Currently, I have two disks that are in a disabled state. A few months ago, I had a drive (currently working fine) drop into a disabled state but rebuilding the array solved the problem. Now, I have one drive (sdd) that is disabled and was unable to rebuild the array with it as it would return to a disabled state. Currently, it also appears to randomly not show up in unraid after restarts.

 

The second disabled drive (sdi/disk3) went to a disabled state a few days ago and has been unable to rebuild. Usually, when I try to stop the array, it hangs forever on "unmounting disk shares"

 

To add to the issues, today when rebooting after booting into safe mode the docker service failed to start. After rebooting a few more times testing SATA cabling to problem drives, docker containers are no longer present. Below diagnostic is from today, when all hard drives were showing attached and docker service wasn't starting.

 

I'm thinking this could be a HW issue either with my HBA or other old parts. Would love input.

 

 

Specs:

Unraid 6.12.4

i5-7600K

16GB RAM

Asus Z170-k

HBA: LSI SAS 2008

x5 8TB HDD IronWolf

x1 10TB Parity Drive

x1 12TB Parity Drive

x1 480GB SSD Cache

 

 

wt-nas-diagnostics-20241124-1034.zip

Edited November 24, 20241 yr by CorCain
clarity

turn off any and all autostarts n vms and doekrs.

Settings > docker off
setting >vm off

stop the array
start the array in maintenance mode smart test and check file sytems on all disks.

for parity rebuild you may need to do a new config keping parts of the setup but replacing the disabled disk...

https://docs.unraid.net/unraid-os/manual/storage-management/#replacing-faileddisabled-disks

23 hours ago, bmartino1 said:

turn off any and all autostarts n vms and doekrs.

Settings > docker off
setting >vm off

stop the array
start the array in maintenance mode smart test and check file sytems on all disks.

for parity rebuild you may need to do a new config keping parts of the setup but replacing the disabled disk...

https://docs.unraid.net/unraid-os/manual/storage-management/#replacing-faileddisabled-disks

 

Performed smart tests on all drives, did a Read Check found zero errors. Had some further docker issues prior to performing the above steps, Plex wasn't responding and attempting a restart caused a "Server Error" pop up. I have two drives on the way to test a replacement, would this lack of parity cause this docker instability? Also curious if anyone notices any issues with my smart tests, I'm not certain what to look for.

 

wt-nas-diagnostics-20241125-1718.zip

Having ditched the array due to mover and parity issues. and running multiple disk without parity. my answer is no parity only protects the disk by simulating a mirror aid array between the disk. Reviewing diagnose will see what i find. plex resource and server eror may be due to a failing disk.

 

in meant time you can review:

 

 

some smart stuff:

 

SMART Data Analysis

 

Disk2 (ST8000VN004-2M2101, Serial WSD8YNN6):

Raw_Read_Error_Rate: High value (7874576), but normalized value (69) is above threshold.

Seek_Error_Rate: High raw value (194279759), normalized value (83), within limits.

Overall Health: Passed(ST8000VN004-2M2101_WSD8…).

 

Parity Disk (ST10000NE0008-2JM101, Serial ZPW0M5M3):

Raw_Read_Error_Rate: Elevated raw value (90727384), normalized value (80), within acceptable limits.

Seek_Error_Rate: High raw value (600690657), normalized value (88), still within thresholds.

Overall Health: Passed(ST10000NE0008-2JM101_ZP…).

 

Parity2 (WDC_WD120EDAZ-11F3RA0, Serial 5PH4BELE):

All SMART attributes indicate normal functioning with no reallocated sectors.

Power_On_Hours: 19447 hours, indicating extended use(WDC_WD120EDAZ-11F3RA0_5…).

 

Cache Disk (KINGSTON_SA400S37480G, Serial 50026B7784DB8DBD):

No significant issues observed.

Bad_Block_Rate: 10, not concerning for an SSD of this age(KINGSTON_SA400S37480G_5…).

 

Disk4 (ST8000VN004-2M2101, Serial WSD8VT8H):

Raw_Read_Error_Rate: High raw value (192334096), normalized value (83), within threshold.

Seek_Error_Rate: High raw value (244783451), within acceptable limits.

Overall Health: Passed(ST8000VN004-2M2101_WSD8…).

 

Disk1 (ST8000VN004-2M2101, Serial WSD8VT40):

Similar observations to Disk4 with high raw values for error rates but normalized values are fine.

Overall Health: Passed(ST8000VN004-2M2101_WSD8…).

 

Disk (ST8000VN004-2M2101, Serial WSD8YGAB):

SMART Data Unavailable due to device not being ready. Indicates a potential failure(ST8000VN004-2M2101_WSD8…).

 

Syslog Analysis

Notable Errors:

Kernel logs show no immediate disk errors but no entries confirm smooth functioning for the problematic disk (WSD8YGAB).

Errors may indicate unresponsiveness or power issues with the disk(syslog.1).

Docker Instability:

Docker application crashes (e.g., Plex) might be related to intermittent failures in accessing Disk (WSD8YGAB).

Yeah disk failures.. bakup and replace disk

 

Recommendations

Replace Disk (WSD8YGAB):

Immediate replacement is recommended as SMART data is unavailable, indicating severe issues.

Monitor Parity Disks:

Keep an eye on Raw_Read_Error_Rate and Seek_Error_Rate for parity disks, but no immediate action is necessary.

Docker Instability:

Verify if Plex or other Docker containers are accessing the problematic disk. Consider reconfiguring or isolating workloads.

General Maintenance:

Schedule regular SMART checks and consider extended tests for all drives.

1 hour ago, bmartino1 said:

some smart stuff:

 

SMART Data Analysis

 

Disk2 (ST8000VN004-2M2101, Serial WSD8YNN6):

Raw_Read_Error_Rate: High value (7874576), but normalized value (69) is above threshold.

Seek_Error_Rate: High raw value (194279759), normalized value (83), within limits.

Overall Health: Passed(ST8000VN004-2M2101_WSD8…).

 

Parity Disk (ST10000NE0008-2JM101, Serial ZPW0M5M3):

Raw_Read_Error_Rate: Elevated raw value (90727384), normalized value (80), within acceptable limits.

Seek_Error_Rate: High raw value (600690657), normalized value (88), still within thresholds.

Overall Health: Passed(ST10000NE0008-2JM101_ZP…).

 

Parity2 (WDC_WD120EDAZ-11F3RA0, Serial 5PH4BELE):

All SMART attributes indicate normal functioning with no reallocated sectors.

Power_On_Hours: 19447 hours, indicating extended use(WDC_WD120EDAZ-11F3RA0_5…).

 

Cache Disk (KINGSTON_SA400S37480G, Serial 50026B7784DB8DBD):

No significant issues observed.

Bad_Block_Rate: 10, not concerning for an SSD of this age(KINGSTON_SA400S37480G_5…).

 

Disk4 (ST8000VN004-2M2101, Serial WSD8VT8H):

Raw_Read_Error_Rate: High raw value (192334096), normalized value (83), within threshold.

Seek_Error_Rate: High raw value (244783451), within acceptable limits.

Overall Health: Passed(ST8000VN004-2M2101_WSD8…).

 

Disk1 (ST8000VN004-2M2101, Serial WSD8VT40):

Similar observations to Disk4 with high raw values for error rates but normalized values are fine.

Overall Health: Passed(ST8000VN004-2M2101_WSD8…).

 

Disk (ST8000VN004-2M2101, Serial WSD8YGAB):

SMART Data Unavailable due to device not being ready. Indicates a potential failure(ST8000VN004-2M2101_WSD8…).

 

Syslog Analysis

Notable Errors:

Kernel logs show no immediate disk errors but no entries confirm smooth functioning for the problematic disk (WSD8YGAB).

Errors may indicate unresponsiveness or power issues with the disk(syslog.1).

Docker Instability:

Docker application crashes (e.g., Plex) might be related to intermittent failures in accessing Disk (WSD8YGAB).

Yeah disk failures.. bakup and replace disk

 

Recommendations

Replace Disk (WSD8YGAB):

Immediate replacement is recommended as SMART data is unavailable, indicating severe issues.

Monitor Parity Disks:

Keep an eye on Raw_Read_Error_Rate and Seek_Error_Rate for parity disks, but no immediate action is necessary.

Docker Instability:

Verify if Plex or other Docker containers are accessing the problematic disk. Consider reconfiguring or isolating workloads.

General Maintenance:

Schedule regular SMART checks and consider extended tests for all drives.

 

 

Thank you very much for the detailed responses. Replacement Drive is arriving mid next week and will try replacing the drive, will report back.

 

Will investigate Docker containers further given your recommendations will also look into scheduling smart checks.

Replacement disks arrived, similar issues.

 

I shutdown unraid, swapped the disks and booted the PC.

Array was already started, stopped it and added the disks, showed as new devices

Disks almost immediately showed as disabled after the array was started

 

Was able to run smart tests on WWZ6WENA but not WWZ6ZZXE, the short and long just stop instantly.

 

Starting to think there's something further down the line causing issues. Either bad cabling, bad HBA or possibly other hardware components. Thinking it might be the HBA. New Diagnostics attached.

wt-nas-diagnostics-20241204-1738.zip

Edited December 5, 20241 yr by CorCain

this soulds like the hba. Check cabling in secure. HBA usual require 8 lanes make sure it is plugined into the coercet PCI slot ont eh board.

also some HBA are cord order specific meaning slot 0 port 1 -4 and slot 1 port 1-4 need to be plugged in order

this means slot 0 p1 needs pluged in and all of slot 0 connected before slot 1 port 1 is used...
Example: if you have 4 disk and you plug slot 0 port 1 and 2, then slot 1 port 1 and 2 you may encounter issues on some HBA...

Also make sure power is good, and avoid splitters.

I ended up using this as an excuse to order new hardware, did a complete rebuild aside from the drives, all new cabling, HBA, mobo, cpu, etc. This of course solved the problem, have a data rebuild currently underway.

 

Both above posts are good suggestions, in the previous setup I had slot 0 and slot 1 filled as I have 7 drives and 1 cache.

I did have a molex --> sata power cable in play but am using it currently and it's functioning.

 

It was likely the previous HBA giving me issues and a swap would've likely fixed the problem, otherwise a rebuilt would've been warranted.