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
Once there are multiple Marathon/Mesos controllers, there should still only be one instance of Consular running.
If we containerize Consular it is easier to run it from Marathon. If it is run from Marathon, when a node falls over, Marathon can move Consular over to a new node and things can carry on running relatively uninterrupted.
The text was updated successfully, but these errors were encountered:
After speaking to Jeremy, there are potentially different ways to improve availability without running as a container.
The idea of running as a container scheduled by Marathon I got from Mesos-DNS and Kubernetes. They both do something similar with their DNS services (although obviously Kubernetes doesn't use Marathon).
Once there are multiple Marathon/Mesos controllers, there should still only be one instance of Consular running.
If we containerize Consular it is easier to run it from Marathon. If it is run from Marathon, when a node falls over, Marathon can move Consular over to a new node and things can carry on running relatively uninterrupted.
The text was updated successfully, but these errors were encountered: