CPU load is every time at 0%

[ITCommander]

Hi!

 

I have a fresh unRAID Installation with version 6.9.2. At my dashboard I can't see CPU load. Dashboard shows always 0%. If I start a putty session I can see the load. Is this a known topic or is there any solution for the problem?

 

BR
Jens

[outrigger999]

Are you using Nginx Proxy Manager? I've found that if I use Nginx Proxy Manager to access my servers through a subdomain which I've setup, that I can access it but the CPU loads look just like what your'e showing. I'm puzzled why this would be. I'm using Brave Browser but I've tried Safari as well and the same issues occurs. Also, another issue which presents is none of the popups work, even if I turn the blocking off for the site. So it's not Brave browser specifically because Safari exhibits the same thing. For example, if I install another docker app, the progress of the download and subsequent, "finished successfully" doesn't display and just a blank screen is there. Once the whole process is finished, then it comes up and displays.

 

If I access the servers directly on my local network, then everything works the correct way on both browsers. It seems as if I'm not routing all the protocols which are needed but I'm not sure what's going on. Anyone have a suggestion? I use PFSense as my router and for port 80 and 443 I route TCP/UDP. Any help would be appreciated.

[ITCommander]

Thank you for your informations. But i don't use a proxy. I use browser in my internal network to connect the unraid server. Only thing what I have that the internal VNC is not working for me. Each VM I can't connect to VM. I have to use external VNC client from my PC.

[jonp]

Please disable any ad blockers or other services running in your browser that may interfere with the webGui.  Let us know if after doing so the issue is resolved.  This definitely feels like a browser-specific issue.  Can you recreate the issue when using a different browser or device?

[WowaDriver]

For anyone who cares, I think I fixed this! I think it may be due to nginx not working with Websockets by default.

 

I added these lines to the site config:

 

proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "Upgrade";

 

And it worked.

 

E.g., this is the location settings for my unRAID proxy:

 

	location / {
		auth_basic "Restricted";
		auth_basic_user_file /config/nginx/.htpasswd;
		include /config/nginx/proxy.conf;
		proxy_set_header	Host $host;
		proxy_set_header	X-Real-IP $remote_addr;
		proxy_set_header	X-Forwarded-For $proxy_add_x_forwarded_for;
		proxy_set_header	X-Forwarded-Proto $scheme;
		proxy_pass http://192.168.1.106:8080;	
    	proxy_set_header Upgrade $http_upgrade;
    	proxy_set_header Connection "Upgrade";
	}

 

And now the CPU stats update in real time!

 

Fingers crossed this is the fix.

[Goldjacket]

I have the CPU 0% issue as well. I see the above code to fix it. But I don't want to attempt that myself. It worked before. But I don;t remember it working from a remote location. 

[mgutt]

It seems I'm having the same problem on my test server with Unraid 6.12.4

 

 

Error message of Chrome Dev Tools:

Quote

Unchecked runtime.lastError: A listener indicated an asynchronous response by returning true, but the message channel closed before a response was received

 

This server is relatively empty, so it does not have any Docker installed at all. But I think it could be because of my in my router (fritz.box). I had a similar issue in the past which got solved after disconnecting the server and deleting all network devices with the server name from my routers webgui network devices overview. At the moment some downloads are running. I will check this later and update my post.

 

EDIT: And as I thought it was because of my fritz box.

This router does crazy things if a mac address changes for a device, which has active port redirects. A similar issue causes the problems with MACVLAN and docker containers.

 

 

My environment:

- thoth, my productive server with the fixed ip .8 and active port redirects for the ports 80 and 443

- tower, my test server with an dynamic DHCP address, at the moment .48

 

A few days ago I changed the network card of "thoth". So it got a new mac address. This causes this mess:

 

As you can see the productive server with the correct name is listed as an "offline device", but with active port redirects ("Portfreigabe").

 

In addition I got two "Tower" servers, which absolutelly does not make sense as my productive server never had this name. It is unchanged "thoth" since several years.

 

How to solve this mess:

- shutdown both servers

- remove the port redirects of "thoth"

- boot "thoth"

- add the port redirects again

 

Now I have my two network devices as expected:

 

 

And my CPU dashboard works, too:

 

But why does it solve the above problem?

I think the reason is, that the Fritz Box randomly answers DNS requests for "tower.fritz.box" with the IP .8 or .48 as it knows about two different network devices with the same name.

 

This bug (?) is present as long I can remind. But even if AVM solves this I'm still confused why my productive server sometimes is detected as "Tower". Does the server boot with this name and changes it after reading the config files from the USB drive?