GameCult/CultLib
Watch
9
0
Fork
You've already forked CultLib
 0
Code Issues Pull requests 3 Projects Releases Packages Wiki Activity Actions 4

GameCult.Networking: answer an undecodable RUDP frame message_invalid and end the session at 8 in a row #46

Merged
libby merged 1 commit from libby/cs-refused-frames-cap into main 2026-11-08 09:47:22 +00:00
Conversation 0 Commits 1 Files changed 2 +82 -17
libby commented 2026-11-08 09:47:09 +00:00
Member
Copy link

Heimdall's parity ruling (2026-11-08), C# side; Rust #43 (4828c416), TS #44, Python #45. C# is the odd one out: RudpCultNetSchemaServer ended the session on the first undecodable frame. It now answers message_invalid (details.field = "message") and keeps the session; RudpCultNetSchemaServer.MaxConsecutiveRefusedFrames = 8 consecutive undecodable frames end it after the 8th is answered. Any decoded frame resets the run. Handler faults still end the session at once (unchanged).

No write-authority decision is touched: the change is in the transport dispatch, below the database server. Not done here: the refused-delete half of the ruling. CultNetDatabaseServer.HandleDeleteAsync answers write_not_authorized inside the handler, where the dispatcher cannot count it or end a session; that needs a handler-to-dispatch signal (follow-up filed).

Tests: three changed (a bad frame is now answered and kept; a frame queued behind it is handled; the poll-input class doc) and EightConsecutiveUndecodableFramesEndTheSessionAfterTheEighthIsAnswered (cap pinned at 8, 6 more answered, a valid frame resets, 7 more answered, 8th answered then ended). Four hand mutants fail it (cap 7, cap 9, run never resets, run never counts). Mutants ran under prlimit --cpu=500 and DOTNET_GCHeapHardLimit=4GiB; prlimit --as was not used because the .NET runtime reserves far more address space than it uses. GameCult.Networking.Tests 505 pass, GameCult.Mesh.Tests 342 pass.

🤖 Generated with Claude Code

Heimdall's parity ruling (2026-11-08), C# side; Rust #43 (4828c416), TS #44, Python #45. C# is the odd one out: `RudpCultNetSchemaServer` **ended the session on the first undecodable frame**. It now answers `message_invalid` (`details.field = "message"`) and keeps the session; `RudpCultNetSchemaServer.MaxConsecutiveRefusedFrames = 8` consecutive undecodable frames end it after the 8th is answered. Any decoded frame resets the run. Handler faults still end the session at once (unchanged). No write-authority decision is touched: the change is in the transport dispatch, below the database server. **Not done here:** the refused-delete half of the ruling. `CultNetDatabaseServer.HandleDeleteAsync` answers `write_not_authorized` inside the handler, where the dispatcher cannot count it or end a session; that needs a handler-to-dispatch signal (follow-up filed). Tests: three changed (a bad frame is now answered and kept; a frame queued behind it is handled; the poll-input class doc) and `EightConsecutiveUndecodableFramesEndTheSessionAfterTheEighthIsAnswered` (cap pinned at 8, 6 more answered, a valid frame resets, 7 more answered, 8th answered then ended). Four hand mutants fail it (cap 7, cap 9, run never resets, run never counts). Mutants ran under `prlimit --cpu=500` and `DOTNET_GCHeapHardLimit=4GiB`; `prlimit --as` was not used because the .NET runtime reserves far more address space than it uses. GameCult.Networking.Tests 505 pass, GameCult.Mesh.Tests 342 pass. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
GameCult.Networking: answer an undecodable RUDP frame message_invalid and end the session at 8 in a row
Some checks failed
CultMesh QUIC native / Build win32-x64 (pull_request) Has been skipped
Details
CultMesh QUIC native / Build linux-x64 (pull_request) Has been skipped
Details
CultMesh QUIC native / Release native binaries (pull_request) Has been skipped
Details
CultMesh portability and performance / ubuntu-latest reactive semantics and leases (pull_request) Failing after 7m10s
Details
CultMesh QUIC native / Verify committed binaries match SHA256SUMS (pull_request) Successful in 1m20s
Details
89bbb277bd 
Heimdall's parity ruling. The schema server ended a peer's session on the first
frame it could not decode. It now answers message_invalid (field "message") and
keeps the session; RudpCultNetSchemaServer.MaxConsecutiveRefusedFrames (8)
consecutive undecodable frames end it after the last is answered, and any
decoded frame resets the run. Handler faults still end the session at once.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
libby merged commit 51d2d0cf46 into main 2026-11-08 09:47:22 +00:00
Sign in to join this conversation.
Reviewers
No reviewers
Labels
Clear labels
No items
No labels
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
GameCult/CultLib!46
No description provided.