When playing superbugs a dialogue box comes up which states 

you have lost connection to the Internet, online game features will no longer be available.  

Any nodes you have completed within the game are lost. All connections to the Internet are fine, no loss of connection. To rectify, game has to be closed and restarted.  It appears to be linked to the game and steam. It only started happening this week. I am worried this, will Impact on museum.  Could someone please let me know if this is something you already know about, or, what can be done to investigate this issue.  

Hi, Rosie.

I've been casually looking at log files for the issues affecting the superbug network disconnections and sync failures. After collecting and reading several log files over a few play sessions with verbose entries enabled, I'd like to share a breakdown of what's happening under the hood and why this may be occurring. Hopefully this can assist the developers and help others experiencing similar issues understand whats happening.

Just how many requests are we talking about here?

From a single verbose log file 13,224 PlayFab requests were sent in just 21 minutes. That’s an average of over 10 requests per second. To put it in perspective, each time you load or update superbug progress, your game might send 2–3 requests per superbug node. If you have 15 active nodes, that’s 30–45 rapid requests. Then it asks for user info and leaderboard positions for (in my case) dozens of players (seems like everyone on my Steam friend list including those that don't own the game) and it does this separately, not in groups.

What happens when this happens?

​Once PlayFab starts refusing the requests player info doesn’t load, leaderboards don’t update. The superbug network can’t keep up and eventually, your game gives up and disconnects you from the collaboration network. Sometimes it looks like nothing happened, other times it completely stops your Superbug progress.

​Requests to look up players like GetPlayFabIDsFromGenericIDs fail repeatedly, even though the IDs are valid. The game tries again and again without any pause making things worse. Many of these requests could likely have been combined into one big request, but aren’t. These failures aren’t because your game is broken it’s because it’s being too enthusiastic talking to the server.

What can be done?

Slow it down a bit? Perhaps add short delays (like 0.1 seconds) between these server calls.
Group things together? Instead of asking for 50 users one by one, ask for all 50 at once.
Reuse info? If a leaderboard hasn’t changed, don’t ask again right away.
Show real error messages. If PlayFab says 'you’re sending too much' the game should log that clearly.

​If your superbug keeps disconnecting it’s not your fault. it’s not your internet, it’s a server-side issue caused by too many requests going out too fast. Avoid clicking on too many superbug tasks all at once, and wait for a patch from the developers.

- A concerned hospital administrator.

--------------------------------------------

The game performs massive bursts of PlayFab API calls, especially GetLeaderboardAroundPlayer and GetPlayerStatistics, for multiple SuperBug nodes in rapid succession. This pattern repeats almost continuously, dozens of requests made within milliseconds of each other and sometimes in the same millisecond or frame.

The log does not indicate any intentional delay between requests, backoff strategy, or retry-after handling. This likely violates PlayFab’s API rate limits, especially for free or non-tiered usage plans.

Missing Rate Limit or 429 Error Codes. While the logs are very verbose, no explicit "HTTP 429 Too Many Requests" or rate limit errors are directly shown in the snippet. However, the game issues multiple repeated GetPlayFabIDsFromGenericIDs calls for the same Steam IDs. This is a common trigger for rate limiting. Warnings like:

[Playfab] Failed to find GenericID for user : 7656119918598XXXX

could indicate a failure to retrieve expected data, potentially from throttling. Likely root cause? Rate Limiting from PlayFab is being triggered because the game floods the API with back-to-back calls to GetLeaderboardAroundPlayer, GetPlayerStatistics, GetPlayFabIDsFromGenericIDs and GetUserData for dozens of superbug nodes without any pacing. Often executed dozens of times per second (e.g., 2–3 per superbug node, sequentially for 15+ nodes). There is no evidence of request throttling, batching, or backoff mechanisms between them. Perhaps on failure or missing data, wait and retry with exponential backoff instead of retrying instantly. Also check rate limit headers? If PlayFab returns rate limit headers (Retry-After, X-RateLimit-Remaining), the game should honor them.

All in all: The superbug client implementation is issuing requests too fast for PlayFab’s default rate limits. These spikes include unbatched ID resolution and leaderboard fetches per node. PlayFab begins silently throttling (dropping or denying responses), which causes data holes and logic failures downstream in superbug collaboration. Eventually, the client likely considers the data ;unavailable' and disconnects the superbug session.

Despite this being a rather in depth explanation and not wholly understandable by the average player on the technical side of things, the instructions on how to make playing a superbug without putting pressure on a system which clearly is not geared up for growth in users, I find it impossible for either myself, or other players to second guess what keystrokes are being used and in what quick succession.  the whole process of playing the game, especially when playing a a group on one superbug, is to complete a node and move on. there is no communication between the players apart from the messaging system on steam, and it would be impossible for each player to communicate to others asking them to pause before they move onto another node in order to not overload the system.  The system today on 20 May 2025, would not enable me to even load the superbug, let alone play a node, and having closed the game twice, and re loaded, I gave up.

Just to follow up for anyone curious about the technical side, after switching log types I did end up seeing the 429 timeout error (a lot):

2025-05-20 19:50:47.634|7389|INFO||[PlayFab Error] error: APIClientRequestRateLimitExceeded http Code: 429 httpStatus: TooManyRequests

2025-05-20 19:50:47.634|7389|ERRR||[PlayFab] GetUserData Error Got APIClientRequestRateLimitExceeded