Files
meshcore-analyzer/cmd/server/repeater_liveness.go
T
Kpa-clawbot 45f30fcadc feat(repeater): liveness detection — distinguish actively relaying from advert-only (#662) (#1073)
## Summary

Implements repeater liveness detection per #662 — distinguishes a
repeater that is **actively relaying traffic** from one that is **alive
but idle** (only sending its own adverts).

## Approach

The backend already maintains a `byPathHop` index keyed by lowercase
hop/pubkey for every transmission. Decode-window writes also key it by
**resolved pubkey** for relay hops. We just weren't surfacing it.

`GetRepeaterRelayInfo(pubkey, windowHours)`:
- Reads `byPathHop[pubkey]`.
- Skips packets whose `payload_type == 4` (advert) — a self-advert
proves liveness, not relaying.
- Returns the most recent `FirstSeen` as `lastRelayed`, plus
`relayActive` (within window) and the `windowHours` actually used.

## Three states (per issue)

| State | Indicator | Condition |
|---|---|---|
| 🟢 Relaying | green | `last_relayed` within `relayActiveHours` |
| 🟡 Alive (idle) | yellow | repeater is in the DB but
`relay_active=false` (no recent path-hop appearance, or none ever) |
|  Stale | existing | falls out of the existing `getNodeStatus` logic |

## API

- `GET /api/nodes` — repeater/room rows now include `last_relayed`
(omitted if never observed) and `relay_active`.
- `GET /api/nodes/{pubkey}` — same fields plus `relay_window_hours`.

## Config

New optional field under `healthThresholds`:

```json
"healthThresholds": {
  ...,
  "relayActiveHours": 24
}
```

Default 24h. Documented in `config.example.json`.

## Frontend

Node detail page gains a **Last Relayed** row for repeaters/rooms with
the 🟢/🟡 state badge. Tooltip explains the distinction from "Last Heard".

## TDD

- **Red commit** `4445f91`: `repeater_liveness_test.go` + stub
`GetRepeaterRelayInfo` returning zero. Active and Stale tests fail on
assertion (LastRelayed empty / mismatched). Idle and IgnoresAdverts
already match the desired behavior under the stub. Compiles, runs, fails
on assertions — not on imports.
- **Green commit** `5fcfb57`: Implementation. All four tests pass. Full
`cmd/server` suite green (~22s).

## Performance

`O(N)` over `byPathHop[pubkey]` per call. The index is bounded by store
eviction; a single repeater has at most a few hundred entries on real
data. The `/api/nodes` loop adds one map read + scan per repeater row —
negligible against the existing enrichment work.

## Limitations (per issue body)

1. Observer coverage gaps — if no observer hears a repeater's relay,
it'll show as idle even when actively relaying. This is inherent to
passive observation.
2. Low-traffic networks — a repeater in a quiet area legitimately shows
idle. The 🟡 indicator copy makes that explicit ("alive (idle)").
3. Hash collisions are mitigated by the existing `resolveWithContext`
path before pubkeys land in `byPathHop`.

