Questions tagged with Amazon GameLift

Does a collection of GameLift service samples exist? I'm developing a Battle Royale game in Unity using the Unity Plugin for GameLift and I have the basics functioning.. starting with CloudFormation template deployment, FlexMatch with game session queue, and automatic backfill. Now I'm starting to build a Lobby system, where a player can create/join/apply to join/invite to join other online clients. This seems like something that should exist in sample code (CloudFormation template, Lambda files, DynamoDB tables, SNS Topic Event Handlers, etc.). Does it? I may have it built and working by the time I get a response here, but here's what I'm thinking. LobbyTable LobbyId PlayerId # lobby owner TicketId # matchmaking ticket id PlayerLobbyTable PlayerId LobbyId Lambda functions to handle setting up a table items when a player: creates a lobby # creates LobbyTable item for PlayerId as owner application/invitation is accepted # creates PlayerLobbyTable item for PlayerId leaves/booted from lobby # deletes PlayerLobbyTable item of leaving PlayerId, picks new lobby owner and assigns if anyone remains in lobby or deletes LobbyTable for LobbyId if no players remain change lobby owner # changes LobbyTable PlayerId to new lobby owner update matchmaking ticket id # when FlexMatch MatchmakingSucceeded event, add IpAddress, Port, DnsName, etc. attributes to LobbyTable item, send message to all players in lobby that the matchmaking succeeded. Thank you.

I cant seem to get access to my linux gamelift instance, when I call gamelift get-instance-access from aws CLI using fleetid and instance id, i am getting timed out. This is a different experience when i do this in a windows gamelist instance. error is An error occurred (InternalServiceException) when calling the GetInstanceAccess operation (reached max retries: 2): Timeout waiting for reply Why? This is a freshly created instance, deployed a build, created a fleet, and nothing special done. following tutorials and documentations on how to access an instance fleet, everything is correct and followed on the dot, but cannot get the instance access in CLI because of the timing issue i described above.. This is getting frustrating

Are there plans to have iOS / Android SDK for GameSparks? Is it possible to access FlexMatch through GameSparks CloudCode? - if yes, how? - if no, then how to use FlexMatch in iOS/Android?

We are in the process of testing GameLift for our game. We have created an Amazon Linux 2 VM under Virtual Box and have a working sample server form Github working and also have GameLiftLocal running on the machine. We run Gamelift running the following command: ``` [ec2-user@vm_gamelift3 GameLiftLocal-1.0.5]$ java -jar GameLiftLocal.jar ``` This seems to result in a successful startup: ``` 05:22:54,823 INFO || - [SocketIOServer] main - Session store / pubsub factory used: MemoryStoreFactory (local session store only) 05:22:59,344 WARN || - [ServerBootstrap] main - Unknown channel option 'SO_LINGER' for channel '[id: 0x88185e57]' 05:22:59,353 INFO || - [SDKConnection] main - GameLift SDK server (communicates with your game server) has started on http://localhost:5757 05:22:59,390 INFO || - [SocketIOServer] nioEventLoopGroup-2-1 - SocketIO server started at port: 5757 05:22:59,450 INFO || - [SdkWebSocketServer] WebSocketSelector-13 - WebSocket Server started on address localhost/127.0.0.1:5759 05:22:59,655 INFO || - [StandAloneServer] main - GameLift Client server (listens for GameLift client APIs) has started on http://localhost:8080 05:22:59,705 INFO || - [StandAloneServer] main - GameLift server sdk http listener has started on http://localhost:5758 ``` We also get a good connection when starting up the game server sample locally. However, when we attempt to issue the local server a command, it always comes back with an exception ``` [ec2-user@vm_gamelift3 ~]$ aws gamelift create-game-session --endpoint-url http://127.0.0.1:5758 --maximum-player-session-count 2 --fleet-id fleet-123 ``` ``` An error occurred (NullPointerException) when calling the CreateGameSession operation (reached max retries: 4): ``` The server does not completely crash, but does give a warning and a call stack of the exception that was caught ``` 05:25:31,497 WARN || - [GameLiftSdkHttpHandler] GameLiftSdkHttpHandler-thread-0 - Caught exception java.lang.NullPointerException at com.amazon.gamelift.handlers.GameLiftSdkHttpHandler.validateHeaders(GameLiftSdkHttpHandler.java:84) at com.amazon.gamelift.handlers.GameLiftSdkHttpHandler.handle(GameLiftSdkHttpHandler.java:55) at com.sun.net.httpserver.Filter$Chain.doFilter(Filter.java:79) at sun.net.httpserver.AuthFilter.doFilter(AuthFilter.java:83) at com.sun.net.httpserver.Filter$Chain.doFilter(Filter.java:82) at sun.net.httpserver.ServerImpl$Exchange$LinkHandler.handle(ServerImpl.java:700) at com.sun.net.httpserver.Filter$Chain.doFilter(Filter.java:79) at sun.net.httpserver.ServerImpl$Exchange.run(ServerImpl.java:672) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) at java.lang.Thread.run(Thread.java:748) 05:25:31,554 WARN || - [GameLiftSdkHttpHandler] GameLiftSdkHttpHandler-thread-0 - Caught exception null ``` This exception also occurs when we just try to describe the sessions, so it seems to be any command you issue the jar on that socket. How do I resolve this so that I can get to actually testing my code? Thank you! David

I'm attempting to use GameLift Custom Server FlexMatch MatchMaking Configuration with a game session placement queue. The Matchmaking Event Handler Lambda first receives a MatchmakingSearching event, then a PotentialMatchCreated event. The Game Session Queue Placements Started event shows up in the Queue metrics section, but it will then timeout (after 5 minutes). The GameLift Game Session is being created, but the instance process' OnStartGameSession(GameSession gameSession) is NOT being called.. Can someone please help me figure out how to get the OnStartGameSession to be called in the fleet instance process? Before attempting to use FlexMatch, I was able to start the placement manually and the game session was being created on the game server, and the game was playable, but I'm unable to get to that step using the automated FlexMatch matchmaking system. Thank you.

from [this image](https://drive.google.com/file/d/1bdWuDc0BITyey8AZKKidYbgjHQsudlhT/view?usp=sharing), we can see that the game session, 53E1A6ED-EB5B-418A-89E8-D773682B70AB in the fleet fleet-76e80f57-a589-444f-ba49-664672c2f9e6 didn't call process ending before terminate the process. But when I check the logs in auxproxy.log in /local/whitewater/Logs/, [here](https://drive.google.com/file/d/1FNdQBaqWtTGv75phG0z6cXsocyA4MgX4/view?usp=sharing), we can see the game session 53E1A6ED-EB5B-418A-89E8-D773682B70AB is the process fb528fcf-ac19-45d0-9750-a28bebf4da11 in the fleet instance. But from [here](https://drive.google.com/file/d/1tBr5YfWx26eGc8I-2jbPs3GXwRqrQZx2/view?usp=sharing), we do see the it has received the process ending message for process fb528fcf-ac19-45d0-9750-a28bebf4da11. And the process is closed hundreds milliseconds after "Marking process fb528fcf-ac19-45d0-9750-a28bebf4da11 as unavailable for game sessions" log is printed. it only happens on UE5.0.1 release. Can you tell me how it determine the process closed without calling process ending? is there any other condition except process ending is called or not? thanks.

I'm not post this for this account. But if I post with newly create aws account, you guys will think I'm a beginner and ask me to increase service limit and tell me my resource usage is minimal. I have reported this issue many times since two or three months ago via email, support case and feedback, but you guys didn't realize it's actually a bug. **So please, supporter, please don't use your manual book when you see this post, just report it to the real developer team.** All newly created aws account's gamelift service limit will return 0 now in all regions and all instance types, which should be anything other than 0 (1 or 20, normally). which make all new account can't even create its first fleet. You can ask the developer team to create a new account to try and confirm it yourselves, it's very easy to reproduce.

