Not able to get allocated and scheduled time for gameservers

[oll-aroundarmor]

Issue: I was trying to get metrics for gameserver in prometheus by using agones_gameserver_state_duration_sum but it is providing me values for agones-controller/fleet allocated and scheduled state.

What I want: I want to get the agones-sdk metrics in order to get the details for how much time an individual gameserver was in an allocated and scheduled state.

Help-needed:

1. Suggestion if there is any way through which I can get those metrics to reflect for individual gameservers.
2. Is there a way I can get a service and service monitor for agones-sdk?

Any suggestions would be greatly appreciated.

[markmandel]

Thanks for the question!

Suggestion if there is any way through which I can get those metrics to reflect for individual gameservers.

You can't get at individual GameServer metrics as metrics are aggregate values.

If you want to watch individual GameServers, I recommend writing some small code to interact with the Kubernetes API: https://agones.dev/site/docs/guides/access-api/

Is there a way I can get a service and service monitor for agones-sdk?

Sorry, not quite sure what you mean by this - can you expand on it in more detail?

[oll-aroundarmor]

Sure, Mark!

As I earlier stated I wanted the individual gameserver's allocated and scheduled time durations but I checked and found that the service monitor is only configured for the agones-controller which in response gives the output of the whole fleet.

I wanted to add an additional service and servicemonitor which will be providing the outputs for agones-sdk.

For example: In the below presentation, you will find that the gameservers are getting scheduled, and based on it its SDK state changes and gets logged on the Kubernetes events. If I can get the service and service monitor for agones-sdk I will be able to get the metrics for individual gameservers from agones_gameserver_state_duration_sum.

`

LAST-SEEN	TYPE	REASON	OBJECT	MESSAGE
119s	Normal	Allocated	gameserver/project-cricket-fleet-w7pf6-5lbl7	Allocated
26s	Normal	Shutdown	gameserver/project-cricket-fleet-w7pf6-5lbl7	SDK state change
26s	Normal	Shutdown	gameserver/project-cricket-fleet-w7pf6-5lbl7	Deletion started
26s	Normal	Killing	pod/project-cricket-fleet-w7pf6-5lbl7	Stopping container agones-gameserver-sidecar
26s	Normal	Shutdown	gameserver/project-cricket-fleet-w7pf6-5lbl7	Deleting Pod project-cricket-fleet-w7pf6-5lbl7
26s	Normal	Killing	pod/project-cricket-fleet-w7pf6-5lbl7	Stopping container project-cricket

`

Let me know if you have any questions.

[markmandel]

If I can get the service and service monitor for agones-sdk I will be able to get the metrics for individual gameservers from agones_gameserver_state_duration_sum.

Sorry I don't follow how this approach works? Metrics are aggregated, they aren't per GameServer, and you don't really want to do this, because metric services can't handle combinatoric explosion with an essentially infinite number of labels.

I'll put it another way - getting per-GameServer information out of metrics is going to be a non starter.

Watch the Kubernetes API for GameServer changes to do this. Or use the Event stream, that would work as well.

[github-actions[bot]]

'This issue is marked as Stale due to inactivity for more than 30 days. To avoid being marked as 'stale' please add 'awaiting-maintainer' label or add a comment. Thank you for your contributions '

[github-actions[bot]]

This issue is marked as obsolete due to inactivity for last 60 days. To avoid issue getting closed in next 30 days, please add a comment or add 'awaiting-maintainer' label. Thank you for your contributions