When a worker dies while working on a request, said request is reenqueued only if a heartbeat check is triggered on a client.
We should allow to explicitly reenqueue all such requests, which would facilitate certain operations on some clusters we manage.
The implementation should simply be a function that checks the heartbeats immediately and does the same collection as the periodic heartbeat checker on the client.
When a worker dies while working on a request, said request is reenqueued only if a heartbeat check is triggered on a client.
We should allow to explicitly reenqueue all such requests, which would facilitate certain operations on some clusters we manage.
The implementation should simply be a function that checks the heartbeats immediately and does the same collection as the periodic heartbeat checker on the client.