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
{{ message }}
This repository was archived by the owner on Apr 8, 2024. It is now read-only.
At the moment, trying an operation that fails causes the onReady event to fire.
This happens because I forced the client ready state to allow the worker to keep polling after the connection drops (for example, when the broker goes away then comes back).
This is going to require a refactor.
The polling operation can fail, but this should not prevent it polling again.
However, it should not emit ready.
So it needs two representations of ready. One is "ready to retry operation". This is when the channel state is IDLE or CONNECTED. The other is "ready to rock", this is when an operation returns a result.
The Middleware.Events.Ready should only be emitted in the second case, and we need the Internal signal for the first case.
The text was updated successfully, but these errors were encountered:
jwulf
added a commit
to jwulf/zeebe-client-node-js
that referenced
this issue
May 6, 2020
At the moment, trying an operation that fails causes the onReady event to fire.
This happens because I forced the client ready state to allow the worker to keep polling after the connection drops (for example, when the broker goes away then comes back).
This is going to require a refactor.
The polling operation can fail, but this should not prevent it polling again.
However, it should not emit ready.
So it needs two representations of ready. One is "ready to retry operation". This is when the channel state is IDLE or CONNECTED. The other is "ready to rock", this is when an operation returns a result.
The Middleware.Events.Ready should only be emitted in the second case, and we need the Internal signal for the first case.
The text was updated successfully, but these errors were encountered: