[
http://jira.jboss.com/jira/browse/JGRP-686?page=comments#action_12398844 ]
Vladimir Blagojevic commented on JGRP-686:
------------------------------------------
Not that we are not changing at all boolean return value semantics of getState. If state
was received at channel level getState will return true. Channel#getState will return
false in case of a first singleton member requesting state or if state was not received at
channel level. We are not doing anything different. For example, Channel#getState will
still return true even though state installation failed (e.g. threw an Exception) on
application level.
We are going to improve getState semantics in 3.0 release. See
http://jira.jboss.com/jira/browse/JGRP-619 for details.
Inconsistent Channel#getState return semantics between byte and
streaming state transfer
----------------------------------------------------------------------------------------
Key: JGRP-686
URL:
http://jira.jboss.com/jira/browse/JGRP-686
Project: JGroups
Issue Type: Bug
Affects Versions: 2.5, 2.6, 2.5.1, 2.5.2, 2.6.1
Reporter: Vladimir Blagojevic
Assigned To: Vladimir Blagojevic
Fix For: 2.7, 2.6.2
We have inconsistency between byte and streaming based state transfer when it comes to
semantics of returning from Channel#getState call. In streaming state transfer calling
thread returns from Channel#getState when application installs a state (or fails) and for
byte transfer calling thread returns from Channel#getState as soon as state arrives to
channel and *prior* to state installation in client application. We should make
Channel#getState calling thread return *after* state installation for byte based state
transfer as well.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
http://www.atlassian.com/software/jira