In Victoria, Trove moved to Docker images for the datastores. Any datastores that didn't update themselves were removed, regardless of state. I believe the only datastores currently supported are MariaDB, and PostgreSQL, though MySQL might be supported as well.
Personally. I find this to have been a bad example of the "containerize all the things" mentality. Trove now deploys an instance (vm) for each datastore instance, running a single container. The exception is if you request backups, in which case a backup container is added to the instance (vm).
Trove is installed in our OpenStack cloud, but we don't use it.
Thank you,
Dominic L. Hilsbos, MBA Vice President - Information Technology Perform Air International Inc. DHilsbos@PerformAir.com www.PerformAir.com
From: Ришат Азизов [mailto:rishat.azizov@gmail.com] Sent: Sunday, February 6, 2022 11:10 AM To: openstack-discuss@lists.openstack.org Subject: OpenStack Trove experimental datastores
Hello!
Why were the experimental datastores removed in the Trove OpenStack Victoria release? https://review.opendev.org/c/openstack/trove/+/728419
Thanks!