docs: align wire-spec and mark implemented vs proposed

Keep the 45-byte pose frame consistent across docs, simulator, and
prototype, clarify MinIO is not peer sync, and state clearly that the
repo is a from-scratch platform sketch with options rather than mandates.

AGENTS.md

@@ -29,3 +29,52 @@ See [`CONTRIBUTING.md`](CONTRIBUTING.md) for the development workflow and TDD re
 - TypeScript: strict mode, functional components and hooks, no extra UI libraries.
 - YAML: 2-space indent.
 - Diagrams: Mermaid inside Markdown.
+
+## Decision records
+
+Significant architectural decisions are recorded chronologically as **ADRs** in
+[`docs/adr/`](docs/adr/) — one immutable file per decision, in the
+Status / Context / Options / Decision / Consequences format. Start from
+[`docs/adr/README.md`](docs/adr/README.md).
+
+- **ADR** = a point-in-time decision for *this* repo → `docs/adr/`.
+- **ASR** = a standing, ecosystem-wide standard (Go-first, secrets, layout) →
+  lives in the `inventar` repo, **not** here.
+
+When a change alters a boundary, contract, or trade-off, add an ADR (copy
+[`docs/adr/TEMPLATE.md`](docs/adr/TEMPLATE.md), take the next number, update the
+index) and cross-link it from the affected design doc. A later reversal gets a
+new ADR that supersedes the old one — never edit an Accepted record.
+
+## Runtime & operations (sim environment)
+
+The k3d ground miniature is `swarm-sim`. Services are exposed as NodePorts:
+
+| Service | URL |
+| --- | --- |
+| Explorer (T1 live lake) | `http://localhost:30088` |
+| Warehouse explorer (T3) | `http://localhost:30089` |
+| Grafana (dashboards `swarm-fleet`, `swarm-platform`) | `http://localhost:30300` |
+| Prometheus | `http://localhost:30990` |
+| MinIO console | `http://localhost:30901` |
+| Prototype dev server (`npm run dev`) | `http://localhost:5173` |
+
+Rebuild the simulator image and roll it out:
+
+```bash
+cd simulator && docker build -t swarm-house/simulator:dev .
+k3d image import swarm-house/simulator:dev -c swarm-sim
+kubectl rollout restart -n swarm statefulset/drone deployment/explorer deployment/exporter
+cd infra/terraform/sim-env && terraform apply
+```
+
+**CPU budget knobs** (see [ADR-0009](docs/adr/ADR-0009-bounded-lake-scans.md)):
+`METRIC_FLIGHT_WINDOW` (scan only recent flights), `SCAN_INTERVAL_S`,
+`VIEW_REFRESH_S`, `TREE_CACHE_S`, `KEEP_ALIVE=1` (drones idle after seal instead
+of restarting), `IMU_HZ`. Never scan the full lake on a hot path — it grows with
+pod churn. Long-horizon analysis is a T3 warehouse query, not an exporter job.
+
+Runtime data lives under `var/t1` / `var/t3`, owned by uid **10001**
+([ADR-0007](docs/adr/ADR-0007-non-root-image.md),
+[ADR-0008](docs/adr/ADR-0008-runtime-data-paths.md)); both are gitignored.
+Terraform state is local and never committed.

CONTRIBUTING.md

@@ -81,9 +81,18 @@ identifying names of people, companies, or locations.
 - [ ] Tests added or updated; `pytest tests/ -q` passes locally
 - [ ] `docker build` in `simulator/` succeeds (test stage included)
 - [ ] Documentation updated if behaviour or boundaries changed
+- [ ] **ADR added** if the change alters an architectural boundary, contract, or trade-off (see [`docs/adr/`](docs/adr/))
 - [ ] English only; no identifying references
 - [ ] Conventional commit message
 
+## Decisions
+
+Architectural decisions are recorded chronologically as ADRs in
+[`docs/adr/`](docs/adr/). Read [`docs/adr/README.md`](docs/adr/README.md) before
+changing a boundary or contract; add a new ADR (from
+[`docs/adr/TEMPLATE.md`](docs/adr/TEMPLATE.md)) rather than editing an Accepted
+one when you reverse course.
+
 ## Questions
 
 Open design questions live in [`docs/09-open-questions.md`](docs/09-open-questions.md).

LICENSE

@@ -0,0 +1,19 @@
+PROPRIETARY LICENSE
+
+Copyright (c) 2026 Andriy Oblivantsev. All rights reserved.
+
+This material is proprietary and confidential.
+
+No person or organization may, without prior written permission from the copyright holder:
+
+- Publish or republish this material, in whole or in part.
+- Copy, reproduce, distribute, transmit, forward, or share this material by any means.
+- Use, modify, adapt, translate, create derivative works from, or incorporate this material into any other work.
+- Make this material available to any third party, whether publicly or privately.
+- Store this material in any repository, archive, or knowledge base accessible to others.
+
+Permission must be obtained in writing before any of the above activities are undertaken.
+
+Unauthorized use, publication, forwarding, reproduction, distribution, or disclosure of this material is strictly prohibited and may result in civil and criminal penalties under applicable copyright and intellectual property laws.
+
+This license grants no rights except the right to view the material for its intended purpose by an authorized recipient.

README.md

@@ -17,6 +17,25 @@ This repository describes how to build, deliver, test, and operate that platform
 7. **SQL is the contract.** Peer data access is read-only DuckDB SQL over SSH forced commands — the query language already lives on both ends, so no service, port, or protocol is invented for it.
 8. **No debug API.** The bench and the flight use the same channel: an engineer debugging on the ground runs the identical query through the identical wrapper, permissions, and output format a peer drone would use. What you test is what flies.
 
+None of the concrete tool picks above are mandates. This repo is a **from-scratch platform sketch**: enough structure to hire and build against, with every decision recorded so the team can replace a piece when a better fit appears.
+
+## Implemented now vs proposed next
+
+Honest map so a reader knows what runs today versus what is design intent.
+
+| Area | Implemented in this repo (runnable PoC) | Proposed for a production air-gapped fleet |
+| --- | --- | --- |
+| On-board layout | Hive-partitioned Parquet writer, seal step, DuckDB views | Same contract; Compose services under systemd |
+| Pose path | Fixed **45-byte** UDP frame + 2D bandwidth visualisation | Zenoh pub/sub (UDP kept as degraded minimal profile) |
+| Peer query | Read-only SQL **gate** over HTTP explorer (keyword allow-list) | Same gate idea via **SSH forced command** + OS/engine hardening |
+| Bulk sync | Visualised opportunistic transfer volume | rsync/rclone over persistent SSH between peers |
+| Mesh trust | Ansible templates for WireGuard + ed25519 forced commands | Provisioned per-device keys; nothing joins at runtime |
+| Ground segment | k3d/Terraform sim: lake, Grafana, offload CronJob, optional MinIO | k3s warehouse, GitOps overlays, post-flight mirror |
+| CI / delivery | GitHub Actions: pytest, smoke flight, Trivy, semver release + fleet manifest artifact | Self-hosted GitLab + registry inside the air gap (same stages) |
+| Docs | Problem, architecture, ADRs, design journey, open questions | Living ADRs owned by the team |
+
+Start from the [design journey](docs/12-design-journey.md) for the story; use the table above when reviewing scope.
+
 ## Documentation
 
 | Document | Contents |
@@ -33,6 +52,10 @@ This repository describes how to build, deliver, test, and operate that platform
 | [09 — Open questions](docs/09-open-questions.md) | Known unknowns and proposed answers |
 | [10 — Domain context](docs/10-domain-context.md) | Swarm autonomy principles this design builds on |
 | [11 — CI/CD & delivery](docs/11-cicd-delivery.md) | Pipelines, registry, dev containers, fleet releases |
+| [12 — Design journey](docs/12-design-journey.md) | **Start here** — a narrative walk-through of how the design came together, linking into the code |
+| [ADRs](docs/adr/) | Architecture Decision Records — the decisions behind the above, in the order they were made |
+
+New here? Read the [design journey](docs/12-design-journey.md) first: it tells the story chronologically and links straight into the code and decisions. The design principles below are the *what*; the [ADRs](docs/adr/) are the *why and when* — each principle traces to a dated, immutable decision record.
 
 ## Runnable parts
 
@@ -73,3 +96,5 @@ Contributing (TDD-first workflow): [`CONTRIBUTING.md`](CONTRIBUTING.md).
 ## CI & releases
 
 Every push runs the pipeline in [`.github/workflows/release.yaml`](.github/workflows/release.yaml): **pytest unit tests**, a compile check, a 10-second virtual smoke flight verified with DuckDB, and a Trivy vulnerability/misconfiguration scan. The simulator Docker build also runs `pytest` — a failing test blocks the image. Pushes to `main` then cut a semantic version from the conventional-commit type (`feat:` → minor, `!`/`BREAKING CHANGE` → major, else patch), tag the commit, and publish a release with a **fleet release manifest** (versioned artifact list with checksums — the delivery unit described in [11 — CI/CD & delivery](docs/11-cicd-delivery.md)) plus a source bundle.
+
+[LICENSE](LICENSE)

docs/02-architecture.md

@@ -20,8 +20,9 @@ graph TB
         subgraph serving [Layer 3 — Serving and sync]
             HOOK["event hook<br/>fires on new derived data"]
             PUB["state publisher<br/>pub/sub broadcast"]
-            MINIO["MinIO<br/>derived datasets bucket"]
-            QAPI["query API<br/>peer data requests"]
+            BULK["bulk sync<br/>rsync over SSH"]
+            MINIO["MinIO<br/>optional derived bucket"]
+            QAPI["query API<br/>SQL-over-SSH"]
         end
     end
 
@@ -33,12 +34,14 @@ graph TB
     WRITER -->|derived rows| HOOK
     HOOK --> PUB
     HOOK --> MINIO
+    HOOK --> BULK
     QAPI --> DUCK
-    PUB -.->|mesh| PEERS["peer drones"]
-    MINIO -.->|replication| PEERS
+    PUB -.->|mesh pose| PEERS["peer drones"]
+    BULK -.->|sealed partitions| PEERS
     QAPI -.->|on demand| PEERS
 ```
 
+MinIO stays available where an S3 API helps (on-board derived datasets, ground warehouse). **In-flight peer bulk sync is SSH/rsync**, not object-store replication — see [04 — Swarm sync](04-swarm-sync.md).
 ### Layer 1 — Ingestion
 
 - `sensor-ingest` subscribes to sensor sources (ROS 2 topics where available, raw drivers otherwise) and normalizes them into typed streams: IMU, barometer, temperature, LiDAR, RSSI, power, and so on.
@@ -53,10 +56,10 @@ graph TB
 
 ### Layer 3 — Serving and sync
 
-- The **event hook** is the on-board "lambda": when the writer lands new *derived* rows (state, detections), it triggers registered actions — broadcast, MinIO upload, or a local mission-logic callback. Nothing polls.
-- The **state publisher** broadcasts compact position/attitude/detection payloads over the mesh pub/sub (transport analysis in [04 — Swarm sync](04-swarm-sync.md)).
-- **Bulk sync** pulls sealed derived partitions from peers over persistent SSH (rsync delta transfer); MinIO remains optional where an S3 API is wanted.
-- **Peer queries** are read-only DuckDB SQL over SSH forced commands — SELECT-only gate, read-only OS user, columnar responses (the design decision and its reasoning are in [04](04-swarm-sync.md)).
+- The **event hook** is the on-board "lambda": when the writer lands new *derived* rows (state, detections), it triggers registered actions — broadcast, optional MinIO put, bulk-sync hint, or a local mission-logic callback. Nothing polls.
+- The **state publisher** broadcasts compact position/attitude/detection payloads over the mesh pub/sub (UDP in the PoC; Zenoh proposed — [04](04-swarm-sync.md)).
+- **Bulk sync** pulls sealed derived partitions from peers over persistent SSH (rsync delta transfer). MinIO is optional where an S3 API is wanted; it is **not** the in-flight peer replication path.
+- **Peer queries** are read-only DuckDB SQL — HTTP explorer gate in the PoC; **SSH forced commands** proposed for flight (SELECT-only gate, read-only OS user, columnar responses — [04](04-swarm-sync.md)).
 
 ## Communication planes
 

docs/04-swarm-sync.md

@@ -16,19 +16,27 @@ Bandwidth is the scarcest resource in the system. Every message class gets an ex
 
 ## Broadcast payload: small on the wire, precise at rest
 
-The `state` broadcast is a fixed compact frame:
+The `state` broadcast is a fixed compact frame. The runnable PoC encodes it as
+**45 bytes** little-endian (`simulator/virtual_drone/broadcast.py`); that size is
+what the prototype uses for bandwidth estimates.
 
-| Field | Type | Notes |
+| Field | Wire type (PoC) | Notes |
 | --- | --- | --- |
-| `drone_id` | uint16 | Fleet-scoped registry |
+| `magic` + `version` | 2s + uint8 | `b"SH"`, version `1` |
+| `drone_id` | 8s ascii | Zero-padded; a fleet `uint16` registry id is a natural production swap |
 | `ts_ns` | int64 | Epoch nanoseconds, same clock domain as storage |
-| `pos_x/y/z` | int32 | **Millimeters** in the mission frame — quantized only here, storage keeps full float precision |
-| `att_roll/pitch/yaw` | int16 | Centi-degrees |
-| `vel_x/y/z` | int16 | cm/s |
+| `pos_x/y/z` | 3 × int32 | **Millimeters** in the **mission frame** — quantized on the wire only; storage keeps full float precision |
+| `att_roll/pitch/yaw` | 3 × int16 | Centi-degrees — attitude stays on the wire so peers need no local shape model |
+| `vel_x/y/z` | 3 × int16 | cm/s |
 | `frame_ref` | uint8 | Frame of reference id (GPS-denied: local/visual-odometry frames must be explicit) |
 | `flags` | uint8 | Battery-low, returning, degraded-sensors, … |
 
-~40 bytes per frame → a 50-drone swarm at 5 Hz is ~10 KB/s of pose traffic before transport overhead. Trivial even on a congested mesh.
+45 bytes × 5 Hz × 50 drones ≈ 11 KB/s of pose traffic before transport overhead — still trivial on a congested mesh.
+
+An earlier sketch used relative coordinates and a bounding sphere (no attitude).
+It was dropped: mission-frame pose + attitude is simpler to fuse post-flight and
+costs almost nothing at this frame size. Relative localization remains a
+consumer concern when `frame_ref` differs across peers ([09](09-open-questions.md)).
 
 `detections` events are slightly larger (class, confidence, bounding volume, ego-pose) but event-shaped and rare by comparison.
 
@@ -51,25 +59,27 @@ graph LR
     SB -->|"store-and-forward relay"| SC
 ```
 
-### Recommended: Zenoh
+### Recommended (proposal): Zenoh
 
 - Designed exactly for constrained, dynamic networks: built-in peer discovery, brokerless peer-to-peer mode, store-and-forward, and a query layer on top of pub/sub.
 - First-class robotics citizenship: an official ROS 2 RMW implementation exists, so the ingestion side and the sync side can share one middleware.
 - Tiny footprint, ARM64-native.
 
+**PoC today:** the simulator and the 2D prototype exercise the **raw UDP** pose path only — the minimal degraded profile below. Zenoh is the proposed production pub/sub, not yet wired into the runnable stack.
+
 ### Alternatives considered
 
 | Option | Verdict |
 | --- | --- |
 | **DDS multicast** (ROS 2 default) | Works, battle-tested; but discovery storms and tuning pain on lossy wireless meshes are well documented. Keep as fallback since ROS 2 speaks it natively |
 | **MQTT** | Needs a broker — a per-drone broker bridge is possible but adds moving parts for no gain over Zenoh |
-| **Raw UDP multicast** | Perfect as a last-resort minimal profile for the pose broadcast alone (fixed frame, no discovery); no query layer, no reliability — documented as the degraded mode |
-| **MinIO bucket replication** | Wrong tool for the 5 Hz pose path, right tool for bulk derived datasets — see below |
+| **Raw UDP multicast** | **Implemented in the PoC** for pose broadcast (fixed frame, no discovery); no query layer, no reliability — also the documented degraded mode |
+| **MinIO bucket replication** | Wrong tool for the 5 Hz pose path; optional on board for derived datasets and primary on the ground warehouse — not the in-flight bulk path |
 
 ## Two sync mechanisms, deliberately separate
 
-1. **Fast path — pub/sub (Zenoh):** pose frames and detection events. Fire-and-forget with bounded staleness; consumers keep a peer-state cache.
-2. **Bulk path — rsync over persistent SSH:** sealed `detections`/`state` Parquet partitions are pulled opportunistically between drones when links allow. This is how a drone that was out of range catches up on mission history without anyone re-sending events.
+1. **Fast path — pub/sub (Zenoh proposed; UDP in the PoC):** pose frames and detection events. Fire-and-forget with bounded staleness; consumers keep a peer-state cache.
+2. **Bulk path — rsync over persistent SSH (proposed):** sealed `detections`/`state` Parquet partitions are pulled opportunistically between drones when links allow. This is how a drone that was out of range catches up on mission history without anyone re-sending events. The PoC visualises bulk volume; it does not yet run real rsync between virtual drones.
 
 Why SSH-based bulk sync over object-store replication:
 
@@ -105,7 +115,7 @@ SQL access must not become a write channel. A single "read-only connection" flag
 | Layer | Mechanism | What it stops |
 | --- | --- | --- |
 | 1. Key = operation | Forced command: the query key can only invoke the query wrapper, nothing else | Arbitrary exec, lateral movement |
-| 2. Statement gate | Wrapper accepts a single statement, parses it, rejects anything but `SELECT` (no `COPY`, `ATTACH`, `INSTALL`, `SET`, multi-statements); parameters bound, not interpolated | SQL-as-a-write-channel, config tampering |
+| 2. Statement gate | Wrapper accepts a single statement, rejects anything but `SELECT`/`WITH`/… (no `COPY`, `ATTACH`, `INSTALL`, `SET`, multi-statements); production should prefer a real parser + bound parameters, not keywords alone | SQL-as-a-write-channel, config tampering |
 | 3. OS permissions | Wrapper runs as a dedicated user with **read-only filesystem access** to the data root and write access to nothing | Any write that slips past layer 2 |
 | 4. Engine hardening | `:memory:` database, external access disabled except the data-root glob, extension loading off | Reaching outside the store |
 | 5. Resource caps | Timeout, memory cap, niced CPU (flight software always wins), response size budget | Denial of service via expensive queries |

docs/05-network-security.md

@@ -2,6 +2,8 @@
 
 Zero trust, fully air-gapped, everything provisioned before wheels-up.
 
+> **Decision:** [ADR-0005 — WireGuard beneath SSH](adr/ADR-0005-wireguard-beneath-ssh.md).
+
 ## Trust model
 
 - **Nothing joins the swarm at runtime.** Every device receives its identity (key pair + certificate) during ground provisioning, signed by the fleet's offline CA. There is no trust-on-first-use, no runtime enrollment endpoint, no exception path.

docs/06-environments.md

@@ -2,6 +2,8 @@
 
 Three infrastructures, one data contract. The layout, schemas, and pipelines are identical everywhere; only scale and lifetime differ.
 
+> **Decisions:** [ADR-0006 — IaC boundaries](adr/ADR-0006-iac-boundaries.md), [ADR-0008 — runtime data paths](adr/ADR-0008-runtime-data-paths.md).
+
 ```mermaid
 graph LR
     subgraph dev [3 — Dev and simulation]

docs/07-observability.md

@@ -2,6 +2,8 @@
 
 The platform watches itself with the same discipline it applies to sensor data — and largely through the same pipeline.
 
+> **Decision:** [ADR-0009 — bound observability lake scans to a flight window and cache](adr/ADR-0009-bounded-lake-scans.md).
+
 ## Two kinds of signals, one storage
 
 | Signal | Examples | Where it goes |
@@ -53,6 +55,13 @@ graph LR
 | Which link degraded first, and was it distance or interference? | RSSI telemetry joined with pose distance |
 | Is the new writer version flushing slower on ARM64? | CI simulation runs emit the same metrics; diff across fleet releases |
 
-A working miniature of this doctrine ships in [`simulator/`](../simulator/): the `monitoring` Compose profile runs a DuckDB-based exporter over the generated Parquet plus Prometheus and a provisioned Grafana dashboard — fleet statistics derived from the data platform itself, with no agent on the (virtual) drones.
+A working miniature of this doctrine ships in [`simulator/`](../simulator/): the `monitoring` Compose profile runs a DuckDB-based exporter over the generated Parquet plus Prometheus and two provisioned Grafana dashboards:
+
+| Dashboard | Focus |
+| --- | --- |
+| **Swarm Fleet — generated data** | Row counts, detections, pose frames, Parquet bytes, battery, RSSI |
+| **Swarm Platform — CPU & scan health** | Node CPU/memory (node-exporter), exporter/explorer scan durations, flight-window gauge |
+
+The exporter and explorer **scan only the most recent flight partitions** (`METRIC_FLIGHT_WINDOW`, default 5) and cache tree/views — otherwise CPU climbs as every pod restart appends a new `flight=` tree to the shared lake. Drones in k3d set `KEEP_ALIVE=1` after sealing so they idle instead of exiting and spawning another flight.
 
 Alerting on the ground follows standard practice (Prometheus alert rules for infrastructure, CI gates for regression in simulated staleness/throughput budgets). In flight there is nobody to page — the platform's job is to degrade in the documented order and record everything for the post-mortem.

docs/09-open-questions.md

@@ -61,3 +61,9 @@ Who owns the device registry (drone ids, key issuance, revocation) organizationa
 Sub-250 g class units cannot run the full stack (no GPU, minimal CPU/storage).
 
 *Proposal:* define a minimal profile early — state publisher + UDP-multicast pose broadcast, no local Parquet store, no video — so the swarm protocol never assumes full-stack peers.
+
+## 11 — Lake retention and partition pruning
+
+The observability path now scans only the most recent flights ([ADR-0009](adr/ADR-0009-bounded-lake-scans.md)), but old `flight=` partitions still accumulate on T1 disk after offload to T3.
+
+*Proposal:* a scheduled prune of T1 partitions whose flights are confirmed present in the T3 warehouse (offload as the retention gate), configurable by age and free-space watermark. On the drone the same policy is bounded by the NVMe quota ladder ([03](03-data-platform.md)); in the sim environment it is a ground CronJob alongside the offload job.

docs/11-cicd-delivery.md

@@ -2,6 +2,8 @@
 
 How software, models, and configuration reach the fleet — reproducibly, scanned, versioned, and atomically rollback-able. Everything lives inside the air gap.
 
+> **Decisions:** [ADR-0006 — IaC boundaries & fleet manifest](adr/ADR-0006-iac-boundaries.md), [ADR-0007 — non-root image](adr/ADR-0007-non-root-image.md).
+
 ## Source and pipelines: GitLab
 
 GitLab (self-managed, on-prem) is the backbone: repositories, CI/CD, and the container registry in one system.

docs/12-design-journey.md

@@ -0,0 +1,199 @@
+# 12 — Design journey
+
+How this design came together, told in the order the thinking actually happened.
+It is a walk-through, not a report — and **not a mandate**. The goal is a
+from-scratch platform sketch that shows how the pieces fit; every concrete
+choice is an option with trade-offs recorded as
+[Architecture Decision Records](adr/README.md). Replace any piece if a better
+fit shows up. This is the story behind the decisions, and each chapter links
+straight into the code it produced.
+
+---
+
+## 1. Starting point
+
+At the start there was only a rough brief and a few adjacent interests to lean
+on. Key rotation matters in every pipeline: a cipher that is safe today may not
+be safe next year, so the infrastructure has to leave room to react, without
+spending too much latency or bandwidth to do it. That balance, security against
+speed and volume, is a recurring trade-off. Geospatial data is older ground:
+routing, indexing, fast spatial lookups at scale. So the first instinct was that
+the value might be less about new storage and more about optimizing access to
+data that already exists.
+
+## 2. From words to a model, then to a picture
+
+Once the shape was clear, an autonomous swarm where each unit talks to any peer
+in range and [no controller spans the swarm](adr/ADR-0001-no-swarm-orchestrator.md),
+the next step was to make the model visible. A small 2D prototype was enough to
+check it: units patrolling, static and moving obstacles, links forming and
+breaking, and a rough count of how much data crosses the air. Speed was not the
+question yet, only volume, because volume is what you size the infrastructure
+for. Starting from a picture gives an anchor before building anything.
+
+> **Read the code**
+> - [`prototype/src/sim.ts`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/prototype/src/sim.ts#L219-L287) — the tick: links form and break by range, pose broadcasts and bulk sync accumulate bytes
+> - [`prototype/src/sim.ts`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/prototype/src/sim.ts#L57-L63) — the pose-frame size and rate that drive the volume estimate
+
+The picture is running, not just described: the visual prototype is live at
+[swarm.produktor.io](https://swarm.produktor.io/) (same access as this
+repository). It draws units patrolling, links forming and breaking by range, and
+a running estimate of how much data crosses the air.
+
+## 3. Naming the limits
+
+Before any code, the limits had to be named. Maximum distance between units in
+the air. The fidelity data is stored at. The fidelity it can be transmitted at.
+Those drive volume, and volume drives the whole shape of the system. The
+conclusion: constrain the transport, but do not constrain on-board capture. Keep
+everything locally. Send only what peers truly need.
+
+> **Read more**
+> - [01 — Problem statement](01-problem-statement.md) and [03 — Data platform](03-data-platform.md) — the constraints and the storage tiers
+
+## 4. Three tiers inside each unit
+
+That gave [three layers](03-data-platform.md) on every drone:
+
+1. **Raw capture**, never transformed.
+2. **ETL / reduction**, where data is cut down to what has to be shared:
+   millimetres to centimetres, thinning time series where that is enough,
+   dropping what nobody downstream reads.
+3. **A bidirectional interface**: on one side it broadcasts position into the
+   shared channel, on the other it lets peers pull data.
+
+The ETL is deliberately not over-specified. How sensor data is transformed is a
+data-engineering decision. The platform's job is the contract, meaning what gets
+stored and in which structure, not to step into that work.
+
+> **Read the code**
+> - [`simulator/virtual_drone/writer.py`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/simulator/virtual_drone/writer.py#L23-L80) — the Hive-partition layout and the seal step that compacts a flight
+> - [`simulator/virtual_drone/sensors.py`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/simulator/virtual_drone/sensors.py) — the raw capture that feeds tier one
+
+## 5. What actually needs to sync
+
+The most time-critical item is where each peer is, so every unit has time to
+react. The wire frame carries **mission-frame position** (millimetres on the
+wire, full float at rest), **attitude**, **velocity**, a `frame_ref`, and flags —
+45 bytes at 5 Hz. That is cheap enough that shrinking further is not worth the
+fusion pain. An earlier sketch used relative coordinates and a bounding sphere
+with no orientation; it was dropped. When peers disagree on frames,
+`frame_ref` makes the mismatch explicit for consumers
+([09 — Open questions](09-open-questions.md)).
+
+> **Read the code**
+> - [`simulator/virtual_drone/broadcast.py`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/simulator/virtual_drone/broadcast.py#L1-L45) — the 45-byte pose frame, position quantized on the wire only
+> - [`prototype/src/sim.ts`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/prototype/src/sim.ts#L60-L63) — `POSE_BYTES = 45` drives the volume estimate
+> - [`prototype/src/sim.ts`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/prototype/src/sim.ts#L260-L277) — pose broadcasts at 5 Hz and opportunistic bulk sync, made visible
+
+## 6. Security as nested layers, not one wall
+
+The assumption was an ad-hoc wireless mesh. Broadcasting positions in the clear,
+even over encrypted Wi-Fi, is not acceptable: anyone who joins can reconstruct
+trajectories over time, and wireless encryption has been broken by ordinary
+people for two decades. Relying on that layer alone is risky, and pushing heavy
+encryption there is wasteful. So the trust model is [layered](05-network-security.md):
+
+1. Wireless transport (WPA class) as the outer shell.
+2. **WireGuard over UDP** as the mesh overlay: proven, peer-authenticated by
+   static keys, cheap.
+3. **SSH inside the tunnel** as the application layer, instead of HTTP/HTTPS. SSH
+   is one of the most tested remote-access technologies we have. HTTPS would pull
+   in a certificate authority and extra parts, and it is the first surface a
+   security researcher probes. [Fewer moving parts, on purpose](adr/ADR-0005-wireguard-beneath-ssh.md).
+
+Identity and encryption solve together at provisioning time: a per-flight,
+per-unit key pair recorded in a flight registry. Each unit holds its own key pair
+plus the public keys of its peers. A separate key registry for SSH means that if
+one WireGuard key leaks, SSH access does not follow automatically.
+
+> **Read the code**
+> - [`infra/ansible/drone-provision.yml`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/infra/ansible/drone-provision.yml#L20-L70) — ed25519 identity, peers authorized only through a forced command, WireGuard brought up
+> - [05 — Network & security](05-network-security.md) and [ADR-0005](adr/ADR-0005-wireguard-beneath-ssh.md)
+
+## 7. SQL as the contract
+
+Beyond the broadcast, people and peers want richer access. The whole progression
+is familiar: fully dynamic APIs, then JSON, OpenAPI/Swagger, GraphQL, then heavier
+schemes. Each time the same pain: hard to debug in the moment, and a lot of
+infrastructure to support, test, and agree on. The simpler conclusion:
+[**SQL is the contract**](adr/ADR-0004-sql-over-ssh-contract.md). Everyone who
