[
https://issues.apache.org/jira/browse/FLINK-10020?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16568542#comment-16568542
]
ASF GitHub Bot commented on FLINK-10020:
----------------------------------------
kailashhd commented on a change in pull request #6482: [FLINK-10020] [kinesis]
Support recoverable exceptions in listShards.
URL: https://github.com/apache/flink/pull/6482#discussion_r207622963
##########
File path:
flink-connectors/flink-connector-kinesis/src/main/java/org/apache/flink/streaming/connectors/kinesis/proxy/KinesisProxy.java
##########
@@ -409,7 +416,7 @@ private ListShardsResult listShards(String streamName,
@Nullable String startSha
int attemptCount = 0;
// List Shards returns just the first 1000 shard entries. Make
sure that all entries
// are taken up.
- while (listShardsResults == null) { // retry until we get a
result
+ while (attemptCount <= listShardsMaxAttempts &&
listShardsResults == null) { // retry until we get a result
Review comment:
The earlier contract was to wait till we get a result.
https://issues.apache.org/jira/browse/FLINK-10020 does not talk about breaking
this contract. I personally believe maxAttemptCount is better since listShard
works in a periodic thread and we are bound to try again after 'X' seconds.
Just wanted to point this out. I like this approach better.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
> Kinesis Consumer listShards should support more recoverable exceptions
> ----------------------------------------------------------------------
>
> Key: FLINK-10020
> URL: https://issues.apache.org/jira/browse/FLINK-10020
> Project: Flink
> Issue Type: Improvement
> Components: Kinesis Connector
> Reporter: Thomas Weise
> Assignee: Thomas Weise
> Priority: Major
> Labels: pull-request-available
>
> Currently transient errors in listShards make the consumer fail and cause the
> entire job to reset. That is unnecessary for certain exceptions (like status
> 503 errors). It should be possible to control the exceptions that qualify for
> retry, similar to getRecords/isRecoverableSdkClientException.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
