Branch: refs/heads/master
Home: https://github.com/jenkinsci/throttle-concurrent-builds-plugin
Commit: 6e9129fc99d0c1eaa2214b9aea6281ee97da3388
https://github.com/jenkinsci/throttle-concurrent-builds-plugin/commit/6e9129fc99d0c1eaa2214b9aea6281ee97da3388
Author: Robert Sandell <[email protected]>
Date: 2012-05-07 (Mon, 07 May 2012)
Changed paths:
M
src/main/java/hudson/plugins/throttleconcurrents/ThrottleQueueTaskDispatcher.java
Log Message:
-----------
Optimizing by changing from getComputers to getNodes.
In environments with hundreds of slaves, hundred of builds in the queue
to be throttled and hundreds of simultaneous web requests.
Jenkins then spends a lot of time in the unnecessary effort to slowly
sort the list of computers for this plugin to count the number of builds
causing the web requests to wait for the queue to release its monitor
so they can render it on the page giving the impression that the master
is overloaded while in fact it is just waiting.
Changing to getNodes() should speed responsiveness up a bit.
Change-Id: I09e0f2b2f2d94fc4366ed84b2e51ecaea961efc1
Commit: d17f210653219f58d16fb5d9879a33b391e80ffe
https://github.com/jenkinsci/throttle-concurrent-builds-plugin/commit/d17f210653219f58d16fb5d9879a33b391e80ffe
Author: Andrew Bayer <[email protected]>
Date: 2012-05-07 (Mon, 07 May 2012)
Changed paths:
M
src/main/java/hudson/plugins/throttleconcurrents/ThrottleQueueTaskDispatcher.java
Log Message:
-----------
Merge pull request #2 from rsandell/master
Optimizing by changing from getComputers to getNodes.
Compare:
https://github.com/jenkinsci/throttle-concurrent-builds-plugin/compare/e6c07a2...d17f210