Hi, I am wondering if there is any way that a game server can activate itself without receiving the `onStartGameSession` event? The scenario I am looking at is roughly like this: I am using a 3rd part matchmaking library; This library allows the game server starts a session and wait for the players. I am planning to make the stand-by game servers create their sessions and wait until a player connects to them; At this point I want to make a call to a GameLift api like `ActivateGameSession` to tell the GameLift the server is activated so that it adds another game server to stand-by pool. The problem I have here is that in the documentation it is mentioned that the `ActivateGameSession` should be called in response to the `onStartGameSession` event. Is the scenario I described possible to implement with the GameLift? If it is, how could I do it? Are there any caveats? Look forward to hearing from you. Regards, Aydin

Hi, --- When I attempted to create my first fleet, I got this error: "Error! Cannot create the fleet. Reason: Current limit of fleets of 0 have been reached.." I tried several ways to solve it, but failed: Change regions or reload builds doesn't work. Change Quotas of fleet doesn't work. --- Anyone can help?

When trying to create a new gamelift fleet, the following error occurs: "An error occurred (LimitExceededException) when calling the CreateFleet operation (reached max retries: 2): Current limit of fleets of 0 have been reached." This happens independently of instance type. On the service quota dashboard, the limit is 20. Does anyone have any idea on what could be causing this issue? We have been using gamelift for over a year and this is the first time anything like this has occurred.

**Hello!** I am trying to deploy the Gamelift test suite, but am getting the following error: **Current limit of fleets of 0 have been reached..** We are a video game development company and we want to deploy our games on your infrastructure. Please get in touch with us. If there is someone who can give us support in **Spanish**, it would be the best option. **Thanks.**

Hello! Whenever I attempt to create a new fleet, I get this error: `Error! Cannot create the fleet. Reason: Current limit of fleets of 0 have been reached..` Was working fine last week, and I have no fleets currently running. Any ideas?

hi. I tried to make freet, “Current limit of fleets of 0 have been reached…” I keep getting an error message. Currently, the number of fleets generated is zero. In the past, there was no problem with generating fleets several times. Is there a if process required to delete it? It's been more than 3 days since the problem lasted. Can you solve this problem on your own? thanks

I have checked with support and confirmed this is not a payment issue. Yesterday, I was able to create new fleets in AWS Gamelift without a problem. In the past, I have gotten a similar error once I have 5 fleets, "Current limit of fleets of 5 have been reached.." Currently I have 2 active fleets, and am unable to make a third.

Hello! I've started getting this error when calling aws gamelift create-game-session for GameLiftLocal `"ERROR || - [GameLiftHttpHandler] Thread-36 - No available process."` This is the full command I'm using. Previously this hasn't given me any issues: `aws gamelift create-game-session --endpoint-url http://localhost:9080 --maximum-player-session 8 --fleet-id fleet123` Any ideas why this would suddenly give me issues? Have tried restarting my PC multiple times.

I can remote into my server, see that the logs exist there, but when I try to use the CLI Console to download the logs at locations /local/game/GameName/Saved/Logs/GameName.log it tells me it cannot be found. Any ideas?

In addition to my earlier issue (see AWS Gamelift Fleet Instance Capacity not updating on terminated fleet), I'm now having an issue where I have zero builds visible, and if I upload a build it does not appear. Any ideas?

I'm testing build compatibility with the free tier for Gamelift. Deleted a fleet and am now receiving a too many instances error when I try to build another one. My fleet console shows empty, but the CLI Describe-fleet-capacity shows me as having an active fleet nor active instances.

Hey! My server listens on 2 ports and uses port 7777 for client connection and port 7787 for beacon connection. When starting a game session, if any of these ports are busy, IpNetDriver increments the port numbers and starts listening. This could be, for example, "7778 and 7788" or "7780 and 7791". There are no dependencies. The beacon client can only connect to the expected beacon port, but it changes every new game session. Is there a way to get the gaming session beacon port together with the server's IP and port as a search result?

