MeshMonitor v4.17.2-rc7

⚠️ Pre-release. This is a release candidate for testing. Do not run it against production systems.

This RC bundles the first batch of issue fixes merged after rc6 — ten PRs across MeshCore, telemetry, aircraft detection, and docs. MeshCore gains a scheduled-telemetry retry burst with cooldown (per-source settings, with the same policy applied to the Meshtastic scheduler), a Channels landing pane with click-to-deselect, a "purge non-favourite nodes" action in device config, aligned node-list header controls, and own outbound channel reactions that now fold into chips instead of showing as raw text. Virtual-node default-key MQTT-proxy traffic now lands on Primary #0 (plus a new purge button for virtual channels). Replayed firmware-2.8 NodeDB telemetry and position frames no longer fire false reboot, power-change, or became-likely-aircraft events. A manual aircraft mark can now apply across every source that sees a node (opt-in). And the archived-MeshManager references were removed from the docs and the script gallery. No security changes here — those shipped in rc6.

Features

• MeshCore scheduled telemetry: retry burst (90s × 5) then a 30-min cooldown after a missed/timed-out pull, exposed as per-source settings and applied to the Meshtastic scheduler too; retry state is DB-persisted so a restart or settings save never re-arms it — #6113 (#6033)
• MeshCore Channels: no channel auto-selected on open, a "Select a channel" landing pane, and click-the-selected-channel to deselect (desktop) — #6112 (#6049)
• Optional all-sources manual aircraft mark — apply aircraft / not_aircraft / clear to every source that sees the node, skipping sources you can't write to — #6111 (#6046)
• Purge non-favourite MeshCore nodes from the device configuration tab — #6110 (#6041)

Bug Fixes

• Fold own outbound MeshCore channel reactions into chips instead of raw text bubbles (receive-side parser now accepts the space-separated hash form) — #6108 (#6039)
• Skip reboot / power-change / became-likely-aircraft detectors on replayed telemetry and position frames — #6107 (#6045, #6099)
• Match default-key primary MQTT-proxy traffic to Primary #0 via the modem-preset name instead of splitting it onto a virtual "LongFast #101" channel — #6109 (#6044)
• Add a "Purge all messages" button to the virtual channel info modal — #6106 (#6044)
• Align the MeshCore node-list header controls into a clean toolbar grid — #6105 (#6040)

Docs

• Remove user-facing MeshManager references and the stale solar-forecast gallery script (MeshMonitor covers it natively) — #6104 (#6048)

Issues Resolved

Closes #6033, #6039, #6040, #6041, #6044, #6045, #6046, #6048, #6049, #6099.

Upgrade notes

• No breaking changes. The new per-source telemetry-retry settings default to 90s / 5 attempts / 30-min cooldown.
• Purging non-favourite MeshCore nodes is admin / nodes:write-gated and only removes stored nodes from MeshMonitor — the radio's own contact list is not touched.
• The all-sources aircraft mark is opt-in (a checkbox on the mark menu); the default stays single-source.

Full changelog: v4.17.2-rc6...v4.17.2-rc7

🚀 MeshMonitor v4.17.2-rc7

📦 Installation

Docker (recommended):

docker run -d \
  --name meshmonitor \
  -p 8080:3001 \
  -v meshmonitor-data:/data \
  ghcr.io/Yeraze/meshmonitor:4.17.2-rc7

🧪 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.