[
https://issues.jboss.org/browse/ISPN-1323?page=com.atlassian.jira.plugin....
]
Galder Zamarreño commented on ISPN-1323:
----------------------------------------
Right, so if the channel has been stopped, TEST_PING.stop() has been called, so an
instance variable checking whether stop was called on TEST_PING should be enough.
That's assuming that when the node is resurrected, the same TEST_PING instance is
used. Dan, is that right?
TEST_PING doesn't remove the discovery info for a node if
findInitialMembers() is called after stop()
-----------------------------------------------------------------------------------------------------
Key: ISPN-1323
URL:
https://issues.jboss.org/browse/ISPN-1323
Project: Infinispan
Issue Type: Bug
Components: Test Suite
Affects Versions: 5.0.0.FINAL
Reporter: Dan Berindei
Assignee: Galder Zamarreño
Fix For: 5.1.0.ALPHA1, 5.1.0.FINAL
Because of
https://issues.jboss.org/browse/JGRP-1336, sometimes a node is
"resurrected" and TEST_PING.findInitialMembers() is called even after the
channel has been stopped.
If that happens the node is not removed from the discovery registry and the next node to
start from the same test will hang trying to contact the dead node and elect a
coordinator.
--
This message is automatically generated by JIRA.
For more information on JIRA, see:
http://www.atlassian.com/software/jira