+touches data already speaks SQL, and we do not know in advance which slice each
+of them needs, so give them a declarative, extensible way in rather than guessing
+endpoints ahead of time.
+
+Concretely: an SSH connection whose forced command runs a read-only **DuckDB**
+query directly. No extra auth handshake between the engine and the data, the way
+a long-running database needs. No new port or protocol. When access control is
+needed, it rides on Linux file permissions, because the data is discrete
+Hive-partitioned files and permissions can be set per file. Not new invention:
+coming back to foundations DevOps and sysadmins already trust.
+
+> **Read the code**
+> - [`simulator/explorer/server.py`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/simulator/explorer/server.py#L43-L53) — the read-only gate that rejects anything that writes
+> - [`simulator/explorer/server.py`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/simulator/explorer/server.py#L111-L131) — the query path a peer or an engineer hits identically
+> - [`simulator/tests/test_sql_gate.py`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/simulator/tests/test_sql_gate.py#L13-L35) — the gate is safety-critical, so it is tested first
+
+## 8. Replication and offload
+
+Moving accumulated data uses rsync-class tooling (rclone) over the same SSH
+channel. It transfers diffs of the Hive tree efficiently. This matters because a
+unit can be lost, and if it is, we want its data to already exist elsewhere.
+When peers have a stable link the partition tree is pulled between them, which
+gives redundancy and a way to cross-check later. On the ground, when the fleet
+returns, the same path unifies every unit's data into a
+[local warehouse](06-environments.md). The warehouse is read-mostly for
+analytics, so transaction contention is not a concern, which opens a
+DuckDB / DuckLake approach with MinIO underneath, the
+[same storage model on both ends](adr/ADR-0002-one-storage-format.md). One
+uniform structure is what lets an analyst trust a single picture even when a
+unit's data has a gap.
+
+> **Read the code**
+> - [`infra/terraform/ground/main.tf`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/infra/terraform/ground/main.tf#L124-L160) — the post-flight T1 to T3 offload CronJob
+> - [06 — Environments](06-environments.md) and [ADR-0002](adr/ADR-0002-one-storage-format.md)
+
+## 9. A proof of concept, not slides
+
+A concept can be handed over and be wrong. Instead there is a generator: a
+virtual drone producing real sensor data into the Hive layout, a small frontend
+to watch it fill, and a **Grafana** dashboard on top, not to get ahead of the
+analysts, but to see the infrastructure work end to end. Closing the loop
+(Terraform, connector, exporter, dashboard) is how a misconfiguration shows up
+early. A panel that renders wrong points at exactly which layer broke, before an
+analyst has to report that the infra is up but nothing shows. Better to exercise
+every layer now and drop any layer or technology later if there is a sound
+argument against it.
+
+Watching that loop is also where the [CPU cost surfaced](adr/ADR-0009-bounded-lake-scans.md):
+scanning the whole lake on every query does not scale, so the scan is
+[bounded to a flight window and cached](07-observability.md).
+
+> **Read the code**
+> - [`simulator/monitoring/exporter.py`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/simulator/monitoring/exporter.py#L38-L107) — the DuckDB scan that turns Parquet into Prometheus metrics
+> - [`simulator/monitoring/lake.py`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/simulator/monitoring/lake.py#L13-L33) — the flight-window scan that keeps CPU bounded as the lake grows
+> - [`simulator/monitoring/grafana/dashboards/`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/simulator/monitoring/grafana/dashboards) — the fleet and platform dashboards
+> - [`.github/workflows/release.yaml`](https://git.produktor.io/eSlider/swarm-house/src/branch/main/.github/workflows/release.yaml) — tests, smoke flight, Trivy, semantic release
+
+## 10. Working method
+
+A note on how this is meant to be built, not just what. Working together needs a
+shared language, and when the working language is nobody's first language,
+answering live is expensive. The form that scales is written-first: lay the idea
+down carefully, let others read and answer in their own time, keep the history. A
+new colleague, or an agent, can then load the full context of why a decision was
+made by reading the discussion. That is exactly why the decisions here are
+[ADRs](adr/README.md) and why the [contribution flow](../CONTRIBUTING.md) is
+test-first: the reasoning is captured where the next person will look for it.
+
+The mental image is a city planner. Draw perfectly straight roads and people
+still cross the grass where it is faster. The job is not to decide for them, but
+to understand the real paths, the pain points and limits of each team, and pave
+those. So constraints get checked with the other DevOps, data engineers, and
+analysts before infrastructure is laid, instead of building roads nobody uses.
+
+---
+
+The platform is air-gapped and on-prem by design: nothing leaves the swarm, and
+nothing joins it at runtime. Start from the [README](../README.md) for the map,
+or the [ADR index](adr/README.md) for the decisions in full.

docs/adr/ADR-0001-no-swarm-orchestrator.md

@@ -0,0 +1,38 @@
+# ADR-0001: No swarm-wide orchestrator; coordinate through data
+
+## Status
+
+Accepted (2026-07-08)
+
+## Context
+
+The fleet flies fully autonomously with no internet uplink and only
+intermittent mesh connectivity between drones. A conventional control plane
+(Kubernetes across the swarm, a leader election, a central scheduler) assumes
+stable membership and low-latency links — exactly what a swarm does not have.
+Partitions are the normal case, not the exception.
+
+## Options
+
+| Option | Pros | Cons |
+| --- | --- | --- |
+| A — Cluster orchestrator spanning the swarm | Familiar tooling | Assumes stable membership; a partition stalls coordination; single point of failure in the air |
+| B — Each drone autonomous, coordinating through exchanged data | Survives partitions; no air-side control plane to fail | No global view; behaviour must be derivable from local state |
+
+## Decision
+
+Every drone is an autonomous node. There is **no orchestrator spanning the
+swarm**. Coordination happens by exchanging derived state (see
+[ADR-0003](ADR-0003-sync-derived-state-only.md)), and each drone makes its own
+in-flight decisions against its local data window. Orchestration tooling
+(Kubernetes, Flux) is confined to the **ground** segment
+([ADR-0006](ADR-0006-iac-boundaries.md)).
+
+## Consequences
+
+- The platform degrades gracefully under partition: a drone that loses peers
+  keeps flying and recording.
+- There is no single "fleet state" to query in the air; health questions are
+  answered from each drone's own store and reconciled on the ground.
+- Design principle 1 in [`../../README.md`](../../README.md) and the on-board
+  model in [`../02-architecture.md`](../02-architecture.md) follow from this.

docs/adr/ADR-0002-one-storage-format.md

@@ -0,0 +1,32 @@
+# ADR-0002: One storage format on every tier — Parquet + DuckDB, Hive layout
+
+## Status
+
+Accepted (2026-07-08)
+
+## Context
+
+Data lives in three places: on the drone's NVMe in flight (T1), in the ground
+warehouse after landing (T3), and in the dev/simulation environment. If each
+tier used a different format or layout, offload would need transformation code,
+schemas would drift, and a query proven on the bench would not run unchanged on
+flight data.
+
+## Decision
+
+Use **one storage format on every floor**: columnar **Parquet** files in an
+identical **Hive-partitioned** layout (`dataset=…/flight=…/drone=…/…`), queried
+with **DuckDB**. Flight offload T1 → T3 is therefore a plain **mirror**
+(copy/rsync of partition directories), not an ETL step.
+
+## Consequences
+
+- Offload is a file operation, testable with a checksum; no format converters
+  to maintain or version.
+- The same SQL runs against T1, T3, and simulated data — the simulator
+  ([`../../simulator/`](../../simulator/)) generates the *real* layout, so tests
+  exercise the production format.
+- Partitioning choices are load-bearing; changing them is a schema migration.
+  Detailed in [`../03-data-platform.md`](../03-data-platform.md).
+- DuckDB's single-file, no-server model fits an air-gapped drone with no room
+  for a database daemon.

docs/adr/ADR-0003-sync-derived-state-only.md

@@ -0,0 +1,33 @@
+# ADR-0003: Sync derived state only; raw telemetry stays local
+
+## Status
+
+Accepted (2026-07-08)
+
+## Context
+
+Each drone produces high-rate sensor telemetry and on-board video detections —
+far more than the mesh can carry. But peers only need enough to coordinate
+flight: where everyone is, where they are heading, what was detected. Pushing
+raw feeds across the air would saturate the radio and drain batteries for data
+nobody reads in flight.
+
+## Decision
+
+Only **derived state** crosses the air — position, attitude, detections — as
+small pub/sub broadcasts (~5 Hz). **Raw telemetry stays on local NVMe** until
+the drone lands, then travels to the warehouse during offload
+([ADR-0002](ADR-0002-one-storage-format.md)). The design is **event-driven**:
+new derived data triggers downstream action through hooks; nothing polls asking
+"anything new yet?".
+
+## Consequences
+
+- Air bandwidth scales with the number of peers and the broadcast rate, not
+  with sensor resolution.
+- The warehouse is the only place with the full picture; in-flight decisions use
+  the local window plus peer broadcasts.
+- Broadcast staleness per peer becomes a first-class health signal
+  ([`../07-observability.md`](../07-observability.md)).
+- What syncs and what does not is specified in
+  [`../04-swarm-sync.md`](../04-swarm-sync.md).

docs/adr/ADR-0004-sql-over-ssh-contract.md

@@ -0,0 +1,46 @@
+# ADR-0004: Read-only SQL-over-SSH as the peer query contract; no debug API
+
+## Status
+
+Accepted (2026-07-08). Supersedes the earlier sketch of a bespoke query service.
+
+## Context
+
+Beyond the 5 Hz broadcast, a peer (or an engineer on the bench) sometimes needs
+to *ask* a drone a richer question: "show me your detections in the last minute
+near this position". Inventing a service for this means a new port, a new
+protocol, an auth layer, and a second code path that only runs during
+debugging — the classic "works on the bench, untested in flight" trap.
+
+## Options
+
+| Option | Pros | Cons |
+| --- | --- | --- |
+| A — Custom query microservice + REST API | Flexible | New port/protocol/auth to secure; a debug-only path that never flies |
+| B — Read-only DuckDB SQL over an SSH forced command | Reuses SSH trust + the query language already on both ends; identical path on bench and in flight | SQL must be gated to read-only |
+
+## Decision
+
+Peer data access is **read-only DuckDB SQL executed through an SSH forced
+command**. SQL is already the query language on both ends
+([ADR-0002](ADR-0002-one-storage-format.md)), and SSH already carries the trust
