Github user zhuoliu commented on a diff in the pull request:
https://github.com/apache/storm/pull/765#discussion_r40731960
--- Diff: storm-core/src/clj/backtype/storm/daemon/worker.clj ---
@@ -137,14 +138,11 @@
check highWaterMark and lowWaterMark for backpressure"
(disruptor/disruptor-backpressure-handler
(fn []
- "When worker's queue is above highWaterMark, we set its backpressure
flag"
- (if (not @(:backpressure worker))
- (do (reset! (:backpressure worker) true)
- (WorkerBackpressureThread/notifyBackpressureChecker
(:backpressure-trigger worker))))) ;; set backpressure no matter how the
executors are
+ (reset! (:transfer-backpressure worker) true)
+ (WorkerBackpressureThread/notifyBackpressureChecker
(:backpressure-trigger worker)))
--- End diff --
Will this cause the WorkerBackpressureThread gets too frequently notified
even when it is not necessary to do so?
(In most cases, both the :backpressure and :transfer-backpressure have not
changed, we may not really need to notify the backpressure-thread to check all
executors's flags, which can be additional overhead.)
---
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.
---
