Task Manager Direct Memory Used
This alert is triggered when the task manager direct memory used is higher than 100 MB during 5 minutes.
This alert is triggered when the failed checkpoints is higher than 3 during 5 minutes.
Job Manager Heap Memory Usage
This alert is triggered when the job manager heap memory usage is higher than 10 MB during 5 minutes.
Shuffle Service Used Memory
This alert is triggered when the shuffle netty used memory is higher than 2 MB during 5 minutes.
Comprehensive monitoring quickstart for Apache Flink
New Relic detects your Apache Flink metrics like job manager (checkpoints, direct, mapped, metaspace memory usage), task manager (shuffle service, operators), heap memory usage and more.
Why monitor Apache Flink?
An ideal Apache Flink performance monitoring provides an easy way for effective monitoring and alerting of your Apache Flink jobs.
What’s included in this quickstart?
New Relic's Apache Flink monitoring quickstart boasts instant full-stack observability out-of-the-box:
- High value alerts for your Apache Flink (errors, usage metrics and JVM).
- Easy to use performance monitoring dashboards.
- Custom alerts that proactively inform developers about the status of their Apache Flink jobs.