I have a Java Client I am attempting to integrate with GameLift. I am able to use it to search for Active Fleets, Search Game Sessions, Create Game Sessions and Create Player Sessions. I do not know where to go from there. What API or Java method is used to communicate with the Server. I have the example Server setup and running. The example server code is shown here: https://docs.aws.amazon.com/gamelift/latest/developerguide/realtime-script.html The Java code I have written: //////////////////////////////////////// AmazonGameLift gameLift = AmazonGameLiftClientBuilder.standard() .withCredentials(new AWSStaticCredentialsProvider(credentials)) .withRegion("us-east-2") .build(); DescribeFleetAttributesRequest dfar = new DescribeFleetAttributesRequest(); DescribeFleetAttributesResult describeFleetResult = gameLift.describeFleetAttributes(dfar); CreateGameSessionRequest cgsRequest = new CreateGameSessionRequest().withFleetId(describeFleetResult.getFleetAttributes().get(0).getFleetId()) .withName("Sim Game") .withMaximumPlayerSessionCount(MAX_PLAYERS); CreateGameSessionResult cgsResult = this.gameLift.createGameSession(cgsRequest); byte[] array = new byte[[PLAYER_ID_LENGTH]]; // length is bounded by PLAYER_ID_LENGTH new Random().nextBytes(array); String playerID = new String(array, Charset.forName("UTF-8")); CreatePlayerSessionRequest cpsRequest = new CreatePlayerSessionRequest().withGameSessionId(cgsResult.getGameSession().getGameSessionId()) .withPlayerId(playerID); CreatePlayerSessionResult cpsResult = this.gameLift.createPlayerSession(cpsRequest); //////////////////////////////////////////////////////////////////////////////////////// I am assuming there is a way to communicate using to the Server using some call to cpsResult. My theory is to use cpsResult.getPlayerSession().marshal() But I do not understand how to use the marshall function. void marshall(ProtocolMarshaller protocolMarshaller) I am also not sure that is even the correct function to use. If anyone can point me in the right direction to start attempting Server communication I would be very grateful.

The GameLift Reference guide shows information on actions available using the C# SDK. It shows methods for "NewMessage" and "SendMessage". I am building my client with Java, and I can't seem to find any equivalent function in Java to send a message to the Server from the Client. Does anyone know what the equivalent methods in Java are for sending messages to a RealTime server? Thank you. https://docs.aws.amazon.com/gamelift/latest/developerguide/realtime-sdk-csharp-ref-actions.html

Hello!) I have an issue with client/server connection. I created a game session, reserved a player session, got the URL for a client. But when the UE4 client calls ClientTravel("ip+port+?+PlayerSessionId=...), the server's game mode doesn't fire either Login() or PreLogin() functions. Because of that, I can't get a PlayerSessionId from the "Options" argument and validate the PlayerSesssion on the server and Gamelift closes it after 1 minute. When I run a server locally without Gamelift, "Login()" fires as expected. Gamelift health check works great in the server's game mode as well. Please help!

Is it possible to get a copy of the .proto files used by the Realtime client SDKs? I want to port the SDK to JVM, but I am having trouble decompiling the protobuf generated code from the C# SDK.

Steps to reproduce: 1. Open gamelift dashboard in us-west-1 2. Click "create fleet" 3. Populate the fleet creation fields 4. Click initialize fleet Expected: The UI loads briefly then shows provisioning details such as logs. The fleet should be visible in the list of fleets, perhaps in a provisioning state. Actual: The fleet creation UI closes and is replaced with the amazon loading icon (two arrows that form a circle, spinning). This has persisted for over an hour. I've tried different fleet details such as spot vs on-demand to no avail. There are no errors/logs present to suggest a failure. Any ideas? Thanks!

My title says it all. There is no action in this forum, so where can i get any support on my bugs?

I am using the Unreal Engine Gamelift Server SDK Plugin. After packaging a build, I can test it locally with the Gamelift-local jar and I can create a game session and join it just fine. When I upload the build to gamelift however, the fleet validation process fails after successfully finding the runtime path. None of the logs indicate an error, and I am left grasping at straws as to what the problem might be here. Is there any way I can retrieve additional information as to what caused it to fail? Any help is greatly appreciated.

I'm getting "Server process exceeded 3 minutes without reporting healthy" fairly regularly from my gamelift event log, but as far as I can tell it is responding "true" in a timely fashion to every healthcheck request received. I'm running 20 processes on a single c4.large, and the server has plenty of free memory and cpu (generally only a few of the processes actually processing game data at any given time). Load on the server is very low. Yet processes are getting killed by gamelift as unhealthy at a pretty steady clip... I'm not entirely sure how long each process lives or what causes it to suddenly start failing health checks. The Longest lived of my processes is about 4 hours while some are only a few minutes old.