Can someone explain to me how to achieve truly distributed architecture ? We are evaluating Hangfire as a solution in our SOA implementation.
We have multiple redundant services (eg Service A, Service B) each running their own Hangfire server. We would like to have one dashboard / database controlling all of these jobs. Queues are excellent for this !
The stumbling block we have is that the dashboard needs a reference to all assemblies that the jobs reference. Can someone explain why ? Just to produce a call tree ? Shouldn’t that be the responsibility of the server, not the dashboard? This is really ties all of our services to the hangfire dashboard, having to have cross solution/repository references to many, many assemblies that have nothing to do with the dashboard application.
I would have assumed that the dashboard could operate completely independently of the server - simply by reading the persisted data. This is a real show stopper