Sudden maxed out CPU and Memory, now getting 504 Bad Gateway

[MrPatton75]

I have attached my diagnostics.  I had to hard power it off and then on again.  Then after everything came back up I did a fresh reboot.  Waited for everything to come back up (Docker Containers).  I'm unable to access any of my web gui's by FQDN.  I can get to them using IP:PORT though.  I can get to Nginx Proxy Manager by IP:PORT and FQDN (that uses itself to proxy).  I'm not sure what is going on.  Everything was working great, watching a movie via Jellyfin and suddenly it froze on my TV.  Checked the server and noticed my CPU and Memory were completely maxed out.  Took forever to just load the dashboard to see that.

I looked through the logs, but honestly don't what I'm looking for.  Didn't notice anything, but that doesn't mean anything, all looks greek to me anyway.

Any help with them would be greatly appreciated.

nas-diagnostics-20240206-2023.zip

[astronax]

this sounds very likely to be a dns issue?

 

what domain are you using, a valid internet domain or something locally?

 

are you using a custom dns on your unraid or the dns supplied by your router?

are you running windows or linux or mac? you could dig/dnslookup the fqdn

 

i had that issue, cause of caching failure, so under windows you could try ipconfig /flushdns in console, this forces a relookup for that entry

 

 

[MrPatton75]

Thank you for the suggestion.  My primary DNS server on my UNRAID server is my OPNsense firewall.  The 2nd and 3rd are Cloudflare's.  I did an NSLOOKUP for jelly.mpcloud.io and it returned what I expect to see.  I will remove my firewall as a DNS server on my UNRAID server and see what that does.  The odd thing is that everything was working great without any changes to any of the systems involved.

[MrPatton75]

I finally just blew it all away and started over with a new NPM setup.  Now, I'm working on migrating to Traefik.  Tired of NPM breaking all the time just out of the blue.