You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
For longer polling durations, the task will spend most of its time asleep. If the task is asleep when it is getting killed, such as when a rebalance is happening or the config is changed, the task is likely not to return within the configured graceful timeout period, instead causing Graceful stop of task connector_name-0 failed. errors.
We suspect this may be leading to some issues we are seeing with connectors seemingly "hanging", (with only the table monitor thread running).
For longer polling durations, the task will spend most of its time asleep. If the task is asleep when it is getting killed, such as when a rebalance is happening or the config is changed, the task is likely not to return within the configured graceful timeout period, instead causing
Graceful stop of task connector_name-0 failed.
errors.We suspect this may be leading to some issues we are seeing with connectors seemingly "hanging", (with only the table monitor thread running).
FWIW the confluent connector seems to have also experienced this issue: confluentinc/kafka-connect-jdbc#677
I have a PR with a fix: #59
The text was updated successfully, but these errors were encountered: