Github user serranom commented on the issue:
https://github.com/apache/twill/pull/23
I found that there is still a race condition in
ApplicationMasterService.launchRunnable. If the number of instances is
increased right after the original request is fullfilled, the current logic can
result in the original request not being polled resulting in future requests
hanging. This seems to be a fairly unlikely case in the real world, but I'll
file a JIRA for it. For now, I've reworked the test to wait until
launchRunnable has done the polling for the original request. This PR is now
complete.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
