You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Found DurableExecutorService error prone in cases when application is submitting larger amount of Tasks.
Example:
Having singe member application with 10 partitions and DurableExecutorService capacity parameter is 3.
I expect DurableExecutorService to be able to store 30 Tasks (Capacity * Partitions).
But due to random partition selection nature of DurableExecutorServcieProxy.getTaskPartitionId() where some partitions being full while others are not, I end up having "RejectExecutionException: Capacity (n) is reached!" exception, causing the specific task to be discarded, even if the application has not reached its maximum storage capacity.
I propose to add some logic that will count with filled partitions when it is generating partition id for Tasks.
I Know about about PartitionAware option, but this in not letting me fill partitions evenly.
The text was updated successfully, but these errors were encountered:
Hello @Jarikkomarik , thanks a lot for the notice. I'd kindly suggest you to inform our engineers through https://slack.hazelcast.com/ for any workaround or alternative options in a quicker way. Then, accordingly, we can update the documentation.
Hi, I have some feedback about this page
Found DurableExecutorService error prone in cases when application is submitting larger amount of Tasks.
Example:
Having singe member application with 10 partitions and DurableExecutorService capacity parameter is 3.
I expect DurableExecutorService to be able to store 30 Tasks (Capacity * Partitions).
But due to random partition selection nature of DurableExecutorServcieProxy.getTaskPartitionId() where some partitions being full while others are not, I end up having "RejectExecutionException: Capacity (n) is reached!" exception, causing the specific task to be discarded, even if the application has not reached its maximum storage capacity.
I propose to add some logic that will count with filled partitions when it is generating partition id for Tasks.
I Know about about PartitionAware option, but this in not letting me fill partitions evenly.
The text was updated successfully, but these errors were encountered: