Shares disappear when docker is running

[halt95]

Hello, after changing my CPU, and adding a new drive, i have found that im having problems with docker containers making all shares go missing/disappear. in krusader the user folder empties. a restart fixes this for a short period of 10 minutes until shares go missing again.

 

i'm very unsure as to what would be causing this as the new disk is still being pre cleared.

 

Thanks

Halt95.

arthur-diagnostics-20231102-0109.zip

Edited November 2, 2023 by halt95

[JorgeB]

Where the diags downloaded after the problem? I'm not seeing anything out of the ordinary logged.

[halt95]

I believe so, here’s diagnostics from just now with server being on with the problem occurring overnight.

arthur-diagnostics-20231102-1046.zip

[JorgeB]

Post the output of

ls -la /mnt/user

 

[halt95]

 

I did see the thread to this potential permission problem but it looks like mine is fine?

[halt95]

My Disk 5 is showing: I have unplugged and replugged the data and power cables and restarted unraid.  same error still showing

[JorgeB]

I don't see anything wrong, post a screenshot of the shares page.

[itimpi]

CrC errors are normally related to the power or SATA cabling. CRC errors never get reset so all you can do is stop increasing.  You should acknowledge it on the Dashboard by clicking the orange icon and then you only get told again if it increases.

[halt95]

All the shares are back with an uptime of 1hr.20 mins. Changes made since restart: Updated to 6.12.4 - reseated power/sata cable for disk 5. Hard power cycled.

 

I'm now able to access and utilize all docker containers as normal without the shares disappearing?

Edited November 2, 2023 by halt95

[JorgeB]

1 hour ago, halt95 said:

I'm now able to access and utilize all docker containers as normal without the shares disappearing?

Are you asking us?

 

If now is working we can't see what the problem was, if it happens again grab new diags before rebooting.

[halt95]

It's more than i'm confused as to what keeps making it disappear and reappear. shares have disappeared again.

 

 

https://i.imgur.com/zTmZnBs.png

arthur-diagnostics-20231102-1431.zip

[JorgeB]

I still don't see anything in the logs, they disappear from where? If you go to Shares page they are not there?

[halt95]

 

shares missing and containers become unresponsive

[JorgeB]

I'm sorry but I don't see anything in the logs that would explain that, maybe someone else can.

[halt95]

No problem thanks for looking! Looking deeper into this when i try to open sonarr/radarr i get this error while shares are missing:

 

Could it be something to do with the open file limit being hit and shares drop out?

[halt95]

Tried this script but didn't work for me. still getting open file limit exceeded on all containers.

 

I have a replacement samsung USB drive,motherboard & CPU arriving on saturday.

The issue of missing shares and open file limit on docker containers started occuring after i changed CPU & added a new data drive (still needs to be precleared.)

[halt95]

Woke up today after leaving a parity check to do overnight with disk 1 in error state - disabled and content emulated.

 

rebooted the server around 14 hours ago. disk 1 disabled around 9 hours ago. logs say there's nothing wrong with disk?

 

I have only been running 4-5 containers such as z2m/homebridge/codeproject ai/mosquitto & Plex - these do not make the shares disappear.

I am able to run my VM with no problem - does not interact with the array or cache pool.

 

 

With disk 1 disabled, now trying to see if shares drop with all containers running as needed.

arthur-diagnostics-20231104-1034.zip

[itimpi]

1 hour ago, halt95 said:

logs say there's nothing wrong with disk?

What makes you say this - there are lots of errors relating to disk1 in the syslog.  They started off as read errors but eventually you got write errors as well which would be why the drive was disabled.

 

could just be a power or SATA cabling issue so worth checking these out.

 

 

[JorgeB]

It does look more like a power/connection issue.

[halt95]

23 minutes ago, itimpi said:

What makes you say this - there are lots of errors relating to disk1 in the syslog.  They started off as read errors but eventually you got write errors as well which would be why the drive was disabled.

 

could just be a power or SATA cabling issue so worth checking these out.

 

 

i see the syslog disk errors but i dont see any SMART changes?

 

 

All shares disappeared again with all containers running

Edited November 4, 2023 by halt95

[itimpi]

20 minutes ago, halt95 said:

i see the syslog disk errors but i dont see any SMART changes?

That is one reason why i suggested that you check the cabling.

 

it is quite possible for a disk to fail with no obvious SMART indication.   You can try the extended SMART test as if that fails the drive needs replacing.

[halt95]

ah ok understood now. Reseated cabling and HBA's:

disk1 is back to normal and working in the array. will run a extended test on it. i have restarted the containers mentioned before.

 

[halt95]

Replacement parts have arrived. First trying a samsung 32GB Bar Plus - manually installed UNRAID and copied over only my config file and logs folder. Server is back up and running, CPU utilization has dropped from around 80% to 20%. Shares still here.

 

upon plugging in my old usb drive, windows asked me to scan & repair the usb drive. Wondering if my old one is could of been the cause of shares going missing. Will leave all containers open and vms as normal overnight to see if shares disappear again before upgrading my CPU or swapping mobo.

 

Would CPU being pinned to 100% cause shares to temporarily shutdown?

 

 

EDIT: 20:12 shares disappeared again... frustrating because i don't know what the actual problem is or lies. will be upgrading cpu and replacing mobo now to see if it's hardware related.

 

 

Edited November 4, 2023 by halt95
update

[halt95]

Replaced Motherboard & CPU last night issue still occuring. Saw on the forums that someone had a similiar issue and it was their HBA's starting to die. Have ordered a new HBA but won't come for a few weeks. Tempted to try wiping all appdata and starting fresh with important configs copied over.

 

shares are still missing.

arthur-diagnostics-20231105-1256.zip

[halt95]

Ok so i have ran a parity error correcting check and disk 1 went disabled again part way through again, this was after the cables were replaced. A ssd was showing a few errors but i think i saw that its normal for the CT500MX.

 

hopefully there's something in these logs to indicate where the problem could be.

arthur-diagnostics-20231107-1146.zip

 

Edited November 7, 2023 by halt95