Re: [workflow-plugin] parallel threads greater than matching node label executors silently discard unassigned threads

[Jesse Glick]

On Mon, Nov 9, 2015 at 6:56 PM, Brian Ray <be_...@sbcglobal.net> wrote:
> [Crossposting from Jenkins Users because I have not been lucky there with
> workflow questions.]

This is a developer list. Your message is somewhere between a user
question and a bug report. If a user question, use the user’s list, or
the StackOverflow tag `jenkins-workflow`.

> Is this a bug?

Sounds like a bug. If you can reproduce from scratch in a clean
environment, please file it in JIRA.

[Brian Ray]

Figured it out and deleted the original posts because they were effectively noise in the Google Groups. It was developer error. I was looking at a test input that would've generated more branches than free matching slaves, but mistakenly rebuilding my job with another test input where threads < available matching slaves.

When I fed it the "threads > slaves" input, parallel worked as expected: 5 concurrent threads, then 5 more, etc. until all branches in the map were executed.

Thanks,

[Brian Ray]

And I'll stick to GG Jenkins Users and SO jenkins-workflow the next time. I appreciate the feedback on where to post.