- User directive: Soft-delete nodes (inactive flag instead of deletion)
- Merged copilot-directive-soft-delete-nodes.md into Active Decisions section
- Removed processed inbox file
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Replace expensive per-request distance computation (1.2s cold) with
precomputed distance index built during Load() and incrementally
updated on IngestNewFromDB/IngestNewObservations.
- Add distHopRecord/distPathRecord types for precomputed hop distances
- buildDistanceIndex() iterates all packets once during Load(), computing
haversine distances and storing results in distHops/distPaths slices
- computeDistancesForTx() handles per-packet distance computation,
shared between full rebuild and incremental ingest
- IngestNewFromDB appends distance records for new packets (no rebuild)
- IngestNewObservations triggers full rebuild only if paths changed
- computeAnalyticsDistance() now aggregates from precomputed records
instead of re-iterating all packets with JSON parsing + haversine
Cold request path: ~10-20ms (filter + sort precomputed records)
vs previous: ~1.2s (iterate 30K+ packets, parse JSON, resolve hops,
compute haversine for each).
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Channels with garbage-decrypted names (pre-#197 data still in DB) are now
filtered at the API level using the same non-printable character heuristic
from #197. Applied in both Node.js server.js and Go server (store.go, db.go).
No data is deleted — only filtered from API responses.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
The subpaths analytics endpoint iterated ALL packets on every cold query,
taking ~900ms. The TTL cache only masked the problem.
Fix: maintain a precomputed raw-hop subpath index (map[string]int) that
is built once during Load() and incrementally updated during
IngestNewFromDB() and IngestNewObservations().
At query time the fast path iterates only unique raw subpaths (typically
a few thousand entries) instead of all packets (30K+), resolves hop
prefixes to names, and merges counts. Region-filtered queries still
fall back to the O(N) path since they require per-transmission observer
checks.
Expected cold-hit improvement: ~900ms → <5ms for the common no-region
case.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
- Create inactive_nodes table with identical schema to nodes
- Add retention.nodeDays config (default 7) in Node.js and Go
- On startup: move nodes not seen in N days to inactive_nodes
- Daily timer (24h setInterval / goroutine ticker) repeats the move
- Log 'Moved X nodes to inactive_nodes (not seen in N days)'
- All existing queries unchanged — they only read nodes table
- Add 14 new tests for moveStaleNodes in test-db.js
- Both Node (db.js/server.js) and Go (ingestor/server) implemented
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
advert_count was incremented on every upsertNode call, meaning each
observation of the same ADVERT packet inflated the count. Node N6NU
showed 4191 'adverts' but only had 77 unique ADVERT transmissions.
Changes:
- db.js: Remove advert_count increment from upsertNode SQL. Add
separate incrementAdvertCount() called only for new transmissions.
insertTransmission() now returns isNew flag.
- server.js: All three ADVERT processing paths (MQTT format 1,
companion bridge, API) now check isNew before incrementing.
- cmd/ingestor/db.go: Same fix in Go — UpsertNode no longer
increments, new IncrementAdvertCount method added.
InsertTransmission returns (bool, error) with isNew flag.
- cmd/ingestor/main.go: Check isNew before calling IncrementAdvertCount.
- One-time startup migration recalculates advert_count from
transmissions table (payload_type=4 matching node public_key).
Fixes#200
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
The #197 decryption fix added channelKeys parameter to decodePayload and
DecodePacket, but the test call sites were malformed:
- DecodePacket(hex, nil + stringExpr) → nil concatenated with string (type error)
- decodePayload(type, make([]byte, N, nil)) → nil used as make capacity (type error)
Fixed to:
- DecodePacket(hex + stringExpr, nil) → string concat then nil channelKeys
- decodePayload(type, make([]byte, N), nil) → proper 3-arg call
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
1. Update golden shapes.json goRuntime keys to match new struct fields
(goroutines, heapAllocMB, heapSysMB, etc. replacing heapMB, sysMB, etc.)
2. Fix analytics_hash_sizes hourly element shape — use explicit keys instead
of dynamicKeys to avoid flaky validation when map iteration picks 'hour'
string value against number valueShape
3. Update TestPerfEndpoint to check new goRuntime field names
4. Guard +Inf in handlePerf: use safeAvg() instead of raw division that
produces infinity when endpoint count is 0
5. Fix TestBroadcastMarshalError: use func(){} in map instead of chan int
to avoid channel-related marshal errors in test output
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Added 'set -e -o pipefail' to both Go test steps. Without pipefail, the exit code from 'go test' was being lost when piped to tee, causing test failures to appear as successes.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
IngestNewFromDB was appending new transmissions to byPayloadType slices,
breaking the newest-first ordering established by Load(). This caused
GetChannelMessages (which iterates backwards assuming newest-first) to
place newly ingested messages at the wrong position, making them invisible
when returning the latest messages from the tail.
