[6.12.8] No WebGUI - /run full

47 minutes ago, ich777 said:

Can someone with a 64+ core system the output from:

grep -c ^processor /proc/cpuinfo

 

 

 

model name      : AMD EPYC 7V12 64-Core Processor

 

SMT disabled:
 

root@Tower:~# grep -c ^processor /proc/cpuinfo
64

 

SMT enabled:
 

root@Tower:~# grep -c ^processor /proc/cpuinfo
128

 

Edited February 19 by bubbl3

Thanks @bubbl3!

Sorry can't share the diagnostic @JorgeB (there's Engineering Sample equipment on my unraid, i can't do that due to NDAs)

 

model name      : AMD EPYC 7B13 64-Core Processor

root@Megumin:~# grep -c ^processor /proc/cpuinfo
128
 

The big difference from my look at both of them is the multiple GPUs vs my 1 (which is a converged accelerator, which is the ES part)

 

root@Megumin:~# du -sh /run/udev/* | sort -h | tail -3
0       /run/udev/tags
0       /run/udev/watch
17M     /run/udev/data
(this is my udev size)

 

I know there was changes on the kernel on udev for multiple gpus due to issues at booting them "fast" enough on AI clusters. (just don't remember which 6.1 version got it, but it's one between 6.12.6 and 6.12.8, still looking at patches to see where it was applied, as we upgraded to 6.6 at work to have the patch before it was on 6.1) So, one thing i should think is making the udev 64MB for 6.13 (which would help for weird and complex systems.)

 

I know Rome vs Milan also do udev differently, as it was an updated architecture on the cpu side.

 

Mobo: Asrock Rack RomeD8-2T
CPU: Epyc 7B13
Ram: 1TB
PCIe Expensions:
1 x LSI 9400-8i (x8)
3 x ASUS Hyper M.2 card filled with m.2
1 x Nvidia A100X ES
1 x Connectx-5 pro 
2 x Oculink to U.2
2 x M.2 (On-board)

 

Those are all my specs as i posted yesterday on Discord to confirm where the issue could be.

3 hours ago, ich777 said:

Can someone with a 64+ core system the output from:

grep -c ^processor /proc/cpuinfo

 

Model name: AMD Ryzen Threadripper PRO 5995WX 64-Cores

root@unraid:~# grep -c ^processor /proc/cpuinfo
128

 

*edited

 

 

Edited February 19 by Amane

5 minutes ago, Amane said:

root@unraid:~# grep -c .processor /proc/cpuinfo

You've run different commands, I need the output from that:

grep -c ^processor /proc/cpuinfo

 

2 minutes ago, ich777 said:

You've run different commands, I need the output from that:

grep -c ^processor /proc/cpuinfo

 

 

He run the one you need too:

13 minutes ago, bubbl3 said:

He run the one you need too:

Oh sorry, completely overlooked that because of the quotes...

 

Thanks for pointing that out.

Here are my observations from yesterday to now. Almost no changes in /run:

Spoiler

 

Edited February 19 by Amane

Believe the SOLVED label should be removed from this report :D

2 hours ago, bubbl3 said:

Believe the SOLVED label should be removed from this report

Yeah, not really solved, I guess the OP marked it solved.

Changed Status to Open

I have the same Problem. Gui very very slow

12 hours ago, Manyak said:

Gui very very slow

That doesn't look like the same issue being reported here, but please post the diagnostics. 

Can confirm this issue with my 64 core epyc processor.  I had to revert back because disabling SMT made everything much slower, including the webgui.

4 hours ago, Einsteinjr said:

I had to revert back because disabling SMT made everything much slower,

You can do this:

 

https://forums.unraid.net/bug-reports/stable-releases/6128-no-webgui-run-full-r2854/?do=findComment&comment=27334

 

These sorts of issues really shouldn't be in 'stable' releases. This seems to hardly be a 'stable' bug.

Either that or unraid needs another release train like long term, or the release terms need to be re-jigged.

 

e.g.

Development (Alpha/Beta) i.e. 7.xxxx or 6.13.xxxxx

Release-Candidate - installs any newer version release candidates (aka bleeding edge) i.e. 6.12.9-RC1

Latest (newest official release) - installs only latest official releases (basically today's stable) i.e. 6.12.8

Stable (Long-term, known stable releases only) - only installs known very stable, safe releases. i.e. 6.11.5.

10 hours ago, warpspeed said:

These sorts of issues really shouldn't be in 'stable' releases. This seems to hardly be a 'stable' bug.

Problem is that they can be very difficult to detect, in this case it only affects users with 128 CPU threads, so unless any of the beta testers was using one, it would be undetectable, and beta testers cannot possibly run all the different types of hardware all users have.

 

 

Can confirm I'm seeing the same with a AMD Ryzen Threadripper 7980X 64-Core. the /run size change works, just a very frustrating 2 or so hours to figure it out.

 

 

Edit: I'm also having to do killall -HUP dhcpcd to get the networking errors to resolve after updating to 6.12.8 from 6.12.4, no ideas on why that is though.

Edited February 29 by Nikkeh

Is the assumption that this will be fixed in a new release?

It should be.

Damn i love you guys. Was in a weird situation when i upgraded to my Epyc 9554 and my server ws not reachable. Debugged the whole day (not an expert) until i nearly gave up and found this thread. Reverted back to 6.12.6 - since the workaround would not work for me, same issue.

 

Server is now up and running i really considered some issue with my hardware. 

 

Again, thx for finding the root cause ! I will warch rhis Thread and wont update until then.

 

11 hours ago, BillVR said:

since the workaround would not work for me

That is strange, any chance you saved the diagnostics?

Unfortunately not, but i can boot feom 6.12.8 and recreate the issue and post it. I wont be able before saturday tho

But for the cause , im glad i can help

Hi everyone, 

 

First time poster here, but I seem to have a similar type of problem with my Unraid 6.12.8 wherein I am not able to access the GUI. Although I do not have a 64/128 core system. 

I am able to access my dockers and my SMB folders using Tailscale remotely, and I am also able to access my server via SSH. However, the GUI is just blank and unaccessible. 

Also my connect.unraid shows my system as being online and accissable. 

 

Unfortunately I am only able to access the server remotely and have no knowledge about SSH. I am able to access the SSH via the Tailscale services (great application btw. for remote accessing your network). 

 

I have an AMD Radeon 5700G,

6 different hard drives

1 cache drive,

 

MSI MAG X570 Tomahawk WiFi Motherboard

 

I ran diagnostics, but as I mentioned not really familiar with SSH, I do not know how to download the file from the server. 

 

Thanks in advance, 

Edited March 16 by HTJP1984

3 hours ago, HTJP1984 said:

Hi everyone, 

 

First time poster here, but I seem to have a similar type of problem with my Unraid 6.12.8 wherein I am not able to access the GUI. Although I do not have a 64/128 core system. 

I am able to access my dockers and my SMB folders using Tailscale remotely, and I am also able to access my server via SSH. However, the GUI is just blank and unaccessible. 

Also my connect.unraid shows my system as being online and accissable. 

 

Unfortunately I am only able to access the server remotely and have no knowledge about SSH. I am able to access the SSH via the Tailscale services (great application btw. for remote accessing your network). 

 

I have an AMD Radeon 5700G,

6 different hard drives

1 cache drive,

 

MSI MAG X570 Tomahawk WiFi Motherboard

 

I ran diagnostics, but as I mentioned not really familiar with SSH, I do not know how to download the file from the server. 

 

Thanks in advance, 

 

Hi HTJP1984

 

I'm currently thinking about what is the easiest way to get the diagnostics.
Basically they are on the USB (/boot/logs).

I think you should install FileZilla or WinSCP. Then connect to the server, go to the specified path and download the latest file.

You can also copy the diagnostics to your SMB share via SSH and then pull them down there:

cp /boot/logs/*diagnostics* /mnt/user/your_share/

 

I hope it works like that..

Edited March 16 by Amane