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
Is your feature request related to a problem? Please describe.
Currently, users are able to control the activity task dispatch RPS (rate per second) as a configuration option before instantiating a worker. After a worker has started and polls for tasks, this RPS value can only be changed by re-deploying the workers. This has proved to be a pain-point for some users
Describe the solution you'd like
I shall be working on implementing an API which shall allow customers to set this RPS limit. My implementation shall also prevent re-deploying of new/existing workers for changing the RPS value.
Describe alternatives you've considered
NA - This seems to be a good solution to solve the problem at hand.
Additional context
This is how customers set this limit, as of today:
The idea is that the API set values would have the highest preference for these limits. So yes, TaskQueueActivitiesPerSecond will be ignored if you would have set a value using the API. This will effectively stop/remove the flapping behaviour thats present with the worker option.
Care will also be taken to have an "unset" feature for the API for those users who prefer to use the worker option as a way for setting the API limit. In that case, when unsetting is done, the value will be set the way it's being set as of today.
Is your feature request related to a problem? Please describe.
Currently, users are able to control the activity task dispatch RPS (rate per second) as a configuration option before instantiating a worker. After a worker has started and polls for tasks, this RPS value can only be changed by re-deploying the workers. This has proved to be a pain-point for some users
Describe the solution you'd like
I shall be working on implementing an API which shall allow customers to set this RPS limit. My implementation shall also prevent re-deploying of new/existing workers for changing the RPS value.
Describe alternatives you've considered
NA - This seems to be a good solution to solve the problem at hand.
Additional context
This is how customers set this limit, as of today:
The text was updated successfully, but these errors were encountered: