[
https://issues.redhat.com/browse/WFCORE-4967?page=com.atlassian.jira.plug...
]
P K updated WFCORE-4967:
------------------------
Steps to Reproduce:
I created a minimal example project which demonstrates this EJB timer issue:
https://github.com/guestlinelabs/wildfly-timer-issue
h2. Recreation steps
# Deploy sample application to the wildfly server (via cli or just by copy it to
standalone/deployments).
# Start the server
h5. Note
Above order is very important because when app is deployed after RequestController is
resumed, issue is not occurring.
h3. Expected
All timers from SomeTimers class are triggered after the server is resumed.
This can be confirmed by finding an entry in server log at INFO level, which message
starts with timers OK
h3. Incorrect behaviour
Some (random) timers are omitted, there is a message in logs saying timers FAIL or there
is no log entry at all.
was:I will create minimal deployment and provide a link here as I know what is the cause
already.
RequestController drops queued tasks randomly during server start.
------------------------------------------------------------------
Key: WFCORE-4967
URL:
https://issues.redhat.com/browse/WFCORE-4967
Project: WildFly Core
Issue Type: Bug
Components: Server
Affects Versions: 11.0.0.Final
Reporter: P K
Assignee: Jeff Mesnil
Priority: Major
Our deployment is quite complex, EAR app containing several modules.
After move to different data-center some of EJB timers will not be triggered at all after
server restart (randomly).
There was no change to wildfly server version and application itself.
Only hardware/network changed and issue started occurring.
Classified as : WEIRDO and some race conditions was assumed.
--
This message was sent by Atlassian Jira
(v7.13.8#713008)