Getting PHP errors in docker settings page after downgrade from 6.12.x to 6.11.5

[VicePrez]

Hello!

 

Hope someone can help me through this issue I've been having. Can't seem to find the root cause after searching online and through my logs.

 

Background:

 

I recently downgraded from 6.12.x to 6.11.5 due to stability issues, I've been stable on 6.11.5 for quite some time and the recent upgrade to 6.12.x had me crashing at least once a week or so.

 

What I noticed after the downgrade:

 

I followed the official docs on downgrading to an earlier version, and noticed that my docker containers weren't running after a clean reboot. Took a quick peak at the docker settings tab and noticed a number of PHP errors (see below).

 

What I've done thus far:

 

• Manual force update for each docker container - this fixed the issue with containers that weren't running
• Several system reboots (php error still persists)
• Deleted docker vdisk image, and reinstalled docker containers from templates (php error still persists)

 

Warning: Invalid argument supplied for foreach() in /usr/local/emhttp/plugins/dynamix/include/DefaultPageLayout.php(694) : eval()'d code on line 57

Warning: Invalid argument supplied for foreach() in /usr/local/emhttp/plugins/dynamix/include/DefaultPageLayout.php(694) : eval()'d code on line 57

[itimpi]

Have you tried booting in Safe Mode in case you have a plugin installed that is not compatible with 6.11.5?

[VicePrez]

10 hours ago, itimpi said:

Have you tried booting in Safe Mode in case you have a plugin installed that is not compatible with 6.11.5?

Not yet; will try that right after work.

[VicePrez]

Same errors after rebooting in safe mode (image: Safe mode - array not started), I did however notice that the last set of php errors are not showing up when the array hasn't started. Once started, its exactly like first picture i posted (image: Non-safe mode). Any other ideas?

 

Non-safe mode:

 

 

Safe mode - array not started:

 

[VicePrez]

Bump

[Squid]

You'll need to post your diagnostics

[VicePrez]

48 minutes ago, Squid said:

You'll need to post your diagnostics

Please see attached. Thanks @Squid

titan-diagnostics-20230810-1757.zip

[Squid]

Unfortunately that error is hard to determine on 6.11 what its from (far easier on 6.12 to figure that out)

 

 

but, it does appear that you *might* have memory issues (run Memtest from the boot menu) and the appdata plugin seems to run about twice a day for an hour or so and it does try and stop everything from running.

[Mainfrezzer]

Ive had a similar issue after downgrading from the GUI just a bit different. After i logged in, nothing except php code would appear. Restoring a backup with the Unraid Flash tool fixed the issue for me.

Saving the config, reflashing the stick and placing the config back on might be worth a shot.

[VicePrez]

thank you both, ill try a memtest run tonight and see where that goes. if that doesnt go anywhere, i'll the other suggestion by @Mainfrezzer

[VicePrez]

Quick update on the memtest run; no issues with 2 passes.

 

Going to try reflashing the USB stick. Am I basically using the USB Creator tool to install 6.11.5 and copying the config file over? Would that nullify the registration key?

[itimpi]

You may find it easier to do a Manual downgrade.

[VicePrez]

27 minutes ago, itimpi said:

You may find it easier to do a Manual downgrade.

Yep, tried that just now. Still the same issue.

Wondering if the php file that's spitting out those errors is a 6.12.x version (i.e. code thats not backwards compatible with 6.11.5)

 

For reference:
 

/usr/local/emhttp/plugins/dynamix/include/DefaultPageLayout.php

 

[itimpi]

30 minutes ago, VicePrez said:

Yep, tried that just now. Still the same issue.

Wondering if the php file that's spitting out those errors is a 6.12.x version (i.e. code thats not backwards compatible with 6.11.5)

 

For reference:
 

/usr/local/emhttp/plugins/dynamix/include/DefaultPageLayout.php

 

that would not be the culprit as it comes from the bz* type files that are on the flash.     The only thing I could think of that could be 6.12.x specific would be code installed by a plugin but you say you get the same problem when booting in Safe Mode which avoids loading plugins.

[VicePrez]

3 hours ago, itimpi said:

that would not be the culprit as it comes from the bz* type files that are on the flash.     The only thing I could think of that could be 6.12.x specific would be code installed by a plugin but you say you get the same problem when booting in Safe Mode which avoids loading plugins.

Correct - was getting the same error even in safe mode. Was looking through some other threads, some were related to network changes. The only thing i can think of is changing ports on my switch from one sfp port to another. Coincidently it was also around the same time i downgraded, so not sure which event caused what.

EDIT:

This thread specifically:
 

 

Edited August 11, 2023 by VicePrez
Adding link

[Lolight]

9 hours ago, VicePrez said:

Quick update on the memtest run; no issues with 2 passes.

It's usually recommended to complete at least 4 passes.

It's not common but sometimes a problematic stick configuration would run fine for a couple of passes but then fail on the 3rd or even 4th pass.

[VicePrez]

19 hours ago, Lolight said:

It's usually recommended to complete at least 4 passes.

It's not common but sometimes a problematic stick configuration would run fine for a couple of passes but then fail on the 3rd or even 4th pass.

 

Yea absolutely, I've seen recommendations anywhere between 2 and 8 passes. I am less inclined to think its memory related; the system has been stable for over 2 years and decided to go kaput after the 6.12.x upgrade. I can do another memtest run next weekend, time permitting. I'm running out of ideas as to what the cause is..

[VicePrez]

Just closing this thread, i ended up reverting back to my original 6.12.3 backup and the error disappeared. I'll just ride it out and see why i was crashing in 6.12.x to begin with. Thank you everyone