- Fixed /voicesentinel unvcmute and /voicesentinel vcmute freezing the server for ~30 seconds on Folia in some situations
- Better player name matching for Bedrock/Floodgate names (e.g. .PlayerName)
- Much lower CPU use from WebSocket reconnects and processor connection retries
- Voice packet processing moved off the main server thread
- Audio uploads no longer block the server tick loop
- Web dashboard + network sync no longer hit Redis on the main thread, fixes dashboard-related lag spikes on multi-server setups
- Custom profanity/mute commands run more reliably after auto-mutes finish applying
- Updated custom command configuration
- Updated docs online
- Misc performance fixes
- Misc backend fixes
V2.3.0 has been released with the following features:
- Multi server support (using redis)
- Bug fixes
- Configuration changes
These changes will be outlined tomorrow in the documentation. For now, you can open a ticket.
- Added leaf compatibility
- Fixed issues with large dependencies
- Fixed issues with session management
- Fixed issues with vcmute commands running on the wrong thread for non folia servers
- Fixed important performance issues for a large playerbase
- Fixed small security issues with web dash
- Fixed issue that could cause processor to crash
- Fixed some report system issues
- Created a config migration so nobody should have to touch their plugin config anymore
Bug fixes and new configuration options:
YAML:custom_commands: mute_commands: [] profanity_commands: [] auto_mute_builtin: luckperms_voice_mute: true player_message: true staff_alert: true discord_webhook: true mute_event_log: true console_log: true
Bug Fixes:
New Features:
- Fixed Folia issues
- Updated docs
- Added legal notice to public processors: Click Here
- Slight modifications to entire codebase for stability
- Updates to license system
- Web Dashboard, added a UI to monitor, review and moderate your server. See here
- Voice Report system, allow users (if configured) to report other users. Administrators will get a notification on server and Web Dashboard if they have permission. See here
- Transcripts are stored in server memory for a configured amount of time (buffer system, see link right above this bullet point)
Processor has been updated to support latest plugin features. Public processors now include audio recordings in responses.
NOTE:
- Heavy configuration changes, BACKUP your configs before updating and fill in the values, as there are a lot of new options
- Processor config has also been slightly updated if you self host, if you want to use the Voice Report system, it is recommended you reset your processor config.json
- This version may have issues, please open a ticket and it will be fixed ASAP
Docs: https://pandascript.gitbook.io/pandascript
