preclear on it twice and got the same result. But I don't know what is the issue (if it is one). Any help would be greatly appreciated. Thanks ================================================================== 1.15 = unRAID server Pre-Clear disk /dev/sdb = cycle 1 of 1, partition start on sector 1 = = Step 1 of 10 - Copying zeros to first 2048k bytes DONE = Step 2 of 10 - Copying zeros to remainder of disk to clear it DONE = Step 3 of 10 - Disk is now cleared from MBR onward. DONE = Step 4 of 10 - Clearing MBR bytes for partition 2,3 & 4 DONE = Step 5 of 10 - Clearing MBR code area DONE = Step 6 of 10 - Setting MBR signature bytes DONE = Step 7 of 10 - Setting partition 1 to precleared state DONE = Step 8 of 10 - Notifying kernel we changed the partitioning DONE = Step 9 of 10 - Creating the /dev/disk/by* entries DONE = Step 10 of 10 - Verifying if the MBR is cleared. DONE = Elapsed Time: 5:39:04 ========================================================================1.15 == == SORRY: Disk /dev/sdb MBR could NOT be precleared == == out4= 00000 == out5= 00000 ============================================================================ 0000000 Quote Link to comment
BobPhoenix Posted April 7, 2016 Share Posted April 7, 2016 I just received a WD RED 3TB drive that I will use to replace my 1TB parity drive. I ran a preclear on it twice and got the same result. But I don't know what is the issue (if it is one). Any help would be greatly appreciated. Thanks ================================================================== 1.15 = unRAID server Pre-Clear disk /dev/sdb = cycle 1 of 1, partition start on sector 1 = = Step 1 of 10 - Copying zeros to first 2048k bytes DONE = Step 2 of 10 - Copying zeros to remainder of disk to clear it DONE = Step 3 of 10 - Disk is now cleared from MBR onward. DONE = Step 4 of 10 - Clearing MBR bytes for partition 2,3 & 4 DONE = Step 5 of 10 - Clearing MBR code area DONE = Step 6 of 10 - Setting MBR signature bytes DONE = Step 7 of 10 - Setting partition 1 to precleared state DONE = Step 8 of 10 - Notifying kernel we changed the partitioning DONE = Step 9 of 10 - Creating the /dev/disk/by* entries DONE = Step 10 of 10 - Verifying if the MBR is cleared. DONE = Elapsed Time: 5:39:04 ========================================================================1.15 == == SORRY: Disk /dev/sdb MBR could NOT be precleared == == out4= 00000 == out5= 00000 ============================================================================ 0000000 I got that when I needed to update the firmware of my SAS Expander. But I think it could be almost anything. Maybe someone else has more ideas. Attaching a Syslog (v5 unRAID) or Diagnostics (v6 unRAID) would help someone else as well. Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.