Error 500 after updating or restarting AURA
I'm not exactly sure if what I'm describing here is one and the same issue. However, the symptoms are the same.
Just yesterday I've updated AURA on my office machine on which I'm developing the migration tool for the migration of our old backend (Radio Orange) to AURA. The steps were the following (updating to commit a06b9b46):
git pull
docker compose pull
docker compose up -d
Then, when trying to login to the dashboard a message
Error: something went wrong when signing in. See console for details.
popped up which obviously refers to a server error 500 which I see when I go to /steering/admin
. The only thing that helped in this case was to re-init the database:
docker-compose run --rm steering ./run.sh init-db
... after which, of course, all content in the database was gone.
While investigating the reason for the problem I've had a look at the steering-postgres
logs with
compose logs -f --tail=100 steering-postgres
that returns a long list of repeating messages:
steering-postgres | 2022-10-28 11:45:58.325 CEST [860] STATEMENT: SELECT "oidc_provider_responsetype"."id", "oidc_provider_responsetype"."value", "oidc_provider_responsetype"."description" FROM "oidc_provider_responsetype" LIMIT 21
So far, so good. For developing it's not a big issue if I have to re-init the database after an update - creating the necessary client in the Steering administration and then running the sync script once recreates the old database content. However, for running AURA in production it certainly is.
Unfortunately the problem doesn't only happen after a system update. It's enough to stop AURA with docker compose down -v
and then start it up with docker compose up -d
- trying to login results in an error 500 and the only solution (that I know) is to reinitialize the database with the above mentioned consequences.