style95 commented on a change in pull request #5194:
URL: https://github.com/apache/openwhisk/pull/5194#discussion_r790246722
##########
File path:
core/controller/src/main/scala/org/apache/openwhisk/core/loadBalancer/FPCPoolBalancer.scala
##########
@@ -46,7 +46,8 @@ class FPCPoolBalancer(config: WhiskConfig,
extends LoadBalancer {
private implicit val executionContext: ExecutionContext =
actorSystem.dispatcher
- private implicit val requestTimeout: Timeout = Timeout(5.seconds)
+ // This value is given according to the total waiting time at QueueManager
for a new queue to be created.
+ private implicit val requestTimeout: Timeout = Timeout(8.seconds)
Review comment:
I just increased this to align it with the retry timeout at the queue
manager.
https://github.com/apache/openwhisk/blob/master/core/scheduler/src/main/scala/org/apache/openwhisk/core/scheduler/queue/QueueManager.scala#L421
In the default configuration, the queue manager waits for a new queue to be
created for around 8 seconds.
Regarding this value, it should not be too high because if a target
scheduler becomes unresponsive, the controller should try to create a queue
against another healthy scheduler.
This change might need to be reverted because controllers would retry to
create queues after 8 seconds but the activations would already be dropped by
the queue manager.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
