Fix a few issues in UDP connection logic #960

New Issue

MrUnknownDE · 2026-04-06T01:37:06+02:00

MrUnknownDE commented

2026-04-06 01:37:06 +02:00

Originally created by @abb128 on 9/1/2023

Makes the following changes:

Stores connections by SocketAddress (ip:port) instead of InetAddress (ip only)
- This behaves properly when a client connects, disconnects, then tries to connect with a different port but same IP (if the udp port is randomly assigned for example)
- Previously re-connecting to the server would work if the client always bound to the same UDP port (as the trackers do: 5fe244423a/src/network/connection.cpp (L608) )
- If the port differed but IP remained the same, setUpNewConnection would see the same InetAddress and re-use the existing UDPDevice without updating the old socketAddress (which still has the old port!)
- With this change the SocketAddress differs so a new UDPDevice is made. If the MAC hasn't changed then the previous connection will be found by MAC and updated by the already existing logic
- This also now allows running multiple clients from the same IP address (as long as the reported MACs differ)
Uses the correct UDPDevice later in the case when a previous connection with same MAC is found and reused
- Previously setUpSensor(connection, 0, handshake.imuType, 1) would be called on the newly created UDPDevice, which never ends up being used because previousConnection is updated instead, so each new connection would add a phantom tracker for a device that's never used (this only would happen if firmwareBuild < 9 and the address differed)
- I also moved VRServer.instance.deviceManager.addDevice(connection) to only the branch when the new UDPDevice is added, although I'm not sure what this is supposed to do, I couldn't find any uses of deviceManager's device list

You can reproduce the bugs on the original server using https://gist.github.com/abb128/15c31fcdda6388818276710d52b08406 , and with this PR the bugs disappear

*Originally created by @abb128 on 9/1/2023* Makes the following changes: * Stores connections by SocketAddress (ip:port) instead of InetAddress (ip only) * * This behaves properly when a client connects, disconnects, then tries to connect with a different port but same IP (if the udp port is randomly assigned for example) * * Previously re-connecting to the server would work if the client always bound to the same UDP port (as the trackers do: https://github.com/SlimeVR/SlimeVR-Tracker-ESP/blob/5fe244423a52ae21d1fc15e4bc057166ff73e91d/src/network/connection.cpp#L608 ) * * If the port differed but IP remained the same, setUpNewConnection would see the same InetAddress and re-use the existing UDPDevice without updating the old socketAddress (which still has the old port!) * * With this change the SocketAddress differs so a new UDPDevice is made. If the MAC hasn't changed then the previous connection will be found by MAC and updated by the already existing logic * * This also now allows running multiple clients from the same IP address (as long as the reported MACs differ) * Uses the correct UDPDevice later in the case when a previous connection with same MAC is found and reused * * Previously `setUpSensor(connection, 0, handshake.imuType, 1)` would be called on the newly created UDPDevice, which never ends up being used because previousConnection is updated instead, so each new connection would add a phantom tracker for a device that's never used (this only would happen if firmwareBuild < 9 and the address differed) * * I also moved `VRServer.instance.deviceManager.addDevice(connection)` to only the branch when the new UDPDevice is added, although I'm not sure what this is supposed to do, I couldn't find any uses of deviceManager's device list You can reproduce the bugs on the original server using https://gist.github.com/abb128/15c31fcdda6388818276710d52b08406 , and with this PR the bugs disappear

MrUnknownDE closed this issue

2026-04-06 01:37:06 +02:00

Sign in to join this conversation.

Branches Tags

main

sapphire/diy-fw-tool-version-sort

aed/grammar-police

battery-0

bscotch/server-rewrite-temp

hannah/steam

pontoon

hannah/keybinds

server-rewrite

bscotch/accel-mount

dependabot/github_actions/gradle/actions-6

dependabot/github_actions/pnpm/action-setup-5

dependabot/github_actions/actions/upload-artifact-7

dependabot/github_actions/actions/download-artifact-8

FloorClipSteamLinkInteractionFix

sapphire/trim-vrm-json

llelievr/trackers-udp-tests

sapphire/linux-bridge-kotlin

sapphire/linux-vrchat-compatdata-path

smol-sleep-state

sapphire/stay-aligned-setup-no-step-skip

sapphire/steamvr-hands-warning

sapphire/proportions-import-skeleton-request

sapphire/bone-mask-update

llelievr/settings-profiles

bscotch/proguard

bscotch/bvh-fix

mitten

bscotch/reset-proof-tests

hannah/install-from-server

smol-buttons

TheButlah-patch-1

sapphire/update-solarxr

sapphire/kotlin-flatbuffers

csv-data-export

feat_flashmode

bscotch/remove-drift-compensation

llelievr/kotlin-config

bscotch/fix-osc-duplication

llelievr/keep-battery-value

rest-calibration-support

tracker-graph

fingertracking

update-eslint

tauri-android

hide-new-resets

height-config

snap-serial

feat/new-updater

support-inches

vrc-fix-ik-key

llelievr/3dmodel-preview

accel-mapping-fixes

llelievr/manual-proportions-rework

llelievr/put-automatic-steamvr-tracker-assigment-first

llelievr/remove-manual-mounting-onboarding

onboarding-usage-step

oldflex

motion-compensation

kotlin-warnings-fix

accessories

tpose

foxvr

No Label

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github/SlimeVR-Server#960