Check the state of running clones and auto-cleanup the stalled clones
Goal
-
check the state of running sessions and remove failed clones -- cleanup the info about failed clones after N days -
----- we do not know how to reproduce it ----- -
first, test this case -
remove clones without sessions (fatal status) after the idle timeout(e.g., failed to start a new session) -
make the reset command more resilient to failed clones
-
In case when container is failed or stopped(joe#101 (closed))
Session: joe-bp0bqkl4h85j6h8o5kfg
ERROR: Pqsl error: psql: error: could not connect to server: could not connect to server: Connection refused
Is the server running on host "localhost" (127.0.0.1) and accepting
TCP/IP connections on port 6000?
TODO / How to implement
- Provision interface: add
CheckClone
/CheckClones
method- Check Docker container is running
- Check Postgres instance can be accessed
- Cloning: run
CheckClone
/CheckClones
on interval
Acceptance criteria
Edited by Nikolay Samokhvalov