Roman Khachatryan created FLINK-38940:
-----------------------------------------
Summary: Wait for all tasks to be running before triggering the
first checkpoint
Key: FLINK-38940
URL: https://issues.apache.org/jira/browse/FLINK-38940
Project: Flink
Issue Type: Improvement
Components: Runtime / Checkpointing
Affects Versions: 2.2.0
Reporter: Roman Khachatryan
Assignee: Roman Khachatryan
Fix For: 2.3.0
Currently, CheckpointCoordinator only checks the job status. The first
checkpoint trigger attempt fails if there's some slow initializing task
(especially if it doesn't wait for the initial delay).
This leads to one more checkpoint delay for the 1st checkpoint.
This change adds an option to wait for all tasks to run before triggering
a checkpoint.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
