[
https://issues.apache.org/jira/browse/BEAM-13203?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17465019#comment-17465019
]
Alexey Romanenko commented on BEAM-13203:
-----------------------------------------
[~echauchot]
bq. If the write is backpressured, there is no way to slow down the source
(because its watermark will got forward) so the records will be buffered in the
write operation indeed
Yes but since there is no internal back-pressure mechanism in Beam, and, in
that way, "sink" never knows about how fast is "source", then it won't be so
easy to deal with this. Well, this topic is quite complicated and out of the
scope of this Jira, so here we just need to make sure that don't potentially
lose the data using async client.
> Potential data loss when using SnsIO.writeAsync
> -----------------------------------------------
>
> Key: BEAM-13203
> URL: https://issues.apache.org/jira/browse/BEAM-13203
> Project: Beam
> Issue Type: Bug
> Components: io-java-aws
> Reporter: Moritz Mack
> Priority: P1
>
> This needs to be investigated, reading the code suggests we might be losing
> data under certain conditions e.g. when terminating the pipeline. The async
> processing model here is far too simplistic.
> The bundle won't ever know about pending writes and won't block to wait for
> any such operation. The same way exceptions are thrown into nowhere. Test
> cases don't capture this as they operate on completed futures only (so
> exceptions in the callbacks get thrown on the thread of processElement).
> {code:java}
> client.publish(publishRequest).whenComplete((response, ex) -> {
> if (ex == null) {
> SnsResponse<T> snsResponse = SnsResponse.of(context.element(), response);
> context.output(snsResponse);
> } else {
> LOG.error("Error while publishing request to SNS", ex);
> throw new SnsWriteException("Error while publishing request to SNS", ex);
> }
> }); {code}
> Also, this entirely removes backpressure from a stream. When used with a much
> faster source we will continue to accumulate more and more memory as the
> number of concurrent pending async operations is not limited.
> Spotify's scio contains a
> [JavaAsyncDoFn|https://github.com/spotify/scio/blob/main/scio-core/src/main/java/com/spotify/scio/transforms/JavaAsyncDoFn.java]
> that illustrates how it can be done.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
