[Traefik] All my services went down or are unreliable after adding Let's Encrypt
This gets hairy and I have no idea what's going on, please excuse me for the sporadic details…
I set up a traefik instance for my domain arvigeus.one with few services: config here.
Because I was already locked out of Let's Encrypt certificate once, I used the staging server.
I set the CNAME domain in Linode as * @ (for wildcard).
After I made sure all was working, it was time to enable https… and it didn't go well. After digging around, the problem turned out to be a problem with CNAME blocking the letsecrypt process. No idea what was going on, I deleted the CNAME record and my site appeared to be secure.
And here my real problem started. Services stopped working. Or at least they became unreachable. Some of them. No errors or whatsoever.
It seems it takes some time for services to "boot up" before they become reachable, but today all of them were unreachable again. Right now I added again the * @ CNAME record and it seems to be mostly working… except Jellyfin that consistently refuses to become reachable. I used it before the Let's Encrypt change, it was alright. Now it simply refuses. No idea if other services will continue working in few hours (there were no crashes in the logs).
I guess I have several questions:
1) Do I need to keep * @ CNAME on, or should remove it in order to keep letsencrypt working?
2) How can I understand why Jellifyn (https://media.arvigeus.one) is not working? Logs indicate nothing wrong, the config used to work. The only thing I changed since the time it was working was removing this line (traefik.http.routers.jellyfin-rtr.tls.certresolver=${CERT_RESOLVER}) and removing the staging server. I also tested by removing all other services and leaving only Jellyfin. Same config is working locally.
I've been having all kinds of problems all week long, opened and closed so many threads, I feel guilty for asking so much. None of the tutorials mentioned it is going to be so much headache. I apologize!