Return Counter and List data when Allocating a GameServer about agones HOT 6 CLOSED

In our case, we prefer that the results of ListAction and CounterAction are also returned.

Can you explain why? I'm curious as to the use case.
Also, since you get the results, and you know what you send, you could do the diff yourself quite easily.

I thought that if we got the result of appended values, simply add them into our storage with the allocated GameServer data without calculating the diff of request and response.
I plan to cache them into our storage to use these roomIDs from other hosts to connect to the GameServer without sending request to allocator service.

Also, since you get the results, and you know what you send, you could do the diff yourself quite easily.

You are right, we can calculate the diff easily so it's not essential.

from agones.

I'm focused on some other stuff at the moment, so if you want to have a go, please feel free!

If you have questions, feel free to drop them here, or come by #development in Slack!

from agones.

We have a similar use case in implementing the High Density GameServer pattern.
When we request a GameServer allocation, we would like to be able to use list to pre-populate the room in GameServer in batches.
This will reduce the number of actual allocation requests, which is more efficient.
However, the current implementation of CountsAndLists does not necessarily append all list items defined. If you try to fill more than the capacity, you will get an error and the GameServer allocation will succeed with only some of the items filled.
In this case, the expected state of the GameServer and the actual state of the GameServer will inevitably be out of sync.
To solve this problem, it would be helpful if counts and lists are returned in the allocation response as suggested.

from agones.

In our case, we prefer that the results of ListAction and CounterAction are also returned.

For example, the following form.

Request:
{
  "lists": {
    "rooms": {
      "addValues": ["room-3", "room-4", "room-5"]
    }
  },
  "gameServerSelectors": [
    {
      "gameServerState": 1,
      "lists": {
        "rooms": {
          "minAvailable": 1
        }
      }
    },
    {
      "gameServerState": 0
    }
  ]
}

Selected GS:
kind: GameServer
metadata:
  name: gs-1
status:
  state: Allocated
  lists: {"rooms":{"capacity":3,"values":["room-1", "room-2"]}}

Result of GS:
kind: GameServer
metadata:
  name: gs-1
status:
  state: Allocated
  lists: {"rooms":{"capacity":3,"values":["room-1", "room-2", "room-3"]}}

Response:
{
  "gameServerName": "gs-1",
  "ports": [],
  "address": "",
  "addresses": [],
  "nodeName": "",
  "source": "",
  "metadata": {},
  "lists": ["room-1", "room-2", "room-3"]
  "listAction": {
    "rooms": {
      "addedValues": ["room-3"]
    }
  }
}

from agones.

In our case, we prefer that the results of ListAction and CounterAction are also returned.

Can you explain why? I'm curious as to the use case.

Also, since you get the results, and you know what you send, you could do the diff yourself quite easily.

from agones.

@markmandel

Are you already working on this issue?
If not, would you mind if I tackle it?

from agones.