What Should I Do If the Error Message "slot request timeout" Is Displayed When I Submit a Flink Job?

Symptom

When a Flink job is submitted, JobManager is started successfully. However, TaskManager remains in the starting state until timeout. The following error information is displayed:

org.apache.flink.runtime.jobmanager.scheduler.NoResourceAvailableException: Could not allocate the required slot within slot request timeout. Please make sure that the cluster has enough resources

Possible Causes

• The resources in the YARN queue are insufficient. As a result, TaskManager fails to start.
• Your JAR files conflict with those in the environment. You can execute the WordCount program to determine whether the issue occurs.
• If the cluster is in security mode, the SSL certificate of Flink may be incorrectly configured or has expired.

Solution

1. Add resources to the YARN queue.
2. Exclude the Flink and Hadoop dependencies in your JAR files so that Flink and Hadoop can depend only on the JAR files in the environment.
3. Reconfigure the SSL certificate of Flink..