[
https://issues.apache.org/jira/browse/FLINK-21661?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17297854#comment-17297854
]
jiamo commented on FLINK-21661:
-------------------------------
In
[https://ci.apache.org/projects/flink/flink-docs-stable/dev/connectors/kinesis.html#polling-default-record-publisher-1]
while the latter modifies the sleep interval between each fetch (default is 200)
> SHARD_GETRECORDS_INTERVAL_MILLIS wrong use?
> --------------------------------------------
>
> Key: FLINK-21661
> URL: https://issues.apache.org/jira/browse/FLINK-21661
> Project: Flink
> Issue Type: Bug
> Components: Connectors / Kinesis
> Affects Versions: 1.12.1
> Reporter: jiamo
> Priority: Major
>
> kinesis `SHARD_GETRECORDS_INTERVAL_MILLIS`
> mean between normal getRecords sleep.
> But at end . The value was used in exception.
> {code:java}
>
> // sleep for the fetch interval before the next getRecords attempt with the
> // refreshed iterator
> if (expiredIteratorBackoffMillis != 0) {
> Thread.sleep(expiredIteratorBackoffMillis);
> }
> {code}
>
>
> {code:java}
> return new PollingRecordPublisher(
> startingPosition,
> streamShardHandle,
> metricsReporter,
> kinesisProxy,
> configuration.getMaxNumberOfRecordsPerFetch(),
> configuration.getFetchIntervalMillis());
> {code}
> But the last arg was *expiredIteratorBackoffMillis.*
>
> Is this a problem?
>
>
>
>
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
