Only started happening since recent up date to 7.0.1

 

Slow web management when docker cpu usage load high causes PHP crash?

 

I did up my zfs ram allowance but there's still about 11GB free so it shouldn't be a ram issue. 

 

All containers still accessible and if I pause NZBGET then I get main unraid webGUI back.. so is it some php bug?

 

moulin-rouge-diagnostics-20250302-2303.zip

 

Edited March 2, 20251 yr by dopeytree

Ping is fine.

Sounds like nzget is being a resource hog and starving the system, try restricting cpu through pinning resources (avoid first 2 cores if possible).

example:


note: advanced view must be toggled to see this option in docker template

Thanks yeah that's helping stop it overtaking. Noticing that the file monitor plugin seems to be eating some cpu up when nzbget is going which is odd?

I've disabled it for now.

Just now had this again.

 

Something causes the PHP bug but only for main Unraid GUI, then gpu says it hang but frigate still working.

Am I being hacked or something?

 

Noticed if I reboot firewall regain access to unraid gui.
 

moulin-rouge-diagnostics-20250307-1629.zip

In my experience, these errors can be the result of the server being close to exhausting the memory, GUI can become extremely slow, like 1 minute to open the dashboard, try limiting the memory for VMs/docker services, or adding a little more RAM.

 

It could also be one or more containers hogging the CPU, try pinning only some cores to them, and leave cores 0/1 available for Unraid.

 

Also, recommend trying a couple of other things, go to Settings - Global share settings and set the Number of fuse File Descriptors to the max, and enable this:

https://docs.unraid.net/unraid-os/release-notes/7.0.0/#excessive-flash-drive-activity-slows-the-system-down

Not highly likely, but wouldn't be a bad idea as a test to disable XMP in the BIOS to run the memory at it's rated speed and also have the system run a memtest for at least a pass

Cheers will keep trying things.
Can always roll back to 7.0

 

Server has 11 - 8GB ram free

 

CPU load average 50%.

Might try putting some more containers onto the hyper threading cores. Most small things are limited to the efficiency cores.

Mar  7 17:24:08 Moulin-rouge nginx: 2025/03/07 17:24:08 [error] 763884#763884: *1744649 connect() to unix:/var/run/syslog.sock failed (111: Connection refused) while connecting to upstream, client: 192.168.22.155, server: , request: "GET /webterminal/syslog/ HTTP/2.0", upstream: "http://unix:/var/run/syslog.sock:/", host: "192.168.22.2", referrer: "https://192.168.22.2/Main"

 

Edited March 7, 20251 yr by dopeytree

Does seem to do doing small reads of USB stick.

How do I downgrade?

 

The downgrade button doesn't seem to work says a downgrade is not available.




There's definitely some docker log bug in 7.0.1 as that is what is also causing the errors with the disk activity plugin

On 3/7/2025 at 6:50 PM, dopeytree said:

Server has 11 - 8GB ram free

 

From my experience on several RAM amounts stability is more percentage-based than absolute quantity, I've only had issues like the one you have when RAM usage was >80% or so, reaching 85-ish% would usually eventually lead to a crash.

 

My guess is if you undo your ZFS ARC increase to stay below 80% use it'll likely be stable again.

Edited March 10, 20251 yr by Kilrah

Thanks ok now set at 40GB instead of 48GB out of 64GB.

Currently seems the ZFS cache is slower to fill. It's opting to use 19GB for cache.

Perhaps before although it says free ram. Perhaps it is not available without using a swap file or similar.

Edited March 10, 20251 yr by dopeytree