activemq-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "pfid (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (AMQCPP-557) ActiveMQ C++ library leaks threads
Date Fri, 12 Dec 2014 08:14:13 GMT

     [ https://issues.apache.org/jira/browse/AMQCPP-557?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

pfid updated AMQCPP-557:
------------------------
    Description: 
Steps to reproduce:

1) open connection
2) attach to the process and note the number of threads (info thr)
3) shutdown the broker
4) your programm should try to automatically reconnect
5) start the broker
6) repeat 2) to see that the number of threads has increased

We had a bigger outage because of this recently, when the library would refuse to reconnect
to the (multiple times) restarted broker, because the maximum number of threads was reached
on OS level.

Attached a very minimal example which will reproduce the problem.

  was:
Steps to reproduce:

1) open connection
2) attach to the process and note the number of threads (info thr)
3) shutdown the broker
4) your programm should try to automatically reconnect
5) start the broker
6) repeat 2) to see that the number of threads has increased

We had a bigger outage because of this recently, when the library would refuse to reconnect
to the (multiple times) restarted broker, because the maximum number of threads was reached
on OS level.

It seems im not able to attach files, so please see this very minimal example which will reproduce
the problem:

http://pastebin.com/szYYE56Y


> ActiveMQ C++ library leaks threads
> ----------------------------------
>
>                 Key: AMQCPP-557
>                 URL: https://issues.apache.org/jira/browse/AMQCPP-557
>             Project: ActiveMQ C++ Client
>          Issue Type: Bug
>    Affects Versions: 3.8.2
>         Environment: Solaris 5.10
>            Reporter: pfid
>            Assignee: Timothy Bish
>         Attachments: main.cc
>
>
> Steps to reproduce:
> 1) open connection
> 2) attach to the process and note the number of threads (info thr)
> 3) shutdown the broker
> 4) your programm should try to automatically reconnect
> 5) start the broker
> 6) repeat 2) to see that the number of threads has increased
> We had a bigger outage because of this recently, when the library would refuse to reconnect
to the (multiple times) restarted broker, because the maximum number of threads was reached
on OS level.
> Attached a very minimal example which will reproduce the problem.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message