[KAFKA-13631] Warning "Ignoring await stop request for non-present connector..." when shutting down connector via API in distributed mode

• Type: Bug
• Status: Open
• Priority: Minor
• Resolution: Unresolved
• Affects Version/s: 2.8.1
• Component/s: connect
• Assignee: Unassigned
• Reporter: Simon Schall
• Created: 31/Jan/22 08:05
• Updated: 07/Sep/22 12:24

Description

• Setup: kafka with version 2.8.1 with scala version 2.13
• Start zookeeper, kafka server and kafka-connect-distributed
• Create a FileStreamSinkConnector (e.g. via the following request:

 curl --location --request PUT 'localhost:8083/connectors/local-file-sink/config'  \
--header 'Content-Type: application/json' \
--data-raw '{
"name":"local-file-sink",
"connector.class":"FileStreamSink",
"tasks.max":1,
"file":"test.sink.txt",
"topics":"connect-test"
}'

• shutdown the connector via the api as followed:

curl --location --request DELETE 'localhost:8083/connectors/local-file-sink/'

 

In the log of connect-distributed will appear the following warnings (for complete log see attachement, warning appears in l.2103ff. ):

[2022-01-28 10:52:42,246] INFO Stopping connector local-file-sink (org.apache.kafka.connect.runtime.Worker:382)
[2022-01-28 10:52:42,247] WARN Ignoring stop request for unowned connector local-file-sink (org.apache.kafka.connect.runtime.Worker:385)
[2022-01-28 10:52:42,247] WARN Ignoring await stop request for non-present connector local-file-sink (org.apache.kafka.connect.runtime.Worker:410)

 

In the log it looks like, it is tried twice to stop the connector. The first attempt is successful (log l.2096ff.), but the second attempt causes the warning, as the connector is already stopped. It is not traceable for us why the connector is tried to stop twice, but it seems to be wrong.

Furthermore, the problem also occurs with any other connector we tried so far, so it is not specific to the FileStreamSinkConnector.