[
https://issues.jenkins-ci.org/browse/JENKINS-13416?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Bruno Meneguello updated JENKINS-13416:
---------------------------------------
Attachment: jenkins-1.459.diff
Description:
I'm using on demand slaves (on amazon) started by a shell script. When starting
a job with all slaves disconnected, all my slaves are started together.
When in debug, I'd noticed that "RetentionStrategy" "Demand" is testing
"Computer.getDemandStartMilliseconds()" to connect the slave, but all slaves
(that 'll take some time to wake up) pass in test.
Shouldn't this strategy take in account if there are executor enough and nodes
starting?
I,ve made a change in RetentionStrategy that solves the problem. Anyone see any
problem with this solution?
was:
I'm using on demand slaves (on amazon) started by a shell script. When starting
a job with all slaves disconnected, all my slaves are started together.
When in debug, I'd noticed that "RetentionStrategy" "Demand" is testing
"Computer.getDemandStartMilliseconds()" to connect the slave, but all slaves
(that 'll take some time to wake up) pass in test.
Shouldn't this strategy take in account if there are executor enough and nodes
starting?
> On demand slave provisioning is starting all available slaves
> -------------------------------------------------------------
>
> Key: JENKINS-13416
> URL: https://issues.jenkins-ci.org/browse/JENKINS-13416
> Project: Jenkins
> Issue Type: Bug
> Components: slave-setup
> Affects Versions: current
> Reporter: Bruno Meneguello
> Assignee: Kohsuke Kawaguchi
> Priority: Minor
> Attachments: jenkins-1.459.diff
>
>
> I'm using on demand slaves (on amazon) started by a shell script. When
> starting a job with all slaves disconnected, all my slaves are started
> together.
> When in debug, I'd noticed that "RetentionStrategy" "Demand" is testing
> "Computer.getDemandStartMilliseconds()" to connect the slave, but all slaves
> (that 'll take some time to wake up) pass in test.
> Shouldn't this strategy take in account if there are executor enough and
> nodes starting?
> I,ve made a change in RetentionStrategy that solves the problem. Anyone see
> any problem with this solution?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.jenkins-ci.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
