Question: is savepointing required?

[hadronzoo]

Thank you for developing and open-sourcing this operator! Maintaining Flink jobs with state on Kubernetes without it has been error prone. Quick question: I'd also like to use this operator for Flink jobs without a persistent state requirement. Is it possible to deploy a new version of a job without restoring from a savepoint when state doesn't matter?

[anandswaminathan]

Hi,

Today, the operator only supports updates to an application by taking a savepoint. We will need to make some changes to the state machine in the operator to support updates without savepoint - something like SavepointInfo - Disabled.

Until we implement that, the only alternate option is to delete the application resource (with deleteMode set to None) and create a new one.

[JonnyIncognito]

Adding my +1 here. We have a number of medium-sized (6M+ records/sec) jobs that are stateless due to idempotency that we'd like to migrate over to FlinkK8sOperator. Cross-linking this issue from an internal backlog as we may help implement it ourselves out of need.

[davidbirdsong]

Does https://github.com/lyft/flinkk8soperator/pull/184 solve this issue?