I’m having some intermittent network issues that result in a few
DBConnection.ConnectionError raised by the Postgres adapter (I guess?!). These errors happen sporadically, but I would like to collect metrics whenever this type of errors happen.
[my_app] [error] 09:27:30.214 application=db_connection module=DBConnection.Connection function=connect/2 Postgrex.Protocol (#PID<0.536.0>) failed to connect: ** (DBConnection.ConnectionError) tcp connect (postgres:5432): non-existing domain - :nxdomain
Currently I can only see them in the logs. I guess these errors are spit by the workers in the Ecto Repo? Is there a way to catch these errors? Maybe by indicating a callback to the Repo, that’s called whenever an error happens?
Thanks in advance!