New settings are not honored unless checkpoint is cleared.

[Yordan Pavlov]

Hello there,
I am running Flink 1.11.3 on Kubernetes deployment. If I change a
setting and re-deploy my Flink setup, the new setting is correctly
applied in the config file but is not being honored by Flink. In other
words, I can ssh into the pod and check the config file - it has the
new setting as I would expect. However the web interface for the job
keeps showing the old configuration and Flink as a whole keep running
with the old setting. The way to have the new setting considered is to
clear the checkpoint for the job stored in Zookeeper. Then I recover
the job using:

--fromSavepoint path_to_savepoint_or_checkpoint

My presumption is that the job configuration is stored in Zookeeper
along with other Flink data. Could someone shed some light on what I
am observing.

Thank you!

[Yun Gao]

Hi Yordan,

What are the settings that are changed during the tests?

Best,

Yun

------------------------------------------------------------------

From:Yordan Pavlov <[hidden email]>

Send Time:2021 Mar. 5 (Fri.) 23:36

To:user <[hidden email]>

Subject:New settings are not honored unless checkpoint is cleared.

Hello there,
I am running Flink 1.11.3 on Kubernetes deployment. If I change a
setting and re-deploy my Flink setup, the new setting is correctly
applied in the config file but is not being honored by Flink. In other
words, I can ssh into the pod and check the config file - it has the
new setting as I would expect. However the web interface for the job
keeps showing the old configuration and Flink as a whole keep running
with the old setting. The way to have the new setting considered is to
clear the checkpoint for the job stored in Zookeeper. Then I recover
the job using:

--fromSavepoint path_to_savepoint_or_checkpoint

My presumption is that the job configuration is stored in Zookeeper
along with other Flink data. Could someone shed some light on what I
am observing.

Thank you!