I am trying to setup Authentik but even when passing the env. var. AUTHENTIK_EXTERNAL_HOST to it, nothing shows up for OpenID Configuration URL inside of Authentik. This sent me down a rabbit hole where I was in the ssh terminal and could not find k3S or Kubectl at all. I tried numerous ways of finding them with no luck.
Eventually I converted my Authentik App to a custom app so I could try to manually add a new ingress with the reverse proxy but I got the error:
Additional property ingress is not allowed
Everything else in TrueNAS is working just fine and ultimately my goal is to get that URL populated. the K3S/Kubectl issue was supposed to be a means to getting this URL working.
I do have snapshots of my data set including the database. So I should be able to reinstall the app and restore pretty quickly. Any insight to get me pointed in the right direction would be appreciated.
this project may help you if the URL you are trying to get is the DNS name of a container itself. this makes the docker containers accessible on an internal network and assigns them names (similar to the old k3s system).
I did and it ran amok. I am setting it up in a docker container on my linux server instead. Way easier. Thank you all for the help though. This is a great community!
99 times out of 100 you can guess when someone has used AI (Artificial Idiocy) to try to take a short cut in learning how to do something.
Rant
AI should stand for Artificial Idiot or Artificial Idiocy rather than Artificial Intelligence because the problem is that AIs have literally ZERO intelligence. They ingest a huge amount of information and look for connecting patterns - they don’t only look for exact matches but also partial matches. And then you ask it a question or give it a task and it tries to match patterns in your question to the information it has ingested, and normally its an approximate match. BUT it doesn’t have any intelligence to judge the information it ingests for accuracy or veracity, it has no concept of the quality of the source, all information is of equal value, both the correct and the incorrect, and it then makes guesses about connections. Oh - and it has no idea what time is so it doesn’t know that information can become out of date or that things might have changed. So when you ask it something, it makes a guess at what you mean and a guess at the answer, and sometimes it will base the answers of false information it was given, and sometimes on out of date information, and sometimes it will make a guess about something it hasn’t actually been told (called hallucinating). And then it spouts the information to you in an authoritative tone as if it is established, indisputable fact because getting people to buy into your answers is how it gets people to use it more.
But there is literally ZERO intelligence, and zero cognitive reasoning - it is nothing more than a very advanced version of the 1962 Eliza chatbot that pattern matched your chat and responded with something that sounded sensible at first. So it is subject to GIGO (Garbage In = Garbage Out) with the added bonus of hallucinations and authoritative tone.
