QQ regarding autoscaling with adaptive scheduler (...
# troubleshooting
a
QQ regarding autoscaling with adaptive scheduler (flink 1.18, k8s operator 1.8) Started with initial parallelism of 64, after some time it scaled down to a maximum of 16. Each TM is configured to have 16 CPU. But instead of releasing the nodes it still use all of them 4. Anyways to improve that (e.g. to make autoscaler release at least one of them?)
g
Yea this is a known issue as there is not really a setting that would make Flink use as few TMs as possible . Could you please open a Jira ticket for this?
👍 1
a
sent a request to open jira account.