+model ([ADR-0005](ADR-0005-wireguard-beneath-ssh.md)), so no new service, port,
+or protocol is invented. A **SQL gate** rejects anything that writes
+(`COPY`, `INSERT`, `ATTACH`, pragmas). There is **no separate debug API**: an
+engineer debugging on the ground runs the identical query through the identical
+wrapper, permissions, and output format a peer drone would use.
+
+## Consequences
+
+- One access path is designed for flight and bench — "what you test is what
+  flies".
+- The statement gate in the PoC (`simulator/explorer/server.py`, covered by
+  `simulator/tests/test_sql_gate.py`) is a **first layer**: keyword allow/deny
+  over HTTP for the explorer. Production still needs the remaining layers in
+  [04 — Swarm sync](../04-swarm-sync.md) (forced-command key, read-only OS user,
+  engine hardening, resource caps) and a real SQL parser rather than keywords
+  alone.
+- The explorer and the prototype's live mode are *stand-in consumers* of the
+  same read-only contract over HTTP today; the proposed flight path is
+  SQL-over-SSH ([`../04-swarm-sync.md`](../04-swarm-sync.md)).
+- Design principles 7 and 8 in [`../../README.md`](../../README.md) restate this.

docs/adr/ADR-0005-wireguard-beneath-ssh.md

@@ -0,0 +1,40 @@
+# ADR-0005: WireGuard beneath SSH for the mesh transport
+
+## Status
+
+Accepted (2026-07-08)
+
+## Context
+
+Peer bulk sync and SQL-over-SSH ([ADR-0004](ADR-0004-sql-over-ssh-contract.md))
+need an encrypted, authenticated channel across an ad-hoc radio mesh. SSH alone
+would work at the application layer, but it leaves the drones' listening
+surface (sshd, any other service port) exposed on the raw mesh network, and it
+gives no cheap, uniform way to authenticate *the peer* before the application
+handshake.
+
+## Options
+
+| Option | Pros | Cons |
+| --- | --- | --- |
+| A — SSH only | One daemon, familiar | Service ports exposed on the raw mesh; per-service auth; no network-layer peer identity |
+| B — WireGuard tunnel, SSH inside it | Network-layer peer auth via static keys; only the WG port is exposed; SSH speaks over a private, encrypted subnet | Two layers to provision |
+
+## Decision
+
+Run **WireGuard as the network layer** and **SSH on top of it**. WireGuard
+authenticates peers by pre-shared static public keys and presents a private
+encrypted subnet; SSH forced commands then carry the SQL and bulk-sync contract
+over that subnet. Keys are issued **per device before deployment** — nothing
+joins the mesh at runtime (design principle 5).
+
+## Consequences
+
+- The only thing exposed on the raw radio is the WireGuard port; sshd and the
+  data services listen only on the WG interface.
+- Peer identity is established once, at the network layer, and reused by every
+  application above it.
+- Provisioning must place both WG and SSH keys during host prep — handled by
+  `infra/ansible/drone-provision.yml` ([ADR-0006](ADR-0006-iac-boundaries.md)).
+- Rationale and the "why not SSH alone" argument live in
+  [`../05-network-security.md`](../05-network-security.md).

docs/adr/ADR-0006-iac-boundaries.md

@@ -0,0 +1,41 @@
+# ADR-0006: IaC split — Ansible hosts, Terraform ground, Flux ground-only, fleet manifest for drones
+
+## Status
+
+Accepted (2026-07-08)
+
+## Context
+
+Three very different things need provisioning: (1) drone hosts, which are
+air-gapped and never reachable by a controller in flight; (2) the ground
+segment (warehouse, observability, offload), a normal k3s cluster; (3) the
+simulation environment that mirrors the ground segment locally. Using one tool
+for all three would force a GitOps controller or a Terraform apply loop onto the
+drone — impossible for an autonomous, disconnected node
+([ADR-0001](ADR-0001-no-swarm-orchestrator.md)).
+
+## Decision
+
+Split infrastructure ownership by lifecycle:
+
+| Layer | Tool | Scope |
+| --- | --- | --- |
+| Host preparation | **Ansible** | Drone bench prep (keys, forced commands, WireGuard, Compose bundle) and the local k3d sim cluster |
+| Ground workloads | **Terraform** | k3s/k3d resources: virtual fleet + observability (`sim-env`), warehouse + T1→T3 offload (`ground`) |
+| Ground continuous delivery | **Flux (GitOps)** | Ground segment **only** — policy labels, dashboard bundles |
+| Drone delivery | **Fleet release manifest** | A versioned, digest-pinned artifact list; no controller pulls to the drone |
+
+Drones are delivered by the **fleet release manifest** (one version for the
+whole fleet, atomic rollback), never by a controller reaching into the air.
+
+## Consequences
+
+- No GitOps agent or `terraform apply` ever targets a drone; the air side stays
+  controller-free.
+- Terraform state is environment-local and kept out of Git
+  ([ADR-0008](ADR-0008-runtime-data-paths.md)).
+- The sim cluster is a faithful miniature: the same Terraform describes it and
+  the ground segment.
+- Boundaries are documented in
+  [`../06-environments.md`](../06-environments.md#iac-boundaries); delivery in
+  [`../11-cicd-delivery.md`](../11-cicd-delivery.md).

docs/adr/ADR-0007-non-root-image.md

@@ -0,0 +1,27 @@
+# ADR-0007: Run the simulator image as a non-root user
+
+## Status
+
+Accepted (2026-07-08)
+
+## Context
+
+The Trivy scan in CI flagged **DS-0002**: the simulator image ran as `root`
+because no `USER` was set. A container writing the data lake as root also
+produces root-owned Parquet on bind mounts, which then blocks a later non-root
+process from the same files.
+
+## Decision
+
+Create a dedicated unprivileged user **`swarm` (uid/gid 10001)** in the
+`simulator/Dockerfile` and switch to it with `USER swarm` before the entrypoint.
+The uid is fixed (not auto-assigned) so host-side ownership of the shared data
+paths ([ADR-0008](ADR-0008-runtime-data-paths.md)) is deterministic.
+
+## Consequences
+
+- Trivy DS-0002 clears; the image follows least-privilege.
+- Bind-mounted data is owned by `10001:10001`; host prep (`sim-cluster.yml`) and
+  Ansible `chown` the `var/t1` / `var/t3` paths to this uid so k3d pods can write.
+- Any future service image reusing this data must run as the same uid or share
+  the group.

docs/adr/ADR-0008-runtime-data-paths.md

@@ -0,0 +1,40 @@
+# ADR-0008: Shared `var/t1` / `var/t3` runtime paths; state out of Git
+
+## Status
+
+Accepted (2026-07-08)
+
+## Context
+
+Compose, the k3d fleet, and the ground offload job all need to read and write
+the same data lake, so that "go live" in the prototype shows the data the
+simulator just generated. Early on, Terraform state and the provider cache were
+accidentally committed (~6 MB of `*.tfstate` and a vendored provider binary),
+polluting history and risking stale/secret data in Git.
+
+## Decision
+
+Standardise two **gitignored** runtime directories at the repo root:
+
+| Path | Tier | Contents |
+| --- | --- | --- |
+| `var/t1/` | T1 live lake | In-flight Parquet from Compose or the k3d fleet |
+| `var/t3/` | T3 warehouse | Offloaded historical partitions |
+
+They are created automatically and overridable via
+`SWARM_T1_DIR` / `SWARM_WAREHOUSE_DATA` / `SWARM_SIM_DATA`. Treat `var/` as
+**runtime-only** per the ecosystem canonical layout. `.gitignore` excludes
+`*.tfstate`, `*.tfstate.backup`, and `.terraform/` while **keeping**
+`.terraform.lock.hcl` (the lock is source, the state is not).
+
+## Consequences
+
+- One lake feeds every component; the live prototype reflects real generated
+  data.
+- Git history carries no machine state or large binaries; `terraform` treats
+  its state as local/remote-backend concern, not versioned source.
+- Ownership of these paths is fixed to uid 10001
+  ([ADR-0007](ADR-0007-non-root-image.md)).
+- Recovery from earlier state drift is documented in
+  [`../../infra/terraform/README.md`](../../infra/terraform/README.md)
+  (`terraform import`).

docs/adr/ADR-0009-bounded-lake-scans.md

@@ -0,0 +1,52 @@
+# ADR-0009: Bound observability lake scans to a flight window and cache
+
+## Status
+
+Accepted (2026-07-08)
+
+## Context
+
+The k3d sim node was running at ~245% CPU. Three causes compounded:
+
+1. **Explorer** rebuilt a DuckDB view over the *entire* lake on every HTTP
+   request (~2000 Parquet files, 200 MB+) — roughly 1.5 cores.
+2. **Exporter** scanned the full lake every 5 s — roughly 0.7 cores.
+3. **Drones** exited after sealing a flight; Kubernetes restarted each pod,
+   and every restart appended a fresh `flight=` partition tree, so the lake grew
+   without bound and every scan got more expensive.
+
+The cost is inherent to scanning an append-only lake whose size grows with pod
+churn, not to any single slow query. Metrics and the live view only ever need
+recent flights.
+
+## Decision
+
+Bound and cache the scans instead of reading the whole lake:
+
+- **Flight window.** The exporter and explorer scan only the most recent
+  `METRIC_FLIGHT_WINDOW` flights (default 5) via a shared helper
+  (`simulator/monitoring/lake.py`). Older partitions stay on disk but out of the
+  hot path.
+- **Caching.** Explorer caches DuckDB views (`VIEW_REFRESH_S=30`) and the
+  partition tree (`TREE_CACHE_S=15`); the exporter scans every
+  `SCAN_INTERVAL_S=30` instead of 5 s.
+- **Stop restart churn.** Drones set `KEEP_ALIVE=1` and idle after sealing
+  instead of exiting, so no new flight tree is created per restart. In k3d,
+  `IMU_HZ=20` and resource limits cap per-drone load; the live prototype polls
+  every 2 s.
+- **Self-observability.** Add node-exporter, a
+  `swarm_exporter_scan_duration_seconds` metric, and a **Swarm Platform**
+  Grafana dashboard (node CPU/memory + scan health) alongside the existing fleet
+  dashboard.
+
+## Consequences
+
+- Steady state dropped to explorer ~29 m, exporter ~16 m CPU; the k3d node to
+  ~56% (from ~245%).
+- Metrics reflect recent flights only; long-horizon analysis is a warehouse
+  (T3) query, not an exporter concern — consistent with
+  [ADR-0003](ADR-0003-sync-derived-state-only.md).
+- Old flight partitions accumulate on disk; pruning them is a follow-up
+  (tracked in [`../09-open-questions.md`](../09-open-questions.md)).
+- Observability design is documented in
+  [`../07-observability.md`](../07-observability.md).

docs/adr/README.md

@@ -0,0 +1,40 @@
+# Architecture Decision Records
+
+This directory records the significant decisions behind Swarm House, in the
+order they were made. Each record is immutable once **Accepted** — a later
+decision that changes course gets a new number and supersedes the old one
+rather than editing it.
+
+## ADR vs ASR
+
+| Kind | What it captures | Where it lives |
+| --- | --- | --- |
+| **ADR** (Architecture Decision Record) | A point-in-time decision for *this* repository, with the options considered and their consequences | Here, in `docs/adr/` |
+| **ASR** (Architecture Standard Record) | A standing, cross-repository policy every project must obey (Go-first, secrets, layout) | The ecosystem `inventar` repo, not here |
+
+The rules in [`../../AGENTS.md`](../../AGENTS.md) are this repo's local standing
+constraints; anything ecosystem-wide belongs in `inventar` as an ASR. Design
+decisions specific to the platform are ADRs and belong here.
+
+## Index
