Skip to main content
Software Engineering

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

10
  • Are you searching for general architecture solution or system/networks one ? On the network side, the simple answer would be to have a each rabbit server double by a fail-over, given that the R2 and monkey won't bother the fact that you might have messages not ordered when they come into R2 or the data processing cluster. If ordering is required, you will necessary need a single point gathering the data and so a failure point. Also if you architecture is behind a router, that router will also be a single failure point. Others solution would involve cache on the device and hamster side. Commented Apr 23, 2018 at 11:32
  • Given your description I would go the system/network solution route.The single entry point is R1 and we don't really care about order at any given step, provided that the time difference is not abysmal ( not greater than X minutes, for example. Each message is parsed rather quickly, so I don't see this happening ). Would you care to elaborate on your cache suggestion? I fail to see how it would help. Commented Apr 23, 2018 at 11:44
  • 1
    Well the cache solution would simply be able to store locally data and push them later in the case you still have a single point of failure, that is suppose your requirments make it so you shouldn't loss any datas or be able to at least have a history tracking one information every X minutes. Commented Apr 23, 2018 at 11:56
  • 4
    RabbitMQ is normally deployed in clusters of 3 or more nodes for high availability. The single point of failure I can see is the ingress point / network connection between the devices and R1. I guess you could either expose multiple nodes in the cluster, or maybe use a load balancer Commented Apr 23, 2018 at 12:16
  • Still, here we have the same issue. If I use the Load Balancer, then the Load Balancer becomes critical point of failure. If I create a Rabbit Cluster then the master Node ( exposed to the outside world ) becomes the point of failure. Commented Apr 23, 2018 at 12:20