Either that or a heartbeat which would reset the timeout timer. This would also require an API to confirm that a service request completed with success or failure status.
Only then the service request would be removed from requests queue. If timeout is reached, the request would become active/visible again, and delivered to next available worker, since it would now be the oldest request still on the queue. Standard message broker behavior, I tought. Sent from my iPad. Regularly foiled by autocorrect. But duck it.. > On May 3, 2016, at 12:11, Michal Vyskocil <[email protected]> wrote: > > Hi, > > By timeout you mean that client expects an answer for service request in $foo > seconds? > > Dne 2. 5. 2016 1:15 PM napsal uživatel "Osiris Pedroso" <[email protected]>: >> Hi, >> >> I am new to Malamute and wanted to exchange words with some other user of >> the Service Requests API in Malamute. >> >> Malamute being a message broker, I expected to find some sort of timeout >> and/or retry logic for in-flight service requests, but from the API I don't >> think those exist. >> I wonder how people use the Service Request API to accomplish that. >> >> Thanks, >> Osiris >> >> _______________________________________________ >> zeromq-dev mailing list >> [email protected] >> http://lists.zeromq.org/mailman/listinfo/zeromq-dev > _______________________________________________ > zeromq-dev mailing list > [email protected] > http://lists.zeromq.org/mailman/listinfo/zeromq-dev
_______________________________________________ zeromq-dev mailing list [email protected] http://lists.zeromq.org/mailman/listinfo/zeromq-dev