Changed append to prepend, matching the existing s.packets prepend pattern
on line 881. Added regression test.
fixes#198
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
#195 — /api/nodes/:pubkey/analytics was hitting SQL (packets_v view)
for all queries. Added store.GetNodeAnalytics(pubkey, days) that uses
the byNode[pubkey] index + text search through decoded_json, computing
all analytics (timeline, SNR trend, type breakdown, observer coverage,
hop distribution, peer interactions, uptime heatmap, computed stats)
entirely in-memory. Route handler now uses store path when available,
falling back to SQL only when store is nil.
#196 — recentPackets from /api/nodes/:pubkey/health were missing the
_parsedPath field that Node.js includes (lazy-cached parsed path_json
array). Added _parsedPath to txToMap() output using txGetParsedPath(),
matching the Node.js packet shape.
fixes#195, fixes#196
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
After decryption produces text, validate it's printable UTF-8.
If it contains more than 2 non-printable characters (excluding
newline/tab), mark as decryption_failed with text: null.
Applied to both Node (decoder.js) and Go (cmd/ingestor/decoder.go)
decoders. Added tests for garbage and valid text in both.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Frontend reads goroutines/pauseTotalMs/lastPauseMs/heapAllocMB/heapSysMB/
heapInuseMB/heapIdleMB/numCPU but Go was returning heapMB/sysMB/numGoroutine/
gcPauseMs. All showed as undefined.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Add defensive type safety to node detail page rendering:
- Wrap all .toFixed() calls with Number() to handle string values from Go backend
- Use Array.isArray() for hash_sizes_seen instead of || [] fallback
- Apply same fixes to both full-screen and side-panel views
- Add 9 new tests for renderHashInconsistencyWarning and renderNodeBadges
with hash_size_inconsistent data (including non-array edge cases)
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
#191: Hash collision matrix now filters to role=repeater only (routing-relevant)
#192: expand=observations in /api/packets now returns full observation details (txToMap includes observations, stripped by default)
#193: /api/nodes/:pubkey/health uses in-memory PacketStore when available instead of slow SQL queries
#194: goRuntime (heapMB, sysMB, numGoroutine, numGC, gcPauseMs) restored in /api/perf response
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
#184: Strip non-printable chars (<0x20 except tab/newline) from ADVERT
names in Go server decoder, Go ingestor decoder, and Node decoder.js.
#185: Add visual (N) badge next to node names when multiple nodes share
the same display name (case-insensitive). Shows in list, side pane, and
full detail page with 'also known as' links to other keys.
#186: Add packetsLast24h field to /api/stats response.
#187#188: Cache runtime.ReadMemStats() with 5s TTL in Go server.
#189: Temporarily patch HTMLCanvasElement.prototype.getContext during
L.heatLayer().addTo(map) to pass { willReadFrequently: true }, preventing
Chrome console warning about canvas readback performance.
Tests: 10 new tests for buildDupNameMap + dupNameBadge (143 total frontend).
Cache busters bumped.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
decodePath() trusted the pathByte hop count without checking available
buffer space. Corrupt packet cbecda1c7d37d4c0 (route_type=3, pathByte
0xAD) claimed 45 hops × 3 bytes = 135 bytes, but only 65 bytes existed
past the header. Node's Buffer.subarray silently returns empty buffers
for out-of-range slices, producing 23 empty-string hops in the output.
Fix: clamp hashCount to floor(available / hashSize). Add a 'truncated'
flag so consumers know the path was incomplete. No empty hops are ever
returned now.
fixes#183
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Add per-payload-type packet detail fixtures captured from production:
- packet-type-advert.json (payload_type=4, ADVERT)
- packet-type-grptxt-decrypted.json (payload_type=5, decrypted GRP_TXT)
- packet-type-grptxt-undecrypted.json (payload_type=5, decryption_failed GRP_TXT)
- packet-type-txtmsg.json (payload_type=1, TXT_MSG)
- packet-type-req.json (payload_type=0, REQ)
Update validate-protos.py to validate all 5 new fixtures against
PacketDetailResponse proto message.
Update CI deploy workflow to automatically capture per-type fixtures
on each deploy, including both decrypted and undecrypted GRP_TXT.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
The /api/observers handler assumed byObserver arrays were sorted
newest-first and used an early break when hitting an old timestamp.
In reality, byObserver is only roughly DESC from the initial DB load;
live-ingested observations are appended at the end (oldest-to-newest).
After ~1 hour of uptime, the first element is old, the break fires
immediately, and every observer returns packetsLastHour=0.
Fix: full scan without break — the array is not uniformly sorted.
The endpoint is cached so performance is unaffected.
fixes#182
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
- #178: Use strftime ISO 8601 format instead of datetime() for observation
timestamps in all SQL queries (v3 + v2 views). Add normalizeTimestamp()
helper for non-v3 paths that may store space-separated timestamps.
- #179: Strip internal fields (decoded_json, direction, payload_type,
raw_hex, route_type, score, created_at) from ObservationResp. Only
expose id, transmission_id, observer_id, observer_name, snr, rssi,
path_json, timestamp — matching Node.js parity.
- #180: Remove _parsedDecoded and _parsedPath from node detail
recentAdverts response. These internal/computed fields were leaking
to the API. Updated golden shapes.json accordingly.
- #181: Use mux route template (GetPathTemplate) for perf stats path
normalization, converting {param} to :param for Node.js parity.
Fallback to hex regex for unmatched routes. Compile regexes once at
package level instead of per-request.
fixes#178, fixes#179, fixes#180, fixes#181
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Deduplicated protobuf contract + fixture directive into single entry.
Protobuf API contract is now single source of truth for all frontend/backend interfaces, with fixture capture running against prod (stable).
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Go ingestor never had channel decryption — GRP_TXT packets were stored
with raw encrypted data while Node.js decoded them successfully.
Changes:
- decoder.go: Add decryptChannelMessage() implementing MeshCore channel
crypto (HMAC-SHA256 MAC verification + AES-128-ECB decryption), matching
the algorithm in @michaelhart/meshcore-decoder. Update decodeGrpTxt(),
decodePayload(), and DecodePacket() to accept and pass channel keys.
Add Payload fields: ChannelHashHex, DecryptionStatus, Channel, Text,
Sender, SenderTimestamp.
- config.go: Add ChannelKeysPath and ChannelKeys fields to Config struct.
- main.go: Add loadChannelKeys() that loads channel-rainbow.json (same
file used by Node.js server) from beside the config file, with env var
and config overrides. Pass loaded keys through the decoder pipeline.
- decoder_test.go: Add 14 channel decryption tests covering valid
decryption, MAC failure, wrong key, no-sender messages, bracket
sender exclusion, key iteration, channelHashHex formatting, and
decryption status states. Cross-validated against Node.js output.
- Update all DecodePacket/decodePayload/decodeGrpTxt/handleMessage call
sites in test files to pass the new channelKeys parameter.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Automated script that compares all 13 major API endpoints between
Go staging (meshcore-staging-go) and Node prod (meshcore-prod)
containers. Uses python3 for JSON field diffing and reports
MATCH/PARTIAL/MISMATCH per endpoint.
Usage: scp to server then run, or pipe via ssh.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
The env var was overriding the config and forcing Go staging to only
connect to its own empty local mosquitto, missing all external data.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
The poller only queried WHERE t.id > sinceID, which missed new
observations added to transmissions already in the store. The trace
page was correct because it always queries the DB directly.
Add IngestNewObservations() that polls observations by o.id watermark,
adds them to existing StoreTx entries, re-picks best observation, and
invalidates analytics caches. The Poller now tracks both lastTxID and
lastObsID watermarks.
Includes tests for v3, v2, dedup, best-path re-pick, and
GetMaxObservationID.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
- Change .traces-page max-width from 1000px to 95vw via CSS variable
(--trace-max-width) and center with margin: 0 auto
- Increase SVG path graph column spacing from 140px to 200px so nodes
and labels don't overlap
- Bump cache busters
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
#165 — build_time in API: already implemented (BuildTime ldflags in
Dockerfile.go, main.go, StatsResponse, HealthResponse)
#168 — subpaths API slow: cache (subpathCache with TTL) and invalidation
already in place; verified working
#169 — distance API slow: cache (distCache with TTL) and invalidation
already in place; verified working
#170 — audio-lab/buckets: in-memory store path already implemented,
matching Node.js pktStore.packets iteration with type grouping and
size-distributed sampling
#171 — channels stale latest message: add companion bridge handling to
Go ingestor for meshcore/message/channel/<n> and meshcore/message/direct/<id>
MQTT topics. Stores decoded channel messages with type CHAN in decoded_json,
enabling the channels endpoint to find them. Also handles direct messages.
#172 — packets page not live-updating: add missing direction field to WS
broadcast packet map for full parity with txToMap/Node.js fullPacket shape.
WS broadcast shape verified correct (type, data.packet structure, timestamp,
payload_type, observer_id all present).
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
db.GetChannels() queried packets_v (observation-level rows) ordered by
observation timestamp and always overwrote lastMessage. When an older
message had a later re-observation, it would overwrite the correct
latest message with stale data.
Fix: query transmissions table directly (one row per unique message)
ordered by first_seen. This ensures lastMessage always reflects the
most recently sent message, not the most recently observed one.
