Now we define CLUSTER_LIFETIME in acceptance-tests (default value 3 days) so Hive will manage the cluster and remove it when the time passes.

Moreover, we run cleanup-cloud-resources task in a cron job that also removes clusters older than 3 days.

If those two params are not synced, it might remove clusters at wrong time or leak credentials.

Probably the best approach would be to exclude Hive clusters in cleanup-cloud-resources but then we need to handle secret removal somehow.