dannycranmer commented on a change in pull request #15157:
URL: https://github.com/apache/flink/pull/15157#discussion_r596638275
##########
File path:
flink-connectors/flink-connector-kinesis/src/main/java/org/apache/flink/streaming/connectors/kinesis/internals/publisher/polling/PollingRecordPublisher.java
##########
@@ -168,8 +177,8 @@ private GetRecordsResult getRecords(String shardItr, int
maxNumberOfRecords)
// sleep for the fetch interval before the next getRecords
attempt with the
// refreshed iterator
- if (expiredIteratorBackoffMillis != 0) {
- Thread.sleep(expiredIteratorBackoffMillis);
+ if (fetchIntervalMillis != 0) {
+ Thread.sleep(fetchIntervalMillis);
Review comment:
It is called after `GetRecords` in the `adjustRunLoopFrequency` at line
216. It does not simply sleep for `fetchIntervalMillis`, it tries to make the
total loop time `fetchIntervalMillis`. For instance, if `GetRecords` takes
200ms and the `fetchIntervalMillis` is set to 1000ms, it would sleep for 800ms.
This allows the interval to be more meaningful, fo example:
- 1000ms: 1 read per second
- 500ms: 2 reads per second
- 250ms: 4 reads per second
- etc
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
