Thanks for pointing it out. You findings are correct, our code depends at the moment on
pg_ctlcluster for restarting DB in order to drop all connections. We were not aware that it is available in Debian-family only. Internally this will be fixed, but I can’t provide estimation about release date of the fix.
From top of my head, I guess you could workaround this issue by doing something like described below. We use the missing command only to restart the database and drop all connections, which is pre-requirement for
DROP DATABASE call. Note that I don’t have at the moment environment similar to you, so I can’t actually test those steps. Hopefully you will be able to take it from here.
- make sure that all connections to postgres (ie. from IDE) are closed.
- stop jenkins service in order to make sure that no running jobs are connected to the DB
- comment out a line with call to pg_ctlcluster in
KoreKontrol Germany GmbH - Spryker hosting