Github user vanzin commented on a diff in the pull request:
https://github.com/apache/spark/pull/2019#discussion_r16731252
--- Diff:
core/src/main/scala/org/apache/spark/network/ConnectionManager.scala ---
@@ -280,42 +280,46 @@ private[spark] class ConnectionManager(
}
while(!keyInterestChangeRequests.isEmpty) {
+ // Expect key interested in OP_ACCEPT is not change its interest
val (key, ops) = keyInterestChangeRequests.dequeue()
-
try {
- if (key.isValid) {
+ key.synchronized {
--- End diff --
Sure, but again, all that is doing is avoiding an exception. An exception
that is handled properly by the code.
So, what is better:
- synchronize on every key to avoid throwing an exception
- don't synchronize and let exceptions be handled
Given that this is a hot loop, and during normal operation keys are not
being cancelled left and right, I think the second approach is better. So I'd
just turn the log messages into `logDebug`, since they're expected.
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
