- 最新
- 最多得票
- 最多評論
I've asked the service team to take a look. If you have technical support on your AWS account and this is urgent, I would suggest opening an issue there for a faster response.
While we wait for the service team, you could check:
- What your matchmaking timeouts are
- If you had actual spare capacity on your fleet, did you have an idle processes at the time which could have taken the game session. If GameLift had to bring up a new instance then that takes minutes.
- Look at how you are managing spare capacity, if its exhausted (because your incoming match rate spiked dramatically) then players will be left waiting for new instances.
- Check your fleet for server instabilities, if your servers start receiving new game sessions and then fail, the match may be left in limbo for a while waiting for the server to be marked as unhealthy so it can be replaced.
Hi!
To further help debug your issue, which region was the queue placing this ticket in and what was the name of that queue?
Thanks!
Region : Oregon(us-west-2) The name of that queue : THEEND-42258-Shp-PlayTest-Guaranteed
Thanks you
Hi there,
At the beginning, the fleet had 1 instance and there are matches formed and submitted to this queue THEEND-42283-Shp-PlayTest-Guaranteed. So the queueDepth metric was high and waitTime metric of this queue is 249 seconds. The scaling policy needs some time to react and the fleet was set to 3 instance at 2020-09-30 17:23:03, which is after this ticket has been in PLACING for 4 minutes. You can check the queue metrics in CloudWatch to see the details.
Best regards
相關內容
- AWS 官方已更新 1 年前