+
+| ADR | Decision | Status |
+| --- | --- | --- |
+| [ADR-0001](ADR-0001-no-swarm-orchestrator.md) | No swarm-wide orchestrator; coordinate through data | Accepted |
+| [ADR-0002](ADR-0002-one-storage-format.md) | One storage format on every tier — Parquet + DuckDB, Hive layout | Accepted |
+| [ADR-0003](ADR-0003-sync-derived-state-only.md) | Sync derived state only; raw telemetry stays local | Accepted |
+| [ADR-0004](ADR-0004-sql-over-ssh-contract.md) | Read-only SQL-over-SSH as the peer query contract; no debug API | Accepted |
+| [ADR-0005](ADR-0005-wireguard-beneath-ssh.md) | WireGuard beneath SSH for the mesh transport | Accepted |
+| [ADR-0006](ADR-0006-iac-boundaries.md) | IaC split: Ansible hosts, Terraform ground, Flux ground-only, fleet manifest for drones | Accepted |
+| [ADR-0007](ADR-0007-non-root-image.md) | Run the simulator image as a non-root user | Accepted |
+| [ADR-0008](ADR-0008-runtime-data-paths.md) | Shared `var/t1` / `var/t3` runtime paths; state out of Git | Accepted |
+| [ADR-0009](ADR-0009-bounded-lake-scans.md) | Bound observability lake scans to a flight window and cache | Accepted |
+
+## Writing a new ADR
+
+1. Copy [`TEMPLATE.md`](TEMPLATE.md) to `ADR-NNNN-short-slug.md` (next free number).
+2. Fill in Context, Options (if more than one was real), Decision, Consequences.
+3. Add a row to the index above.
+4. Cross-link the ADR from the design doc it affects (and vice versa).
+
+Keep the tone laconic: state the decision and why, skip the narrative.

docs/adr/TEMPLATE.md

@@ -0,0 +1,29 @@
+# ADR-NNNN: Short decision title
+
+## Status
+
+Proposed | Accepted | Superseded by ADR-XXXX
+
+## Context
+
+What forces are in play? What problem or constraint triggered the decision?
+Keep it to the facts that actually shaped the choice.
+
+## Options
+
+| Option | Pros | Cons |
+| --- | --- | --- |
+| A — … | … | … |
+| B — … | … | … |
+
+(Drop this section when only one option was ever real.)
+
+## Decision
+
+The choice, stated plainly. One paragraph.
+
+## Consequences
+
+- What becomes easier.
+- What becomes harder or is now forbidden.
+- Follow-ups, links to affected docs.

infra/ansible/sim-cluster.yml

@@ -29,12 +29,16 @@
         path: "{{ data_dir }}"
         state: directory
         mode: "0777"
+        owner: "10001"
+        group: "10001"
 
     - name: Create warehouse directory (T3 host path inside the k3d node)
       ansible.builtin.file:
         path: "{{ warehouse_dir }}"
         state: directory
         mode: "0777"
+        owner: "10001"
+        group: "10001"
 
     - name: List existing clusters
       ansible.builtin.command: k3d cluster list -o json

infra/terraform/sim-env/main.tf

@@ -63,6 +63,14 @@ resource "kubernetes_stateful_set" "drone" {
             name  = "DATA_DIR"
             value = "/data"
           }
+          env {
+            name  = "KEEP_ALIVE"
+            value = "1"
+          }
+          env {
+            name  = "IMU_HZ"
+            value = "20"
+          }
 
           volume_mount {
             name       = "lake"
@@ -118,9 +126,21 @@ resource "kubernetes_deployment" "exporter" {
             name  = "EXPORTER_PORT"
             value = "9105"
           }
+          env {
+            name  = "SCAN_INTERVAL_S"
+            value = "30"
+          }
+          env {
+            name  = "METRIC_FLIGHT_WINDOW"
+            value = "5"
+          }
           port {
             container_port = 9105
           }
+          resources {
+            requests = { cpu = "50m", memory = "64Mi" }
+            limits   = { cpu = "500m", memory = "256Mi" }
+          }
           volume_mount {
             name       = "lake"
             mount_path = "/data"
@@ -182,9 +202,25 @@ resource "kubernetes_deployment" "explorer" {
             name  = "EXPLORER_PORT"
             value = "8088"
           }
+          env {
+            name  = "VIEW_REFRESH_S"
+            value = "30"
+          }
+          env {
+            name  = "TREE_CACHE_S"
+            value = "15"
+          }
+          env {
+            name  = "METRIC_FLIGHT_WINDOW"
+            value = "5"
+          }
           port {
             container_port = 8088
           }
+          resources {
+            requests = { cpu = "50m", memory = "64Mi" }
+            limits   = { cpu = "750m", memory = "256Mi" }
+          }
           volume_mount {
             name       = "lake"
             mount_path = "/data"
@@ -226,11 +262,17 @@ resource "kubernetes_config_map" "prometheus" {
   data = {
     "prometheus.yml" = <<-EOT
       global:
-        scrape_interval: 5s
+        scrape_interval: 15s
       scrape_configs:
         - job_name: swarm
           static_configs:
             - targets: ["exporter:9105"]
+        - job_name: explorer
+          static_configs:
+            - targets: ["explorer:8088"]
+        - job_name: node
+          static_configs:
+            - targets: ["node-exporter:9100"]
     EOT
   }
 }
@@ -298,8 +340,12 @@ resource "kubernetes_config_map" "grafana_provisioning" {
   data = {
     "datasource.yml" = <<-EOT
       apiVersion: 1
+      deleteDatasources:
+        - name: Prometheus
+          orgId: 1
       datasources:
         - name: Prometheus
+          uid: swarm-prom
           type: prometheus
           access: proxy
           url: http://prometheus:9090
@@ -324,6 +370,90 @@ resource "kubernetes_config_map" "grafana_dashboard" {
   }
   data = {
     "swarm.json"          = file("${path.module}/../../../simulator/monitoring/grafana/dashboards/swarm.json")
+    "swarm-platform.json" = file("${path.module}/../../../simulator/monitoring/grafana/dashboards/swarm-platform.json")
+  }
+}
+
+# Host metrics for the k3d node (CPU / memory on the platform dashboard)
+resource "kubernetes_deployment" "node_exporter" {
+  metadata {
+    name      = "node-exporter"
+    namespace = kubernetes_namespace.swarm.metadata[0].name
+    labels    = local.labels
+  }
+
+  spec {
+    replicas = 1
+    selector {
+      match_labels = { app = "node-exporter" }
+    }
+    template {
+      metadata {
+        labels = merge(local.labels, { app = "node-exporter" })
+      }
+      spec {
+        host_network = true
+        host_pid     = true
+        container {
+          name  = "node-exporter"
+          image = "prom/node-exporter:v1.8.2"
+          args = [
+            "--path.procfs=/host/proc",
+            "--path.sysfs=/host/sys",
+            "--path.rootfs=/host/root",
+            "--collector.filesystem.mount-points-exclude=^/(sys|proc|dev|host|etc)($$|/)",
+          ]
+          port {
+            container_port = 9100
+          }
+          volume_mount {
+            name       = "proc"
+            mount_path = "/host/proc"
+            read_only  = true
+          }
+          volume_mount {
+            name       = "sys"
+            mount_path = "/host/sys"
+            read_only  = true
+          }
+          volume_mount {
+            name       = "root"
+            mount_path = "/host/root"
+            read_only  = true
+          }
+          resources {
+            requests = { cpu = "20m", memory = "32Mi" }
+            limits   = { cpu = "200m", memory = "64Mi" }
+          }
+        }
+        volume {
+          name = "proc"
+          host_path { path = "/proc" }
+        }
+        volume {
+          name = "sys"
+          host_path { path = "/sys" }
+        }
+        volume {
+          name = "root"
+          host_path { path = "/" }
+        }
+      }
+    }
+  }
+}
+
+resource "kubernetes_service" "node_exporter" {
+  metadata {
+    name      = "node-exporter"
+    namespace = kubernetes_namespace.swarm.metadata[0].name
+  }
+  spec {
+    selector = { app = "node-exporter" }
+    port {
+      port        = 9100
+      target_port = 9100
+    }
   }
 }
 

prototype/src/App.tsx

@@ -25,7 +25,7 @@ function displayAspect(): number {
   return Math.max(1, Math.min(MAX_ASPECT, q));
 }
 
-const LIVE_POLL_MS = 1000;
+const LIVE_POLL_MS = 2000;
 
 export default function App(): JSX.Element {
   const [droneCount, setDroneCount] = useState(8);
@@ -268,7 +268,7 @@ export default function App(): JSX.Element {
             {busiest.length === 0 && <div style={styles.dim}>no links in range</div>}
           </Section>
           <Section title="Legend">
-            <div style={styles.dim}>blue line — pose broadcast (5 Hz, ~46 B)</div>
+            <div style={styles.dim}>blue line — pose broadcast (5 Hz, 45 B)</div>
             <div style={styles.dim}>green line — bulk sync (sealed partitions)</div>
             <div style={styles.dim}>flash — broadcast event delivered</div>
             <div style={styles.dim}>red circle — transit object crossing the area</div>

prototype/src/live.ts

@@ -20,24 +20,24 @@ export interface LivePose {
 // latest battery reading from telemetry. arg_max keeps it a single scan.
 const LIVE_SQL = `
 WITH pose AS (
-  SELECT drone_id,
+  SELECT drone,
          arg_max(pos_x, ts_ns) AS x,
          arg_max(pos_y, ts_ns) AS y,
          arg_max(yaw, ts_ns)   AS yaw,
          max(ts_ns)            AS ts_ns
   FROM state
   WHERE direction = 'sent'
-  GROUP BY drone_id
+  GROUP BY drone
 ),
 batt AS (
-  SELECT drone_id, arg_max(level_pct, ts_ns) AS battery
+  SELECT drone, arg_max(level_pct, ts_ns) AS battery
   FROM telemetry
   WHERE sensor = 'battery'
-  GROUP BY drone_id
+  GROUP BY drone
 )
-SELECT p.drone_id, p.x, p.y, p.yaw, p.ts_ns, b.battery
-FROM pose p LEFT JOIN batt b USING (drone_id)
-ORDER BY p.drone_id`;
+SELECT p.drone, p.x, p.y, p.yaw, p.ts_ns, b.battery
+FROM pose p LEFT JOIN batt b USING (drone)
+ORDER BY p.drone`;
 
 interface QueryResponse {
   columns?: string[];

prototype/src/sim.ts

@@ -57,7 +57,7 @@ export function areaWidth(): number {
 export const LINK_RANGE = 320;
 const CRUISE = 28;
 const SEPARATION = 55;
-const POSE_BYTES = 46;
+const POSE_BYTES = 45;
 const POSE_HZ = 5;
 // Steady per-link broadcast throughput (both directions), bytes/s
 export const POSE_RATE_BYTES = POSE_BYTES * POSE_HZ;

simulator/Dockerfile

@@ -7,6 +7,10 @@ COPY virtual_drone ./virtual_drone
 COPY monitoring ./monitoring
 COPY explorer ./explorer
 COPY tests ./tests
+RUN groupadd --gid 10001 swarm \
+    && useradd --uid 10001 --gid swarm --home-dir /app --shell /usr/sbin/nologin swarm \
+    && chown -R swarm:swarm /app
+USER swarm
 RUN pytest tests/ -q
 
 FROM python:3.12-slim
@@ -19,6 +23,11 @@ COPY virtual_drone ./virtual_drone
 # them without bind mounts (Compose overrides these with live mounts)
 COPY monitoring ./monitoring
 COPY explorer ./explorer
+RUN groupadd --gid 10001 swarm \
+    && useradd --uid 10001 --gid swarm --home-dir /app --shell /usr/sbin/nologin swarm \
+    && mkdir -p /data \
+    && chown -R swarm:swarm /app /data
 
 ENV DATA_DIR=/data
+USER swarm
 CMD ["python", "-m", "virtual_drone.main"]

simulator/docker-compose.yml

@@ -24,7 +24,8 @@ services:
     command: ["python", "monitoring/exporter.py"]
     environment:
       DATA_DIR: /data
-      SCAN_INTERVAL_S: "5"
+      SCAN_INTERVAL_S: "30"
+      METRIC_FLIGHT_WINDOW: "5"
     volumes:
       - ${SWARM_T1_DIR:-../var/t1}:/data:ro
