[
https://issues.apache.org/jira/browse/FLINK-4197?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15373018#comment-15373018
]
ASF GitHub Bot commented on FLINK-4197:
---------------------------------------
Github user tzulitai commented on a diff in the pull request:
https://github.com/apache/flink/pull/2227#discussion_r70457227
--- Diff: docs/apis/streaming/connectors/kinesis.md ---
@@ -299,3 +299,41 @@ Otherwise, the returned stream name is used.
Other optional configuration keys can be found in `KinesisConfigConstants`.
+### Using non-AWS Kinesis Endpoints
--- End diff --
"N"on-AWS. Also, I think it'd be good to let the title be "Using Non-AWS
Kinesis Endpoints for Testing", to better point out the purpose of doing this.
> Allow Kinesis Endpoint to be Overridden via Config
> --------------------------------------------------
>
> Key: FLINK-4197
> URL: https://issues.apache.org/jira/browse/FLINK-4197
> Project: Flink
> Issue Type: Improvement
> Components: Kinesis Connector
> Affects Versions: 1.0.3
> Reporter: Scott Kidder
> Priority: Minor
> Labels: easyfix
> Fix For: 1.0.4
>
> Original Estimate: 1h
> Remaining Estimate: 1h
>
> I perform local testing of my application stack with Flink configured as a
> consumer on a Kinesis stream provided by Kinesalite, an implementation of
> Kinesis built on LevelDB. This requires me to override the AWS endpoint to
> refer to my local Kinesalite server rather than reference the real AWS
> endpoint. I'd like to add a configuration property to the Kinesis streaming
> connector that allows the AWS endpoint to be specified explicitly.
> This should be a fairly small change and provide a lot of flexibility to
> people looking to integrate Flink with Kinesis in a non-production setup.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
