MMChris Posted April 3, 2020 Share Posted April 3, 2020 So i started my Unraid server today. Installation was very easy and it picked up all my hardware. My array consists of 4TB (new wd red) as parity 3 x 3TB drives (wd red from freenas build 2 years run time) as array. 500GB SSD samsung 860 evo as the Cache ( chose this one as its newer and crystalinfo said it had way less writes on it) 250GB SSD samsung 840 evo as unassigned. Its currently building parity it says it will take 7hrs to do. Everything is working from what i can tell but it has given me some errors First issue is one of the 3TB disks has errors. 509 of them in the main tab. S.M.A.R.T has highlighted this as the issue : 197 Current pending sector 0x0032 200 200 000 Old age Always Never 9 also the array says its used 21GB on each drive (62.9GB in total) but i'm yet to put anything on it. is that normal? Second issue is the ssd cache drive also has errors. 0 errors on main tab. S.M.A.R.T has highlighted this as the issue : 199 CRC error count 0x003e 099 099 000 Old age Always Never 1 it also says the cache is using 22.6GB but again i haven't transferred anything ( or set up shares yet ) is this normal? This weekend i'm just going to be messing around with it before committing all my data to it. Thanks for any help. Quote Link to comment
Hoopster Posted April 3, 2020 Share Posted April 3, 2020 1 hour ago, MMChris said: but i'm yet to put anything on it. is that normal? The filesystem takes some space and that is completely normal. You do have "data" on the drives, the filesystem. CRC errors (especially a high number of them) usually indicates a poor connection with the SATA cable or a bad cable. Pending sectors on a hard drive can be expected as the drive ages. If there are just a couple, not to worry, if the number is growing, that indicates your drive could be close to failing. The only acceptable error number in a parity check is 0. A high number of errors usually indicates a drive cabling problem causing read/write errors. Quote Link to comment
MMChris Posted April 3, 2020 Author Share Posted April 3, 2020 8 minutes ago, Hoopster said: The filesystem takes some space and that is completely normal. You do have "data" on the drives, the filesystem. CRC errors (especially a high number of them) usually indicates a poor connection with the SATA cable or a bad cable. Pending sectors on a hard drive can be expected as the drive ages. If there are just a couple, not to worry, if the number is growing, that indicates your drive could be close to failing. The only acceptable error number in a parity check is 0. A high number of errors usually indicates a drive cabling problem causing read/write errors. Thanks so in short 62gb on array is normal along with 21.6 on cache ? and keep an eye on that disk and maybe check the sata cables? Quote Link to comment
Hoopster Posted April 3, 2020 Share Posted April 3, 2020 57 minutes ago, MMChris said: Thanks so in short 62gb on array is normal along with 21.6 on cache ? and keep an eye on that disk and maybe check the sata cables? Yes 1 Quote Link to comment
JorgeB Posted April 4, 2020 Share Posted April 4, 2020 11 hours ago, Hoopster said: Pending sectors on a hard drive can be expected as the drive ages. If there are just a couple, not to worry, I think maybe @Hoopsterwas thinking of reallocated sectors? A single pending sector is a problem, unless it's a "false positive", i.e. there are no read errors or a failed SMART test. If you're having disk read errors then likely the disk has bad sectors, so any data there can't be read, parity sync will be invalid, if any other disk fails it can't be rebuild correctly, etc, so you should replace any such disk. Also a good idea to post the diagnostics (Tools -> Diagnostics) so we can have a better look, you can also run an extended SMART test on that disk to confirm if it's failing or not. 2 Quote Link to comment
Hoopster Posted April 4, 2020 Share Posted April 4, 2020 7 minutes ago, johnnie.black said: I think maybe @Hoopsterwas thinking of reallocated sectors? Oops, yes, you are correct. I was thinking of reallocated sectors. Thanks for catching that and the correction. Quote Link to comment
MMChris Posted April 4, 2020 Author Share Posted April 4, 2020 (edited) 1 hour ago, johnnie.black said: Also a good idea to post the diagnostics (Tools -> Diagnostics) so we can have a better look, you can also run an extended SMART test on that disk to confirm if it's failing or not. Thanks for catching it. the parity check finished last night. it came up with some notifications but i thought they would have been logged some where and closed them (still a newbie) soos-diagnostics-20200404-0924.zip Is this how i upload the Diag? Will run extended smart test on the disk now. Edited April 4, 2020 by MMChris Quote Link to comment
JorgeB Posted April 4, 2020 Share Posted April 4, 2020 43 minutes ago, MMChris said: Will run extended smart test on the disk now. That's good and it will confirm if the disk is really failing, and it does appear to be, in any case parity isn't 100% valid because of the read errors on that disk. Quote Link to comment
MMChris Posted April 4, 2020 Author Share Posted April 4, 2020 6 minutes ago, johnnie.black said: That's good and it will confirm if the disk is really failing, and it does appear to be, in any case parity isn't 100% valid because of the read errors on that disk. I'm doing a extended smart test on all the drives. will post diag when done. Is there a way to scheduled smart tests and how often should they be done? This weekend is basically me just testing Unraid, It seems very easy to use so i'm pretty much sold on it. i had planned on buying hardware and upgrading so i think i'll buy it now ready for the long weekend coming. so il buy 3 x 4TB wd red ( replace the dying one) i will also need a new case i'm thinking about https://www.servercase.co.uk/shop/server-cases/rackmount/4u-chassis/4u-server-case-w-24x-35-hot-swappable-satasas-drive-bays-6gbs-minisas-sc-4324/ I know its overkill for my needs but future proofing for further up grades. any other recommendations? will also change fans for noctuas and get reverse mini sas to sata cables. Quote Link to comment
JorgeB Posted April 4, 2020 Share Posted April 4, 2020 Is there a way to scheduled smart tests and how often should they be done? With Unraid there's no real need to schedule extended SMART tests (at least for array devices), since a parity check works the same, i.e., it reads all sectors of all disks, usually once a month is the recommended frequency for that. 1 Quote Link to comment
MMChris Posted April 4, 2020 Author Share Posted April 4, 2020 5 hours ago, johnnie.black said: With Unraid there's no real need to schedule extended SMART tests (at least for array devices), since a parity check works the same, i.e., it reads all sectors of all disks, usually once a month is the recommended frequency for that. Thanks here's the diag for all disks after extended smart test. soos-diagnostics-20200404-1841.zip also if my parity is already borked and disk 2 need changing do i stop the array to remove them? then re add the partity disk to the array as part of the array until i swap cases and disks? many thanks still a newbie. Quote Link to comment
MMChris Posted April 4, 2020 Author Share Posted April 4, 2020 Also just had a crash whilst using plex. had to reboot it. it didn't say why it crashed any ideas? soos-diagnostics-20200404-1951.zip Quote Link to comment
JorgeB Posted April 5, 2020 Share Posted April 5, 2020 SMART tests confirm disk2 is failing and needs to be replaced, since there's no data you can just rebuild to a new disk, or do a new config with a new disk2 and re-sync parity. 12 hours ago, MMChris said: Also just had a crash whilst using plex. had to reboot it. Syslog is cleared after a reboot, this might help if it happens again. 1 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.