       - ./monitoring:/app/monitoring:ro
@@ -37,6 +38,9 @@ services:
     command: ["python", "explorer/server.py"]
     environment:
       DATA_DIR: /data
+      VIEW_REFRESH_S: "30"
+      TREE_CACHE_S: "15"
+      METRIC_FLIGHT_WINDOW: "5"
     volumes:
       - ${SWARM_T1_DIR:-../var/t1}:/data:ro
       - ./explorer:/app/explorer:ro

simulator/explorer/server.py

@@ -1,33 +1,28 @@
-"""Data-plane explorer: a web view over the Hive-partitioned Parquet lake.
-
-Serves three things:
-  /            single-page UI (partition tree + read-only SQL console)
-  /api/tree    partition hierarchy with file counts and bytes, live
-  /api/query   gated read-only DuckDB SQL, same statement rules as the
-               peer query channel (SELECT/WITH only, single statement)
-
-The point is doctrinal, not just convenient: the explorer reuses the exact
-read-only SQL contract that drones expose to each other, so "looking at the
-data plane" on the bench exercises the same path a peer would use in flight.
-"""
+"""Data-plane explorer: a web view over the Hive-partitioned Parquet lake."""
 
 from __future__ import annotations
 
 import json
 import os
 import re
+import sys
+import threading
+import time
 from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
 from pathlib import Path
 
 import duckdb
 
+sys.path.insert(0, str(Path(__file__).resolve().parents[1] / "monitoring"))
+from lake import flight_window, iter_parquet_files, parquet_reader  # noqa: E402
+
 DATA_DIR = Path(os.environ.get("DATA_DIR", "./data"))
 PORT = int(os.environ.get("EXPLORER_PORT", "8088"))
 ROW_LIMIT = int(os.environ.get("ROW_LIMIT", "500"))
+VIEW_REFRESH_S = float(os.environ.get("VIEW_REFRESH_S", "30"))
+TREE_CACHE_S = float(os.environ.get("TREE_CACHE_S", "15"))
 STATIC_DIR = Path(__file__).parent
 
-# Same spirit as the forced-command gate on a real drone: one statement,
-# must be a read, no statement that could write, configure, or reach out.
 _ALLOWED_START = re.compile(r"^\s*(SELECT|WITH|DESCRIBE|SUMMARIZE|SHOW)\b", re.IGNORECASE)
 _FORBIDDEN = re.compile(
     r"\b(INSERT|UPDATE|DELETE|CREATE|DROP|ALTER|ATTACH|DETACH|COPY|EXPORT"
@@ -35,9 +30,17 @@ _FORBIDDEN = re.compile(
     re.IGNORECASE,
 )
 
+_lock = threading.Lock()
+_con: duckdb.DuckDBPyConnection | None = None
+_views_at = 0.0
+_tree_cache: dict | None = None
+_tree_at = 0.0
+_last_tree_s = 0.0
+_last_query_s = 0.0
+_query_count = 0
+
 
 def gate(sql: str) -> str | None:
-    """Return a rejection reason, or None if the statement passes."""
     stripped = re.sub(r"--[^\n]*|/\*.*?\*/", " ", sql, flags=re.DOTALL).strip().rstrip(";")
     if not stripped:
         return "empty statement"
@@ -50,27 +53,39 @@ def gate(sql: str) -> str | None:
     return None
 
 
-def connect() -> duckdb.DuckDBPyConnection:
-    """Fresh connection with the three datasets pre-registered as views."""
+def _refresh_views() -> None:
+    global _con, _views_at
     con = duckdb.connect()
     for ds in ("telemetry", "detections", "state"):
-        pattern = f"{DATA_DIR}/dataset={ds}/**/*.parquet"
+        reader = parquet_reader(DATA_DIR, ds)
         try:
-            con.execute(
-                f"CREATE VIEW {ds} AS SELECT * FROM read_parquet("
-                f"'{pattern}', hive_partitioning=true, union_by_name=true)"
-            )
+            con.execute(f"CREATE OR REPLACE VIEW {ds} AS SELECT * FROM read_parquet({reader})")
         except duckdb.Error:
-            pass  # dataset not written yet; view simply won't exist
-    return con
+            pass
+    with _lock:
+        if _con is not None:
+            _con.close()
+        _con = con
+        _views_at = time.monotonic()
+
+
+def connect() -> duckdb.DuckDBPyConnection:
+    if _con is None or time.monotonic() - _views_at > VIEW_REFRESH_S:
+        _refresh_views()
+    assert _con is not None
+    return _con
 
 
 def tree() -> dict:
-    """Partition hierarchy: dataset -> flight -> drone -> leafs, with sizes."""
+    global _tree_cache, _tree_at, _last_tree_s
+    now = time.monotonic()
+    if _tree_cache is not None and now - _tree_at < TREE_CACHE_S:
+        return _tree_cache
+    started = now
     root: dict = {}
     total_bytes = 0
     total_files = 0
-    for f in sorted(DATA_DIR.rglob("*.parquet")):
+    for f in sorted(iter_parquet_files(DATA_DIR)):
         rel = f.relative_to(DATA_DIR)
         size = f.stat().st_size
         total_bytes += size
@@ -80,7 +95,6 @@ def tree() -> dict:
             node = node.setdefault("children", {}).setdefault(part, {})
         leaf = node.setdefault("children", {}).setdefault(rel.parts[-1], {})
         leaf["bytes"] = size
-        # roll sizes up the tree
         node = root
         node["bytes"] = node.get("bytes", 0) + size
         node["files"] = node.get("files", 0) + 1
@@ -88,18 +102,26 @@ def tree() -> dict:
             node = node["children"][part]
             node["bytes"] = node.get("bytes", 0) + size
             node["files"] = node.get("files", 0) + 1
-    return {"tree": root, "total_bytes": total_bytes, "total_files": total_files}
+    _last_tree_s = time.monotonic() - started
+    _tree_cache = {"tree": root, "total_bytes": total_bytes, "total_files": total_files}
+    _tree_at = now
+    return _tree_cache
 
 
 def run_query(sql: str) -> dict:
+    global _last_query_s, _query_count
     reason = gate(sql)
     if reason:
         return {"error": f"rejected by read-only gate: {reason}"}
+    started = time.monotonic()
+    with _lock:
         con = connect()
         try:
             cur = con.sql(sql)
             columns = cur.columns
             rows = cur.fetchmany(ROW_LIMIT)
+            _query_count += 1
+            _last_query_s = time.monotonic() - started
             return {
                 "columns": columns,
                 "rows": [[repr(v) if isinstance(v, bytes) else v for v in row] for row in rows],
@@ -107,8 +129,23 @@ def run_query(sql: str) -> dict:
             }
         except duckdb.Error as exc:
             return {"error": str(exc)}
-    finally:
-        con.close()
+
+
+def metrics_text() -> str:
+    return "\n".join([
+        "# HELP swarm_explorer_query_duration_seconds Wall time of the last SQL query",
+        "# TYPE swarm_explorer_query_duration_seconds gauge",
+        f"swarm_explorer_query_duration_seconds {_last_query_s:.4f}",
+        "# HELP swarm_explorer_tree_duration_seconds Wall time of the last partition tree build",
+        "# TYPE swarm_explorer_tree_duration_seconds gauge",
+        f"swarm_explorer_tree_duration_seconds {_last_tree_s:.4f}",
+        "# HELP swarm_explorer_queries_total Read-only queries served",
+        "# TYPE swarm_explorer_queries_total counter",
+        f"swarm_explorer_queries_total {_query_count}",
+        "# HELP swarm_metric_flight_window Flight partitions in DuckDB views",
+        "# TYPE swarm_metric_flight_window gauge",
+        f"swarm_metric_flight_window {flight_window()}",
+    ]) + "\n"
 
 
 class Handler(BaseHTTPRequestHandler):
@@ -116,13 +153,11 @@ class Handler(BaseHTTPRequestHandler):
         self.send_response(code)
         self.send_header("Content-Type", ctype)
         self.send_header("Content-Length", str(len(body)))
-        # Dev CORS: lets the prototype's live mode poll the API from another
-        # origin. Everything behind this is read-only by construction.
         self.send_header("Access-Control-Allow-Origin", "*")
         self.end_headers()
         self.wfile.write(body)
 
-    def do_OPTIONS(self) -> None:  # noqa: N802 — http.server API
+    def do_OPTIONS(self) -> None:  # noqa: N802
         self.send_response(204)
         self.send_header("Access-Control-Allow-Origin", "*")
         self.send_header("Access-Control-Allow-Methods", "GET, POST, OPTIONS")
@@ -132,15 +167,17 @@ class Handler(BaseHTTPRequestHandler):
     def _json(self, payload: dict, code: int = 200) -> None:
         self._send(code, json.dumps(payload, default=str).encode(), "application/json")
 
-    def do_GET(self) -> None:  # noqa: N802 — http.server API
+    def do_GET(self) -> None:  # noqa: N802
         if self.path in ("/", "/index.html"):
             self._send(200, (STATIC_DIR / "index.html").read_bytes(), "text/html; charset=utf-8")
         elif self.path == "/api/tree":
             self._json(tree())
+        elif self.path == "/metrics":
+            self._send(200, metrics_text().encode(), "text/plain; version=0.0.4")
         else:
             self._send(404, b"not found", "text/plain")
 
-    def do_POST(self) -> None:  # noqa: N802 — http.server API
+    def do_POST(self) -> None:  # noqa: N802
         if self.path != "/api/query":
             self._send(404, b"not found", "text/plain")
             return
@@ -156,6 +193,20 @@ class Handler(BaseHTTPRequestHandler):
         pass
 
 
+def _view_loop() -> None:
+    while True:
+        try:
+            _refresh_views()
+        except Exception:
+            pass
+        time.sleep(VIEW_REFRESH_S)
+
+
 if __name__ == "__main__":
-    print(f"data-plane explorer on :{PORT}, reading {DATA_DIR}")
+    _refresh_views()
+    threading.Thread(target=_view_loop, daemon=True).start()
+    print(
+        f"data-plane explorer on :{PORT}, views refresh every {VIEW_REFRESH_S}s, "
+        f"last {flight_window()} flights, reading {DATA_DIR}"
+    )
     ThreadingHTTPServer(("", PORT), Handler).serve_forever()

simulator/monitoring/exporter.py

@@ -1,13 +1,14 @@
 """Prometheus exporter over the simulator's Parquet output.
 
 Periodically scans DATA_DIR with DuckDB and exposes fleet statistics as
-/metrics. Zero dependencies beyond duckdb: the exposition format is plain
-text, served with the standard library HTTP server.
+/metrics. Scans only the most recent flight partitions by default so CPU
+stays bounded as the lake grows across pod restarts.
 """
 
 from __future__ import annotations
 
 import os
+import sys
 import threading
 import time
 from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
@@ -15,27 +16,29 @@ from pathlib import Path
 
 import duckdb
 
+sys.path.insert(0, str(Path(__file__).resolve().parent))
+from lake import flight_window, iter_parquet_files, parquet_reader  # noqa: E402
+
 DATA_DIR = Path(os.environ.get("DATA_DIR", "./data"))
 PORT = int(os.environ.get("EXPORTER_PORT", "9105"))
-SCAN_INTERVAL_S = float(os.environ.get("SCAN_INTERVAL_S", "5"))
+SCAN_INTERVAL_S = float(os.environ.get("SCAN_INTERVAL_S", "15"))
 
 _lock = threading.Lock()
 _payload = "# swarm exporter starting\n"
+_last_scan_s = 0.0
 
 
 def _q(con: duckdb.DuckDBPyConnection, sql: str) -> list[tuple]:
     try:
         return con.sql(sql).fetchall()
     except duckdb.Error:
-        return []  # partitions may not exist yet while the swarm warms up
+        return []
 
 
 def collect() -> str:
+    global _last_scan_s
+    started = time.monotonic()
     con = duckdb.connect()
-    # union_by_name: sensors have different schemas under one dataset glob
-    glob = lambda ds: (  # noqa: E731
-        f"'{DATA_DIR}/dataset={ds}/**/*.parquet', hive_partitioning=true, union_by_name=true"
-    )
     lines: list[str] = []
 
     def metric(name: str, help_text: str, mtype: str, rows: list[str]) -> None:
