Calling -2 for vote if not too late.

Have git bisected from 3.5.0 to HEAD and found the commit that introduced the behaviour:

ec629f7d511eb910b4e80112a9fbe85ed8786f10 is the first bad commit
commit ec629f7d511eb910b4e80112a9fbe85ed8786f10
Author: Igor Fedorenko ifedorenko@apache.org
Date: Tue Apr 11 07:59:34 2017 -0700

MNG-6209 better executeMojo thread context classloader

Signed-off-by: Igor Fedorenko <ifedorenko@apache.org>

:040000 040000 570fa9308365b0ee98d57ac3f8006691bd9ade4d 3c6791665c9d41f0e4a3893ea99621cc50e8b91b M maven-core

Not exactly sure what/why/how this problem manifests in a testable manner yet however.

Mark

On 11 Sep 2017, at 23:19, Mark Derricutt wrote:

On 11 Sep 2017, at 18:10, Stephen Connolly wrote:

I wonder if mng-6275 is affecting that plugin

Didn't manage to get a chance to look into this tonight :( Tho that ticket mentions nashorn, phantonjs is a C/native headless browser library, so it doesn't feel like it could be related.

If there's a build available with a fix for that, welcome to give it a bash.

I'll try carve out some time in the morning to see if I can make a simple standalone project...


"The ease with which a change can be implemented has no relevance at all to whether it is the right change for the (Java) Platform for all time." — Mark Reinhold.

Mark Derricutt
http://www.theoryinpractice.net
http://www.chaliceofblood.net
http://plus.google.com/+MarkDerricutt
http://twitter.com/talios
http://facebook.com/mderricutt


"The ease with which a change can be implemented has no relevance at all to whether it is the right change for the (Java) Platform for all time." — Mark Reinhold.

Mark Derricutt
http://www.theoryinpractice.net
http://www.chaliceofblood.net
http://plus.google.com/+MarkDerricutt
http://twitter.com/talios
http://facebook.com/mderricutt