Greetings,
Solving an unforeseen problem is especially nerve-wracking if you work on your gate to the internet (kind of).
On my weekly SSH round-trip to my servers, where I update system and software (mostly guided, I only intervene on errors)
I updated my self-hosted Bitwarden instance. this is where the problem begins.
Unknowingly at this point, I successfully set my be-loved password manager offline due to an oversight from my side.
So the days pass and nothing happens. I can still log-on to websites, my TOTPs work just fine.
Today I noticed that I cannot log on to my Bitwarden server on a new device. Weird.
The browser extension seemed fine so I suspected it was server and yeah, a beautiful 502 Bad Gateway, but
the error was coming from the nginx provided by Bitwarden, since I use an Apache HTTP daemon for proxying, so on to fixing the error.
Maybe a Bitwarden update would help? No.
System update? Try again.
Ok. Let's ask the mighty problem solver ... Google.
(Fast-forward an hour: nothing described my situation.)
Fine, time to list docker processes with docker ps. And there you find the culprit some containers were in an unhealthy.
After consulting the Bitwarden manual I verified that at least the mssql container was fine.
Shortly after, I nailed down the culprit to be the admin container not connecting to the mssql server.
After another system update and a reboot, the problem persisted.
On to the manual again.
And then there it was. Docker.
Turns out the Docker installation was slightly out of date (if you agree that one major version is slightly)
So installing the newest Docker version from the Docker repo directly (not Debian), rebuilding Bitwarden and my
sacred blue angel came back to life.
Lessons learned: