RE: Flink operator task opens threads internally

2023-08-04 Thread Kamal Mittal via user
Hello, Can you please share views for below? Rgds, Kamal From: Kamal Mittal Sent: 04 August 2023 11:19 AM To: Kamal Mittal ; Paul Lam Cc: user@flink.apache.org Subject: RE: Flink operator task opens threads internally Hello, I hope you have the context of use case clear now as described in

Re: Flink restored from an initially-specified checkpoint

2023-08-04 Thread Filip Karnicki
update: Our current line of thinking is that need this to be set in order to the checkpoint to live across job manager failures - ExternalizedCheckpointCleanup.DELETE_ON_CANCELLATION: Delete the checkpoint when the job is cancelled. The checkpoint state will only be available if the

Flink task manager failure scenario

2023-08-04 Thread Kamal Mittal via user
Hello, How flink behaves in case one of task manager POD fails out of a set of task managers PODs over say K8s environment? In my case, job remains in failed state even after giving re-start strategy with fixed delay (5 sec) and no. of attempts (5) with error as "Could not acquire minimum