You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@activemq.apache.org by "clebert suconic (Jira)" <ji...@apache.org> on 2019/08/26 15:48:06 UTC

[jira] [Reopened] (ARTEMIS-2368) Fix races on closing consumer

     [ https://issues.apache.org/jira/browse/ARTEMIS-2368?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

clebert suconic reopened ARTEMIS-2368:
--------------------------------------

> Fix races on closing consumer
> -----------------------------
>
>                 Key: ARTEMIS-2368
>                 URL: https://issues.apache.org/jira/browse/ARTEMIS-2368
>             Project: ActiveMQ Artemis
>          Issue Type: Bug
>          Components: Broker
>    Affects Versions: 2.9.0
>            Reporter: yangwei
>            Priority: Major
>             Fix For: 2.10.0
>
>          Time Spent: 20m
>  Remaining Estimate: 0h
>
> There are some race conditions between closing consumer and session failover:
>  # When sending SessionConsumerCloseMessage timeout, conn::fail() is called resulting in current consumer is recreated on server side again but removed on client side.
>  # Between  SessionConsumerCloseMessage sending and session::removeConsume(), session failover happens resulting in consumer inconsistency.
>  # When an iteration over the consumers entrySet is made to recreate consumer during session failover process, consumer::close() is called, maybe leading to consumer inconsistency or ConcurrentModificationException thrown.
> We traverse through the cloned consumers entrySet, add synchronized block and check if consumer is closed while recreating consumer to fix the issue.



--
This message was sent by Atlassian Jira
(v8.3.2#803003)