I have updated the PR for review and also added a new test.
https://github.com/apache/flink-connector-elasticsearch/pull/107
Could you take a look, Ahmed?
Thanks,
On Tue, Jun 11, 2024 at 5:23 AM Ahmed Hamdy wrote:
> Hi Mingliang,
> Yes sounds like a good solution, I am not very familiar with
Hi Mingliang,
Yes sounds like a good solution, I am not very familiar with ElasticSearch
internals and APIs but will try to assist with the PR when ready.
Best Regards
Ahmed Hamdy
On Tue, 11 Jun 2024 at 07:07, Mingliang Liu wrote:
> Thank you Ahmed for the explanation.
>
> The current Elasticse
Thank you Ahmed for the explanation.
The current Elasticsearch 8 connector already uses the
FatalExeptionClassifier for fatal / non-retriable requests [1]. It's very
similar to what you linked in the AWS connectors. Currently this is only
used for fully failed requests. The main problem I was conc
Hi Mingliang,
We already have a mechanism for detecting and propagating
Fatal/Non-retryable exceptions[1]. We can use that in ElasticSearch similar
to what we do for AWS connectors[2]. Also, you can check AWS connectors for
how to add a fail-fast mechanism to disable retrying all along.
> FLIP-45
Hi all,
Currently the Elasticsearch 8 connector retries all items if the request
fails as a whole, and retries failed items if the request has partial
failures [1]. I think this infinitely retries might be problematic in some
cases when retrying can never eventually succeed. For example, if the
re
