September 21, 20205 yr Unraid Parity sync / Data rebuild: 13-09-2020 04:23 Notice [TOWER] - Parity sync / Data rebuild finished (1 errors) Duration: 10 hours, 5 minutes, 33 seconds. Average speed: 110.1 MB/s What do I do with a data rebuild that rebuilt with 1 error? Now when I run parity check I get a single error. How do I handle this?
September 21, 20205 yr Community Expert Go to Tools - Diagnostics and attach the complete Diagnostics ZIP file to your NEXT post in this thread.
September 23, 20205 yr Community Expert Unless you are trying to debug something related to mover I recommend disabling mover logging, it just clutters up syslog and isn't anonymized. I haven't reviewed SMART for each of your large number of disks. Do any disks have SMART warnings on the Dashboard? On 5 Sep you had problems with original disk11 which became disabled. Instead of rebuilding disk11, you did New Config / Trust Parity on 7 Sep, followed soon after by disk11 again disabled. Disk11 replaced on 12 Sep, SMART for still attached original disk11 looks OK. Looks like these may have just been connection issues. How was this disk attached? I didn't notice any I/O errors during rebuild but may have overlooked them. Is there an Error count on Main - Array Devices for any disk? Can you provide any other details?
September 23, 20205 yr Community Expert Rebuild was successful, error was from a corrected block on parity2: Sep 12 18:17:05 Tower kernel: md: recovery thread: Q corrected, sector=128 I would ignore for now, but keep an eye on it, any more sync errors (without a reason like from an unclean shutdown) might indicate a hardware issue.
September 23, 20205 yr Community Expert 6 hours ago, JorgeB said: error was from a corrected block on parity2 right after rebuild started, don't know how I missed.
September 23, 20205 yr Community Expert 1 hour ago, trurl said: don't know how I missed. What I do in this cases is to search the syslog for "thread", this will find all parity check/sync/rebuilds info, since they all start with "md recovery thread"
September 26, 20205 yr Author On 9/22/2020 at 7:14 PM, trurl said: Unless you are trying to debug something related to mover I recommend disabling mover logging, it just clutters up syslog and isn't anonymized. I haven't reviewed SMART for each of your large number of disks. Do any disks have SMART warnings on the Dashboard? On 5 Sep you had problems with original disk11 which became disabled. Instead of rebuilding disk11, you did New Config / Trust Parity on 7 Sep, followed soon after by disk11 again disabled. Disk11 replaced on 12 Sep, SMART for still attached original disk11 looks OK. Looks like these may have just been connection issues. How was this disk attached? I didn't notice any I/O errors during rebuild but may have overlooked them. Is there an Error count on Main - Array Devices for any disk? Can you provide any other details? Yeah this is what was weird the disk SMART was fine so I tried to rebuild the config since it had 4 read errors. But when I rebuilt it it got a lot more errors. The disk is still sitting there but it is not in the array anymore. I have a Norco 24 bay drive case with a sas back plane connected to dell perc cards in IT mode. I will shut down the server and check the cables. EDIT: Not bad cables, bad drive, Error 187. Unraid did no notify me of this, and the drive was green until I rebooted the server. Once I replaced that drive the array is functioning as normal. Edited October 24, 20205 yr by exist2resist
Archived
This topic is now archived and is closed to further replies.