Also fix db.GetChannelMessages() to use first_seen ordering with
schema-aware queries (v2/v3), and add missing distCache/subpathCache
invalidation on packet ingestion.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Use s.store (in-memory PacketStore) instead of direct packets_v SQL query,
matching how the Node.js handler iterates pktStore.packets. This fixes the
endpoint returning empty buckets when packets_v view is missing or the DB
query fails silently.
Key changes:
- Group by decoded_json.type first, fall back to payloadTypeNames
- Evenly-spaced sampling (up to 8 per type) sorted by raw_hex length
- Use actual ObservationCount instead of hardcoded 1
- Reuse payloadTypeNames from store.go instead of duplicating
- Retain DB fallback path when store is nil
fixes#170
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
- Added decision: Proto fixture capture in CI runs against prod (stable reference)
- Added decision: Protobuf API contract architecture (single source of truth for frontend/backend interfaces)
- Merged 2 directives from spawn manifest processing
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
- Add 15s TTL cache to GetAnalyticsSubpaths with composite key (region|minLen|maxLen|limit),
matching the existing cache pattern used by RF, topology, hash, channel, and distance analytics.
Cache hits return instantly vs 900ms+ computation. fixes#168
- Add BuildTime to /api/stats and /api/health responses, injected via ldflags at build time.
Dockerfile.go now accepts BUILD_TIME build arg. fixes#165
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
The proto validation infrastructure was added in commit e70ba44 but used
an invalid --syntax_check flag. Changed to use --descriptor_set_out=/dev/null
which validates syntax without generating files.
Proto validation flow (now complete):
1. go-test job: verify .proto files compile (syntax check) ✅
2. deploy-node job: validate protos match prod API responses ✅
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
The Go server's WebSocket broadcast included first_seen but not
timestamp in the nested packet object. The frontend packets.js
filters on m.data.packet and reads p.timestamp for row insertion
and sorting. Without this field, live-updating silently failed
(rows inserted with undefined latest, breaking display).
Mirrors the pattern already used in txToMap() (store.go:1168)
which correctly emits both first_seen and timestamp.
Also updates websocket_test.go to assert timestamp presence
in broadcast data to prevent regression.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
The build-node job was failing with 'node: not found' because it
runs scripts/validate.sh (which uses 'node -c' for syntax checking)
but didn't have the actions/setup-node@v4 step.
Added Node.js 22 setup before the validate step to match the pattern
used in other jobs.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
The distance analytics endpoint was recomputing haversine distances on
every request (~1.22s). Add a 15s TTL cache following the same pattern
used by RF, topology, hash-sizes, and channels analytics endpoints.
Include distCache in cache stats size calculation.
fixes#169
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Previously only captured 19 simple endpoints. Now captures all 33:
- 19 simple endpoints (stats, health, nodes, etc.)
- 14 dynamic ID endpoints (node-detail, packet-detail, etc.)
Dynamic ID resolution:
- Extracts real pubkey from /api/nodes for node detail endpoints
- Extracts real hash from /api/packets for packet-detail
- Extracts real observer ID from /api/observers for observer endpoints
- Gracefully skips fixtures if DB is empty (no data yet)
WebSocket capture:
- Uses node -e with ws module to capture one live WS message
- Falls back gracefully if no live packets available
The validator already handles missing fixtures without failing, so this
will work even when staging container has no data yet.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Added a CI step that:
- Refreshes Node fixtures from the staging container after deployment
- Runs tools/validate-protos.py to validate proto definitions match actual API responses
- Fails the pipeline if proto drift is detected
This ensures nobody can merge a Node change that breaks the Go proto contract
without updating the .proto definitions.
The step runs after the Node staging healthcheck, capturing fresh responses
from 19 API endpoints (stats, health, nodes, analytics/*, config/*, etc.).
Endpoints requiring parameters (node-detail, packet-detail) use existing
fixtures and aren't auto-refreshed.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Validates all 33 captured Node fixtures against the 10 .proto files.
Parses proto message definitions, maps each fixture to its response
message, and checks field presence, types, and structural shape.
Mismatches found (for Hicks to fix):
1. analytics-channels.json: ChannelAnalyticsSummary.hash is int in
fixture but proto says string
2. analytics-rf.json: PayloadTypeEntry.type is null in fixture but
proto says non-optional int32
3. bulk-health.json: API returns bare array with flat node fields;
proto nests them in BulkHealthEntry.node (structural mismatch)
4. node-analytics.json: activityTimeline uses 'bucket' key but
TimeBucket proto expects 'label'
5. observer-analytics.json: recentPackets are Observation-shaped
(have transmission_id) but proto says repeated Transmission
6. packet-detail.json: ByteRange has 'color' field not in proto
7. websocket-message.json: DecodedResult missing transportCodes,
raw fields; DecodedHeader missing routeTypeName; DecodedPayload
is flat (not oneof-wrapped); WSPacketData.packet is Observation-
shaped, not Transmission-shaped
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
