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
Is your feature request related to a problem? Please describe.
Currently the multi-cluster configuration relies on the "agent" service which watches for GameServer CRDs in a cluster, and a "relay" service which aggregates the information from all agents. Currently the setup uses just a single instance of of agent in each cluster, and a single relay for all agents and proxies. In order to make the setup for resilient, we should allow the following:
Multiple agents in a single cluster communicating the same configuration to a single relay.
Agents sending their configuration to multiple relays.
Proxies communicating to multiple relays simultaneously.
Describe the solution you'd like
Don't have a solid solution that we should pick, the simplest solution is to spawn multiple tasks for each relay in the agent and proxy, and allow there to be redundant data, this is the simplest, but means that memory requirements grow with each redundant service, it would make more sense to do some level of merging of information, to reduce the amount of data being stored.
The text was updated successfully, but these errors were encountered:
Is your feature request related to a problem? Please describe.
Currently the multi-cluster configuration relies on the "agent" service which watches for
GameServer
CRDs in a cluster, and a "relay" service which aggregates the information from all agents. Currently the setup uses just a single instance of of agent in each cluster, and a single relay for all agents and proxies. In order to make the setup for resilient, we should allow the following:Describe the solution you'd like
Don't have a solid solution that we should pick, the simplest solution is to spawn multiple tasks for each relay in the agent and proxy, and allow there to be redundant data, this is the simplest, but means that memory requirements grow with each redundant service, it would make more sense to do some level of merging of information, to reduce the amount of data being stored.
The text was updated successfully, but these errors were encountered: