[
https://issues.apache.org/jira/browse/FLINK-26589?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zichen Liu updated FLINK-26589:
-------------------------------
Description:
h2. Motivation
During the normal course of the KDS and KDF's sink's operation, we will likely
encounter many failed persists to the destination due to throttling or
otherwise. These are currently logged as warn.
*Proposed Solution*
1. We would like these messages to be logged as debug.
2. It would also be beneficial to adjust the wording to clarify that the
entries will be requeued for retry.
*Files to be changed*
* KinesisStreamsSinkWriter
** when a request succeeds with a subset of the records and the remainder
needs to be retried
** when the request has failed
* KinesisFirehoseSinkWriter
** when a request succeeds with a subset of the records and the remainder
needs to be retried
** when the request has failed
was:
h2. Motivation
During the normal course of the KDS and KDF's sink's operation, we will likely
encounter many failed persists to the destination due to throttling or
otherwise. These are currently logged as warn.
*Proposed Solution*
1. We would like these messages to be logged as debug.
2. It would also be beneficial to adjust the wording to clarify that the
entries will be requeued for retry.
*Files to be changed*
* *KinesisStreamsSinkWriter*
** *when a request succeeds with a subset of the records and the remainder
needs to be retried*
** *when the request has failed*
* *KinesisFirehoseSinkWriter*
** *when a request succeeds with a subset of the records and the remainder
needs to be retried*
** {*}when the request has failed{*}{*}{*}
> Update the logging level of Kinesis Streams and Firehose sinks
> --------------------------------------------------------------
>
> Key: FLINK-26589
> URL: https://issues.apache.org/jira/browse/FLINK-26589
> Project: Flink
> Issue Type: Improvement
> Components: Connectors / Kinesis
> Affects Versions: 1.15.0
> Reporter: Zichen Liu
> Assignee: Ahmed Hamdy
> Priority: Minor
>
> h2. Motivation
> During the normal course of the KDS and KDF's sink's operation, we will
> likely encounter many failed persists to the destination due to throttling or
> otherwise. These are currently logged as warn.
> *Proposed Solution*
> 1. We would like these messages to be logged as debug.
> 2. It would also be beneficial to adjust the wording to clarify that the
> entries will be requeued for retry.
> *Files to be changed*
> * KinesisStreamsSinkWriter
> ** when a request succeeds with a subset of the records and the remainder
> needs to be retried
> ** when the request has failed
> * KinesisFirehoseSinkWriter
> ** when a request succeeds with a subset of the records and the remainder
> needs to be retried
> ** when the request has failed
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
