[
https://issues.apache.org/jira/browse/KAFKA-3135?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15214993#comment-15214993
]
Jason Gustafson commented on KAFKA-3135:
----------------------------------------
I made a little progress understanding this problem. The 5 second pause that
we're seeing is caused by the TCP persist timer, which kicks in when the remote
window size gets small enough. You can see this by watching the traffic through
tcpdump and confirming the persist timeout event with 'netstat -s'. Clearly
there are some cases where the client is not reading the data off the socket
fast enough, which is causing the window size to fall and the server to back
off.
> Unexpected delay before fetch response transmission
> ---------------------------------------------------
>
> Key: KAFKA-3135
> URL: https://issues.apache.org/jira/browse/KAFKA-3135
> Project: Kafka
> Issue Type: Bug
> Affects Versions: 0.9.0.0
> Reporter: Jason Gustafson
> Assignee: Jason Gustafson
> Priority: Critical
> Fix For: 0.10.0.1
>
>
> From the user list, Krzysztof Ciesielski reports the following:
> {quote}
> Scenario description:
> First, a producer writes 500000 elements into a topic
> Then, a consumer starts to read, polling in a loop.
> When "max.partition.fetch.bytes" is set to a relatively small value, each
> "consumer.poll()" returns a batch of messages.
> If this value is left as default, the output tends to look like this:
> Poll returned 13793 elements
> Poll returned 13793 elements
> Poll returned 13793 elements
> Poll returned 13793 elements
> Poll returned 0 elements
> Poll returned 0 elements
> Poll returned 0 elements
> Poll returned 0 elements
> Poll returned 13793 elements
> Poll returned 13793 elements
> As we can see, there are weird "gaps" when poll returns 0 elements for some
> time. What is the reason for that? Maybe there are some good practices
> about setting "max.partition.fetch.bytes" which I don't follow?
> {quote}
> The gist to reproduce this problem is here:
> https://gist.github.com/kciesielski/054bb4359a318aa17561.
> After some initial investigation, the delay appears to be in the server's
> networking layer. Basically I see a delay of 5 seconds from the time that
> Selector.send() is invoked in SocketServer.Processor with the fetch response
> to the time that the send is completed. Using netstat in the middle of the
> delay shows the following output:
> {code}
> tcp4 0 0 10.191.0.30.55455 10.191.0.30.9092 ESTABLISHED
> tcp4 0 102400 10.191.0.30.9092 10.191.0.30.55454 ESTABLISHED
> {code}
> From this, it looks like the data reaches the send buffer, but needs to be
> flushed.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
