[
https://issues.apache.org/jira/browse/ARTEMIS-1312?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16108563#comment-16108563
]
ASF GitHub Bot commented on ARTEMIS-1312:
-----------------------------------------
GitHub user franz1981 opened a pull request:
https://github.com/apache/activemq-artemis/pull/1433
ARTEMIS-1312 TimedBuffer doubled timeout with blocking flush
It changes the TimedBuffer timeout expiration using the effective flush
time (if available) to perform the sleep time after a submitted flush.
This wouldn't impact ASYNCIO that is supposed to be nonblocking on flush
submission, but will let NIO/MAPPED to have the timeout matching the configured
IOPS.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/franz1981/activemq-artemis
timed_buffer_sanity_test
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/activemq-artemis/pull/1433.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #1433
----
commit ae77bad17e38138c189b0c574f8541f91a3d780b
Author: Francesco Nigro <[email protected]>
Date: 2017-07-28T14:25:53Z
ARTEMIS-1312 TimedBuffer doubled timeout with blocking flush
----
> TimedBuffer doubled timeout with blocking flush
> -----------------------------------------------
>
> Key: ARTEMIS-1312
> URL: https://issues.apache.org/jira/browse/ARTEMIS-1312
> Project: ActiveMQ Artemis
> Issue Type: Bug
> Reporter: Francesco Nigro
> Assignee: Francesco Nigro
> Priority: Minor
>
> TimedBuffer is not correctly handling the configured timeout when blocking
> flush is used (ie NIO,MAPPED), leading to halved IOPS or that cases.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
