This came up in testing and in the migration of an existing Trident 1.4.6rc2 instance to a new environment. When the DB cannot be reached by the Trident daemon, the Trident web frontend does not display any error indicating the DB connection failed - it simply shows a basis, dead version of the Trident frontend. Subsequently, there is no error indicator in the Trident logs or the webserver logs.
When an issue such as this occurs, Trident should probably throw an error, such as "Could not connect to DB" (and the corresponding exact error that was thrown by the underlying postgres system) in the error logs, and the Frontend should show "An error occurred" or something to indicate there was an error in the connection.
This came up in testing and in the migration of an existing Trident 1.4.6rc2 instance to a new environment. When the DB cannot be reached by the Trident daemon, the Trident web frontend does not display any error indicating the DB connection failed - it simply shows a basis, dead version of the Trident frontend. Subsequently, there is no error indicator in the Trident logs or the webserver logs.
When an issue such as this occurs, Trident should probably throw an error, such as "Could not connect to DB" (and the corresponding exact error that was thrown by the underlying postgres system) in the error logs, and the Frontend should show "An error occurred" or something to indicate there was an error in the connection.