[
https://issues.apache.org/jira/browse/GEODE-915?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15222153#comment-15222153
]
ASF subversion and git services commented on GEODE-915:
-------------------------------------------------------
Commit a7b9b90f8a7b46c488eb30b10499a1f72c57ce04 in incubator-geode's branch
refs/heads/develop from [~dschneider]
[ https://git-wip-us.apache.org/repos/asf?p=incubator-geode.git;h=a7b9b90 ]
GEODE-915: fix FORCE_INVALIDATE_EVENT
- new unit test ClientServerForceInvalidateDUnitTest
- ProxyRegionMap now checks for force invalidate before it throws
EntryNotFound
- AbstractRegionMap invalidate now does force invalidate unless the region is
not initialized
As part of the work also did the following code cleanup work:
1. Removed the isInvalidate parameter from shouldCreateCBEvent since this
method did the exact same thing all values of this parameter.
2. Changed cacheWriteBeforeInvalidate to be serverInvalidate since all this
method does is possibly send the invalidate to the server.
3. EntryEventImpl toString will now include inhibitCacheListenerNotification if
it is true
4. Removed the unused parameter invokeCallbacks and forceNewEntry from
serverInvalidate
> invalidate event when entry does not exist or already invalid
> -------------------------------------------------------------
>
> Key: GEODE-915
> URL: https://issues.apache.org/jira/browse/GEODE-915
> Project: Geode
> Issue Type: New Feature
> Components: regions
> Reporter: Darrel Schneider
> Assignee: Darrel Schneider
> Fix For: 1.0.0-incubating.M2
>
>
> When a Region invalidate finds that the entry does not exist or is already
> invalid it does not invoke afterInvalidate on the CacheListener. Some users
> have requested that afterInvalidate be invoked even in these cases.
> What was done to for them was a system property named
> "gemfire.FORCE_INVALIDATE_EVENT" was added. If it was set to "true" then
> afterInvalidate was invoked even if the entry did not exist or was already
> invalid. In both these cases the oldValue on the EntryEvent passed to
> afterInvalidate was set to null.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
