[
https://issues.apache.org/jira/browse/KAFKA-8850?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Andreas Kohn updated KAFKA-8850:
--------------------------------
Description:
The current documentation for `fetch.min.bytes` says:
{quote}
The minimum amount of data the server should return for a fetch request. _If
insufficient data is available the request will wait for that much data to
accumulate before answering the request._ The default setting of 1 byte means
that fetch requests are answered as soon as a single byte of data is available
or the fetch request times out waiting for data to arrive. Setting this to
something greater than 1 will cause the server to wait for larger amounts of
data to accumulate which can improve server throughput a bit at the cost of
some additional latency.
{quote}
The highlighted sentence made me believe that Kafka would start waiting, and
_as soon as enough data was available would stop waiting and return_. In
reality the behavior is that Kafka will check for enough data, and if there
isn't will wait the full `fetch.max.wait.ms` time.
I would propose rewording the sentence to make that clearer, for example:
"If insufficient data is available the request will wait for
`fetch.max.wait.ms` before answering the request."
was:
The current documentation for `fetch.min.bytes` says:
> The minimum amount of data the server should return for a fetch request. _If
> insufficient data is available the request will wait for that much data to
> accumulate before answering the request._ The default setting of 1 byte means
> that fetch requests are answered as soon as a single byte of data is
> available or the fetch request times out waiting for data to arrive. Setting
> this to something greater than 1 will cause the server to wait for larger
> amounts of data to accumulate which can improve server throughput a bit at
> the cost of some additional latency.
The highlighted sentence made me believe that Kafka would start waiting, and
_as soon as enough data was available would stop waiting and return_. In
reality the behavior is that Kafka will check for enough data, and if there
isn't will wait the full `fetch.max.wait.ms` time.
I would propose rewording the sentence to make that clearer, for example:
"If insufficient data is available the request will wait for
`fetch.max.wait.ms` before answering the request."
> fetch.min.bytes documentation could be misunderstood
> ----------------------------------------------------
>
> Key: KAFKA-8850
> URL: https://issues.apache.org/jira/browse/KAFKA-8850
> Project: Kafka
> Issue Type: Improvement
> Components: documentation
> Reporter: Andreas Kohn
> Priority: Major
>
> The current documentation for `fetch.min.bytes` says:
> {quote}
> The minimum amount of data the server should return for a fetch request. _If
> insufficient data is available the request will wait for that much data to
> accumulate before answering the request._ The default setting of 1 byte means
> that fetch requests are answered as soon as a single byte of data is
> available or the fetch request times out waiting for data to arrive. Setting
> this to something greater than 1 will cause the server to wait for larger
> amounts of data to accumulate which can improve server throughput a bit at
> the cost of some additional latency.
> {quote}
> The highlighted sentence made me believe that Kafka would start waiting, and
> _as soon as enough data was available would stop waiting and return_. In
> reality the behavior is that Kafka will check for enough data, and if there
> isn't will wait the full `fetch.max.wait.ms` time.
> I would propose rewording the sentence to make that clearer, for example:
> "If insufficient data is available the request will wait for
> `fetch.max.wait.ms` before answering the request."
--
This message was sent by Atlassian Jira
(v8.3.2#803003)
