[
https://issues.apache.org/jira/browse/ARTEMIS-642?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15396291#comment-15396291
]
ASF GitHub Bot commented on ARTEMIS-642:
----------------------------------------
Github user clebertsuconic commented on the issue:
https://github.com/apache/activemq-artemis/pull/672
I don't think we can merge this...
this is introducing a new datatype through the protocol and it will break
compatibility with older clients in both directions. You would need to add
handchecks for the version ID on the protocol if that's possible or not.
I wouldn't do a hashMap storing sessions. I would just disable
reconnections or kill all the sessions.
> Disable slow client reconnecting with KILL slow client policy
> -------------------------------------------------------------
>
> Key: ARTEMIS-642
> URL: https://issues.apache.org/jira/browse/ARTEMIS-642
> Project: ActiveMQ Artemis
> Issue Type: New Feature
> Affects Versions: 1.3.0
> Reporter: Chen Maoqian
> Priority: Minor
>
> It is possible to detect and handle(KILL, NOTIFY) slow consumers. In case
> policy KILL is used and connection factory used to create connection of slow
> consumer has attribute reconnect-attempts set to "-1", slow consumer
> reconnects to server. This consumer is then reconnects, is detected as slow
> and disconnected.
> Attribute reconnect-attempts="-1" should be ignored, and slow client should
> not be trying to reconnect once it is disconnected by KILL slow consumer
> policy.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
