MeshMonitor v4.17.1-rc2

Summary

This release candidate is mostly about MQTT. Two bridges attached to one embedded broker no longer relay each other's upstream traffic. Before, upstream A's packets were published to upstream B, and two upstreams that already shared traffic could pass a packet back and forth without end. A new per-bridge option, Relay traffic from other bridges (off by default), keeps that relaying for anyone who wants it. A bridge also no longer sends its own downlinked packet back upstream. The broker's Raise hop count for backhaul now works on encrypted uplinks from a phone's MQTT client proxy, so a "raise text messages" policy finally matches a phone's text. Don't uplink MeshMonitor automation traffic now also covers a node linked straight to a bridge. A new MQTT Setup Guide walks through the common broker and bridge layouts, and the MQTT settings pages link to it. MeshCore remote admin now remembers admin logins, so the console can resume a session with nothing sent over the air, and scheduled clock syncs drop from 4 packets to 2. Auto Acknowledge channel replies now come first from MeshMonitor instances nearer the sender.

Features

• MeshCore: remember and resume remote-admin sessions (#5798, closes #5785). Admin logins are stored per source and survive a restart. The console offers Resume session and Log out, and scheduled jobs reuse the session, with one fresh login and one retry if the repeater has dropped it.
• Auto Acknowledge: nearer nodes answer a channel message first (#5789, closes #5788)

Bug Fixes

• MQTT: bridges stop relaying each other's upstream; close the echo race (#5799, refs #5013)
• MQTT: raise encrypted uplinks by their real portnum (#5795, closes #5791)
• MQTT: drop automation uplinks on the client-proxy path too (#5787, refs #5414)
• Docker: the BLE bridge health check follows the BLE link, not the TCP port (#5802). It works with the bridge's new HEALTH_FILE (meshtastic-ble-bridge#20) in the Docker Configurator and docker-compose.ble.yml.

Documentation

• MQTT Setup Guide, linked from the app (#5800, refs #5013)

CI/DevOps

• Version bump to 4.17.1-rc2 (#5801)

Issues Resolved

• #5785 MeshCore: Resume Admin Session
• #5788 Hop-weighted bias on Auto-Ack channel-reply delay
• #5791 Raise hop limit for text messages does not work on MQTT source

Upgrade Notes

• Several bridges on one broker: if you relied on one upstream being relayed into another through MeshMonitor, tick Relay traffic from other bridges on the bridge that should carry it. Everything that starts locally is uplinked as before.
• BLE bridge users: pull the latest meshtastic-ble-bridge image and add HEALTH_FILE=/tmp/ble-bridge.health plus the new health check (or regenerate your compose file with the Docker Configurator). MeshMonitor now depends on the bridge with service_started, so it starts even while the node is out of range.
• Migration 204 adds the meshcore_admin_sessions table on SQLite, PostgreSQL and MySQL.

Full changelog: v4.17.1-rc1...v4.17.1-rc2

🚀 MeshMonitor v4.17.1-rc2

📦 Installation

Docker (recommended):

docker run -d \
  --name meshmonitor \
  -p 8080:3001 \
  -v meshmonitor-data:/data \
  ghcr.io/Yeraze/meshmonitor:4.17.1-rc2

🧪 Testing

✅ All tests passed
✅ TypeScript checks passed
✅ Docker images built for linux/amd64, linux/arm64, linux/arm/v7

📋 Changes

See commit history for detailed changes.