Github user upthewaterspout commented on a diff in the pull request:
https://github.com/apache/geode/pull/303#discussion_r91392448
--- Diff:
geode-core/src/main/java/org/apache/geode/cache/wan/GatewaySender.java ---
@@ -400,4 +400,19 @@
public int getMaxParallelismForReplicatedRegion();
+
+ /**
+ * Destroys the GatewaySender. Before destroying the sender, caller
needs to to ensure that the
+ * sender is stopped so that all the resources (threads, connection pool
etc.) will be released
+ * properly. Stopping the sender is not handled in the destroy. Destroy
is carried out in
+ * following steps: 1. Take the lifeCycleLock. 2. If the sender is
attached to any application
+ * region, throw an exception. 3. Close the GatewaySenderAdvisor. 4.
Remove the sender from the
+ * cache. 5. Destroy the region underlying the GatewaySender.
+ * <p>
+ * In case of ParallelGatewaySender, the destroy operation does
distributed destroy of the QPR. In
+ * case of SerialGatewaySender, the queue region is destroyed locally.
--- End diff --
Is this still accurate? I see in the code you now call stop before calling
destroy, so does the sender still need to be stopped?
This looks like it's the internal documentation about what happens. Maybe
right something more oriented towards the user for the public method?
Also add an @since Geode 1.1 tag.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
