Hi everybody,
i have an issue with my test suite. When i run each test separatly, everything is ok but in the whole test suite i have some DBConnection issues, althought all test pass.
Here is my case:
I have a Supervisor (a module that use Supervisor). This supervisor starts a scheduler. It is a genserver polling the DB periodically.
In the test, i start_supervised! this supervisor and do a few assert. But i get the following error:

MyXQL.Connection (#PID<0.265.0>) disconnected: ** (DBConnection.ConnectionError) owner #PID<0.567.0> exited

Client #PID<0.572.0> is still using a connection from owner at location:

:prim_inet.recv0/3
(myxql) lib/myxql/client.ex:171: MyXQL.Client.recv_packets/4
(myxql) lib/myxql/connection.ex:102: MyXQL.Connection.handle_execute/4
(ecto_sql) lib/ecto/adapters/sql/sandbox.ex:370: Ecto.Adapters.SQL.Sandbox.Connection.proxy/3
(db_connection) lib/db_connection/holder.ex:316: DBConnection.Holder.holder_apply/4
(db_connection) lib/db_connection.ex:1255: DBConnection.run_execute/5
(db_connection) lib/db_connection.ex:1342: DBConnection.run/6
(db_connection) lib/db_connection.ex:595: DBConnection.execute/4
(ecto_sql) lib/ecto/adapters/myxql/connection.ex:28: Ecto.Adapters.MyXQL.C
....

The Client #PID<0.572.0> is my scheduler and the owner is the test process.
As i read in the Ecto.Adapters.Sql.Sandbox, it seems to be normal: the test exits, the DBConnection lose its owner and notify the client process it doesn’t have owner anymore. But… the scheduler is querying the DB and is waiting for a reply. (https://hexdocs.pm/ecto_sql/Ecto.Adapters.SQL.Sandbox.html#module-faq)

I would like to know how to prevent such errors (too many red in the screen ;( )

Also, start_supervised is aimed to start the process under the test supervisor and the lifecycle of the test process is:

1. the test process is spawned
2. it runs setup/2 callbacks
3. it runs the test itself
4. it stops all supervised processes
5. the test process exits with reason :shutdown
6. on_exit/2 callbacks are executed in a separate process
   (https://hexdocs.pm/ex_unit/ExUnit.Callbacks.html)

It seems that my scheduler is still alive after the process exit. Why it is not killed with my supervisor at step 4?

thank you for your responses