[
https://issues.apache.org/jira/browse/KAFKA-4862?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15933400#comment-15933400
]
Colin P. McCabe commented on KAFKA-4862:
----------------------------------------
+1 for adding a configurable connect() timeout for the {{NetworkClient}}.
> Kafka client connect to a shutdown node will block for a long time
> ------------------------------------------------------------------
>
> Key: KAFKA-4862
> URL: https://issues.apache.org/jira/browse/KAFKA-4862
> Project: Kafka
> Issue Type: Bug
> Affects Versions: 0.9.0.0, 0.10.2.0
> Reporter: Pengwei
> Assignee: Pengwei
> Fix For: 0.11.0.0
>
>
> Currently in our test env, we found after one of the broker node crash(reboot
> or os crash), the client maybe still connecting to the crash node to send
> metadata request or other request, and it need about several minutes to
> aware the connection is timeout then try another node to connect to send the
> request. Then the client may still not aware the metadata change after
> several minutes.
> We don't have a connection timeout for the network client, we should add a
> connection timeout for the client
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
