Hi all,

TL;DR we hopefully have a fix for what might have caused many tests to fail non-deterministically. Please review.

Having investigated a few non-deterministic tests, I think I've finally got to the bottom of what might be causing some/many of them to fail.

I've raised a Jira and a PR that describe and fix the problem:
    https://issues.apache.org/jira/browse/BROOKLYN-291
    https://github.com/apache/brooklyn-server/pull/187

I'd appreciate a careful review. It only changes a few lines of code, but it's quite deep in Brooklyn so might have unforeseen consequences.

---
If this is the right fix, then we should look at (selectively) re-enabling the tests we marked as "Broken".

Aled

Reply via email to