Fixes #662

---------

Co-authored-by: clawbot <bot@corescope.local>
2026-05-05 01:17:52 -07:00

144 lines
4.6 KiB
Go

package main
import (
"strings"
"time"
)
// RepeaterRelayInfo describes whether a repeater has been observed
// relaying traffic (appearing as a path hop in non-advert packets) and
// when. This is distinct from advert-based liveness (last_seen / last_heard),
// which only proves the repeater can transmit its own adverts.
//
// See issue #662.
type RepeaterRelayInfo struct {
// LastRelayed is the ISO-8601 timestamp of the most recent non-advert
// packet where this pubkey appeared as a relay hop. Empty if never.
LastRelayed string `json:"lastRelayed,omitempty"`
// RelayActive is true if LastRelayed falls within the configured
// activity window (default 24h).
RelayActive bool `json:"relayActive"`
// WindowHours is the active-window threshold actually used.
WindowHours float64 `json:"windowHours"`
// RelayCount1h is the count of distinct non-advert packets where this
// pubkey appeared as a relay hop in the last 1 hour.
RelayCount1h int `json:"relayCount1h"`
// RelayCount24h is the count of distinct non-advert packets where this
// pubkey appeared as a relay hop in the last 24 hours.
RelayCount24h int `json:"relayCount24h"`
}
// payloadTypeAdvert is the MeshCore payload type for ADVERT packets.
// See firmware/src/Mesh.h. Adverts are NOT considered relay activity:
// a repeater that only sends adverts proves it is alive, not that it
// is forwarding traffic for other nodes.
const payloadTypeAdvert = 4
// parseRelayTS attempts to parse a packet first-seen timestamp using the
// formats CoreScope writes in practice. Returns zero time and false on
// failure. Accepted (in order):
// - RFC3339Nano — Go's default UTC marshal output
// - RFC3339 — second-precision ISO-8601 with offset
// - "2006-01-02T15:04:05.000Z" — millisecond-precision Z form used by ingest
func parseRelayTS(ts string) (time.Time, bool) {
if ts == "" {
return time.Time{}, false
}
if t, err := time.Parse(time.RFC3339Nano, ts); err == nil {
return t, true
}
if t, err := time.Parse(time.RFC3339, ts); err == nil {
return t, true
}
if t, err := time.Parse("2006-01-02T15:04:05.000Z", ts); err == nil {
return t, true
}
return time.Time{}, false
}
// GetRepeaterRelayInfo returns relay-activity information for a node by
// scanning the byPathHop index for non-advert packets that name the
// pubkey as a hop. It computes the most recent appearance timestamp,
// 1h/24h hop counts, and whether the latest appearance falls within
// windowHours.
//
// Cost: O(N) over the indexed entries for `pubkey`. The byPathHop index
// is bounded by store eviction; on real data this is small per-node.
//
// Note on self-as-source: byPathHop is keyed by every hop in a packet's
// resolved path, including the originator. For ADVERT packets that's the
// node itself, which is filtered above by the payloadTypeAdvert check.
// For non-advert packets a node "originates" rather than "relays" only
// when it is the source; we don't currently have a clean signal for that
// distinction, so the count here is *path-hop appearances in non-advert
// packets*. In practice for a repeater nearly all such appearances are
// relay hops (the firmware doesn't originate user traffic), so this is
// the right approximation for issue #662.
func (s *PacketStore) GetRepeaterRelayInfo(pubkey string, windowHours float64) RepeaterRelayInfo {
info := RepeaterRelayInfo{WindowHours: windowHours}
if pubkey == "" {
return info
}
key := strings.ToLower(pubkey)
s.mu.RLock()
txList := s.byPathHop[key]
// Copy only the timestamps + payload types we need so we can release
// the read lock before doing parsing/compare work below.
type entry struct {
ts string
pt int
}
scratch := make([]entry, 0, len(txList))
for _, tx := range txList {
if tx == nil {
continue
}
pt := -1
if tx.PayloadType != nil {
pt = *tx.PayloadType
}
scratch = append(scratch, entry{ts: tx.FirstSeen, pt: pt})
}
s.mu.RUnlock()
now := time.Now().UTC()
cutoff1h := now.Add(-1 * time.Hour)
cutoff24h := now.Add(-24 * time.Hour)
var latest time.Time
var latestRaw string
for _, e := range scratch {
// Self-originated adverts are not relay activity (see header comment).
if e.pt == payloadTypeAdvert {
continue
}
t, ok := parseRelayTS(e.ts)
if !ok {
continue
}
if t.After(latest) {
latest = t
latestRaw = e.ts
}
if t.After(cutoff24h) {
info.RelayCount24h++
if t.After(cutoff1h) {
info.RelayCount1h++
}
}
}
if latestRaw == "" {
return info
}
info.LastRelayed = latestRaw
if windowHours > 0 {
cutoff := now.Add(-time.Duration(windowHours * float64(time.Hour)))
if latest.After(cutoff) {
info.RelayActive = true
}
}
return info
}