[
https://issues.jboss.org/browse/WFCORE-1074?page=com.atlassian.jira.plugi...
]
Brian Stansberry commented on WFCORE-1074:
------------------------------------------
Emanuel and I were chatting, so this is a follow up to his last comment based on our
chat.
1) If an app tries to deploy and deployment doesn't go well due to some service
problems, but rollback-on-runtime-failure=false is set, that app is still
"deployed" in the management sense. The 'enabled' attribute will be set
to 'true' in the model, and MSC will have a bunch of services installed related to
the app. It just won't be functioning properly. So, sending a "deployed"
notification is valid.
2) The notifications really are only relevant for cases like WFCORE-860 where the end user
is directly using CLI/console to tweak the state of a scanner controlled deployment. There
the scanner needs notifications to be aware of user-initiated changes. But the scanner has
logic to ignore notifications emitted as a result of its *own* activity. Perhaps there is
a bug there, but the logic to ignore definitely exists.
Intermittent failure in DeploymentScannerUnitTestCase
-----------------------------------------------------
Key: WFCORE-1074
URL:
https://issues.jboss.org/browse/WFCORE-1074
Project: WildFly Core
Issue Type: Bug
Components: Domain Management
Reporter: Brian Stansberry
Assignee: ehsavoie Hugonnet
Fix For: 3.0.0.Alpha1
DeploymentScannerUnitTestCase occasionally fails, 4 times in the last six weeks or so per
Team City records:
http://brontes.lab.eng.brq.redhat.com/project.html?projectId=WildFlyCore&...
May not be relevant but the change log associated with the first failure shows WFCORE-903
/ WFCORE-860 as part of the change set:
http://brontes.lab.eng.brq.redhat.com/viewLog.html?buildId=70465&tab=...
--
This message was sent by Atlassian JIRA
(v6.4.11#64026)