Junrui Li created FLINK-35622:
---------------------------------
Summary: Filter out noisy "Coordinator of operator xxxx does not
exist" exceptions in batch mode
Key: FLINK-35622
URL: https://issues.apache.org/jira/browse/FLINK-35622
Project: Flink
Issue Type: Technical Debt
Components: Runtime / Coordination
Reporter: Junrui Li
In batch mode, the Flink JobManager logs frequently print "Coordinator of
operator xxxx does not exist or the job vertex this operator belongs to is not
initialized." exceptions when using the collect() method.
This exception is caused by the collect sink attempting to fetch data from the
corresponding operator coordinator on the JM based on the operator ID. However,
batch jobs do not initialize all job vertices at the beginning, and instead,
initialize them in a sequential manner. If a job vertex has not been
initialized yet, the corresponding coordinator cannot be found, leading to the
printing of this message.
These exceptions are harmless and do not affect the job execution, but they can
clutter the logs and make it difficult to find relevant information, especially
for large-scale batch jobs.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