@@ -43,27 +46,38 @@ def collect() -> str:
         lines.append(f"# TYPE {name} {mtype}")
         lines.extend(rows)
 
+    for ds, name in (
+        ("telemetry", "swarm_rows_total"),
+        ("detections", "swarm_detections_total"),
+        ("state", "swarm_state_frames_total"),
+    ):
+        reader = parquet_reader(DATA_DIR, ds)
+        if ds == "telemetry":
             metric(
-        "swarm_rows_total", "Telemetry rows written per drone and sensor", "gauge",
-        [f'swarm_rows_total{{drone="{d}",sensor="{s}"}} {n}'
-         for d, s, n in _q(con, f"SELECT drone, sensor, count(*) FROM read_parquet({glob('telemetry')}) GROUP BY 1,2")],
+                name, f"Rows in dataset={ds} (recent {flight_window()} flights)", "gauge",
+                [f'{name}{{drone="{d}",sensor="{s}"}} {n}'
+                 for d, s, n in _q(con, f"SELECT drone, sensor, count(*) FROM read_parquet({reader}) GROUP BY 1,2")],
             )
+        elif ds == "detections":
             metric(
-        "swarm_detections_total", "Detection events per drone and class", "gauge",
-        [f'swarm_detections_total{{drone="{d}",cls="{c}"}} {n}'
-         for d, c, n in _q(con, f"SELECT drone, cls, count(*) FROM read_parquet({glob('detections')}) GROUP BY 1,2")],
+                name, "Detection events per drone and class", "gauge",
+                [f'{name}{{drone="{d}",cls="{c}"}} {n}'
+                 for d, c, n in _q(con, f"SELECT drone, cls, count(*) FROM read_parquet({reader}) GROUP BY 1,2")],
             )
+        else:
             metric(
-        "swarm_state_frames_total", "Pose broadcast frames per drone and direction", "gauge",
-        [f'swarm_state_frames_total{{drone="{d}",direction="{dr}"}} {n}'
-         for d, dr, n in _q(con, f"SELECT drone, direction, count(*) FROM read_parquet({glob('state')}) GROUP BY 1,2")],
+                name, "Pose broadcast frames per drone and direction", "gauge",
+                [f'{name}{{drone="{d}",direction="{dr}"}} {n}'
+                 for d, dr, n in _q(con, f"SELECT drone, direction, count(*) FROM read_parquet({reader}) GROUP BY 1,2")],
             )
+
+    telem = parquet_reader(DATA_DIR, "telemetry")
     metric(
         "swarm_battery_pct", "Latest battery level per drone", "gauge",
         [f'swarm_battery_pct{{drone="{d}"}} {v}'
          for d, v in _q(con, f"""
             SELECT drone, arg_max(level_pct, ts_ns)
-            FROM read_parquet({glob('telemetry')})
+            FROM read_parquet({telem})
             WHERE sensor='battery' GROUP BY drone""")],
     )
     metric(
@@ -71,26 +85,24 @@ def collect() -> str:
         [f'swarm_rssi_dbm{{drone="{d}",peer="{p}"}} {v}'
          for d, p, v in _q(con, f"""
             SELECT drone, peer_id, arg_max(rssi_dbm, ts_ns)
-            FROM read_parquet({glob('telemetry')})
-            WHERE sensor='rssi' GROUP BY drone, peer_id""")],
-    )
-    metric(
-        "swarm_peer_distance_m", "Latest inter-drone distance estimate", "gauge",
-        [f'swarm_peer_distance_m{{drone="{d}",peer="{p}"}} {v}'
-         for d, p, v in _q(con, f"""
-            SELECT drone, peer_id, arg_max(distance_m, ts_ns)
-            FROM read_parquet({glob('telemetry')})
+            FROM read_parquet({telem})
             WHERE sensor='rssi' GROUP BY drone, peer_id""")],
     )
 
-    files = list(DATA_DIR.rglob("*.parquet"))
+    files = iter_parquet_files(DATA_DIR)
     metric(
-        "swarm_parquet_bytes", "Bytes on disk per dataset", "gauge",
+        "swarm_parquet_bytes", "Bytes on disk per dataset (recent flights)", "gauge",
         [f'swarm_parquet_bytes{{dataset="{ds}"}} {sum(f.stat().st_size for f in files if f"dataset={ds}" in str(f))}'
          for ds in ("telemetry", "detections", "state")],
     )
-    metric("swarm_parquet_files", "Parquet files on disk", "gauge",
+    metric("swarm_parquet_files", "Parquet files scanned (recent flights)", "gauge",
            [f"swarm_parquet_files {len(files)}"])
+    metric("swarm_metric_flight_window", "Flight partitions included per dataset", "gauge",
+           [f"swarm_metric_flight_window {flight_window()}"])
+
+    _last_scan_s = time.monotonic() - started
+    metric("swarm_exporter_scan_duration_seconds", "Wall time of the last metrics scan", "gauge",
+           [f"swarm_exporter_scan_duration_seconds {_last_scan_s:.4f}"])
     con.close()
     return "\n".join(lines) + "\n"
 
@@ -101,11 +113,11 @@ def scanner() -> None:
         started = time.monotonic()
         try:
             payload = collect()
-        except Exception as exc:  # keep serving stale metrics over dying
+        except Exception as exc:
             payload = f"# collect error: {exc}\n"
         with _lock:
             _payload = payload
-        time.sleep(max(0.5, SCAN_INTERVAL_S - (time.monotonic() - started)))
+        time.sleep(max(1.0, SCAN_INTERVAL_S - (time.monotonic() - started)))
 
 
 class Handler(BaseHTTPRequestHandler):
@@ -123,10 +135,13 @@ class Handler(BaseHTTPRequestHandler):
         self.wfile.write(body)
 
     def log_message(self, *_args: object) -> None:
-        pass  # scrapes every few seconds; keep the log quiet
+        pass
 
 
 if __name__ == "__main__":
     threading.Thread(target=scanner, daemon=True).start()
-    print(f"swarm exporter on :{PORT}/metrics, scanning {DATA_DIR} every {SCAN_INTERVAL_S}s")
+    print(
+        f"swarm exporter on :{PORT}/metrics, scanning last {flight_window()} flights "
+        f"every {SCAN_INTERVAL_S}s under {DATA_DIR}"
+    )
     ThreadingHTTPServer(("", PORT), Handler).serve_forever()

simulator/monitoring/grafana/dashboards/swarm-platform.json

@@ -0,0 +1,90 @@
+{
+  "uid": "swarm-platform",
+  "title": "Swarm Platform — CPU & scan health",
+  "tags": ["swarm", "platform"],
+  "timezone": "utc",
+  "schemaVersion": 39,
+  "version": 1,
+  "refresh": "10s",
+  "time": { "from": "now-30m", "to": "now" },
+  "panels": [
+    {
+      "id": 1, "type": "timeseries", "title": "Node CPU %",
+      "gridPos": { "h": 8, "w": 12, "x": 0, "y": 0 },
+      "datasource": { "type": "prometheus", "uid": "swarm-prom" },
+      "targets": [{
+        "expr": "100 - (avg by (instance) (rate(node_cpu_seconds_total{mode=\"idle\"}[1m])) * 100)",
+        "legendFormat": "{{instance}}", "refId": "A"
+      }],
+      "fieldConfig": { "defaults": { "unit": "percent", "min": 0, "max": 100 }, "overrides": [] }
+    },
+    {
+      "id": 2, "type": "timeseries", "title": "Node memory used %",
+      "gridPos": { "h": 8, "w": 12, "x": 12, "y": 0 },
+      "datasource": { "type": "prometheus", "uid": "swarm-prom" },
+      "targets": [{
+        "expr": "(1 - node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes) * 100",
+        "legendFormat": "used", "refId": "A"
+      }],
+      "fieldConfig": { "defaults": { "unit": "percent", "min": 0, "max": 100 }, "overrides": [] }
+    },
+    {
+      "id": 3, "type": "timeseries", "title": "Exporter scan duration",
+      "gridPos": { "h": 7, "w": 8, "x": 0, "y": 8 },
+      "datasource": { "type": "prometheus", "uid": "swarm-prom" },
+      "targets": [{
+        "expr": "swarm_exporter_scan_duration_seconds",
+        "legendFormat": "scan seconds", "refId": "A"
+      }],
+      "fieldConfig": { "defaults": { "unit": "s" }, "overrides": [] }
+    },
+    {
+      "id": 4, "type": "timeseries", "title": "Explorer query duration",
+      "gridPos": { "h": 7, "w": 8, "x": 8, "y": 8 },
+      "datasource": { "type": "prometheus", "uid": "swarm-prom" },
+      "targets": [{
+        "expr": "swarm_explorer_query_duration_seconds",
+        "legendFormat": "last query", "refId": "A"
+      }],
+      "fieldConfig": { "defaults": { "unit": "s" }, "overrides": [] }
+    },
+    {
+      "id": 5, "type": "timeseries", "title": "Explorer tree build duration",
+      "gridPos": { "h": 7, "w": 8, "x": 16, "y": 8 },
+      "datasource": { "type": "prometheus", "uid": "swarm-prom" },
+      "targets": [{
+        "expr": "swarm_explorer_tree_duration_seconds",
+        "legendFormat": "tree seconds", "refId": "A"
+      }],
+      "fieldConfig": { "defaults": { "unit": "s" }, "overrides": [] }
+    },
+    {
+      "id": 6, "type": "stat", "title": "Parquet files scanned",
+      "gridPos": { "h": 5, "w": 6, "x": 0, "y": 15 },
+      "datasource": { "type": "prometheus", "uid": "swarm-prom" },
+      "targets": [{ "expr": "swarm_parquet_files", "instant": true, "refId": "A" }],
+      "options": { "reduceOptions": { "calcs": ["lastNotNull"] } }
+    },
+    {
+      "id": 7, "type": "stat", "title": "Flight window",
+      "gridPos": { "h": 5, "w": 6, "x": 6, "y": 15 },
+      "datasource": { "type": "prometheus", "uid": "swarm-prom" },
+      "targets": [{ "expr": "swarm_metric_flight_window", "instant": true, "refId": "A" }],
+      "options": { "reduceOptions": { "calcs": ["lastNotNull"] } }
+    },
+    {
+      "id": 8, "type": "stat", "title": "Telemetry rows (window)",
+      "gridPos": { "h": 5, "w": 6, "x": 12, "y": 15 },
+      "datasource": { "type": "prometheus", "uid": "swarm-prom" },
+      "targets": [{ "expr": "sum(swarm_rows_total)", "instant": true, "refId": "A" }],
+      "options": { "reduceOptions": { "calcs": ["lastNotNull"] } }
+    },
+    {
+      "id": 9, "type": "stat", "title": "Min fleet battery %",
+      "gridPos": { "h": 5, "w": 6, "x": 18, "y": 15 },
+      "datasource": { "type": "prometheus", "uid": "swarm-prom" },
+      "targets": [{ "expr": "min(swarm_battery_pct)", "instant": true, "refId": "A" }],
+      "fieldConfig": { "defaults": { "unit": "percent", "min": 0, "max": 100 }, "overrides": [] }
+    }
+  ]
+}

simulator/monitoring/lake.py

@@ -0,0 +1,43 @@
+"""Scan the Hive-partitioned lake without re-reading every historical flight."""
+
+from __future__ import annotations
+
+import os
+from pathlib import Path
+
+
+def flight_window() -> int:
+    return max(1, int(os.environ.get("METRIC_FLIGHT_WINDOW", "5")))
+
+
+def recent_flights(data_dir: Path, dataset: str, limit: int | None = None) -> list[Path]:
+    """Most recently modified flight= partitions for a dataset."""
+    limit = limit or flight_window()
+    base = data_dir / f"dataset={dataset}"
+    if not base.is_dir():
+        return []
+    flights = [p for p in base.iterdir() if p.is_dir() and p.name.startswith("flight=")]
+    flights.sort(key=lambda p: p.stat().st_mtime, reverse=True)
+    return flights[:limit]
+
+
+def parquet_reader(data_dir: Path, dataset: str, *, limit: int | None = None) -> str:
+    """DuckDB read_parquet() source limited to recent flights."""
+    flights = recent_flights(data_dir, dataset, limit)
+    if not flights:
+        path = data_dir / f"dataset={dataset}" / "**" / "*.parquet"
+        return f"'{path}', hive_partitioning=true, union_by_name=true"
+    if len(flights) == 1:
+        return f"'{flights[0]}/**/*.parquet', hive_partitioning=true, union_by_name=true"
+    inner = ", ".join(f"'{f}/**/*.parquet'" for f in flights)
+    return f"[{inner}], hive_partitioning=true, union_by_name=true"
+
+
+def iter_parquet_files(data_dir: Path, *, flight_limit: int | None = None) -> list[Path]:
+    """Parquet paths under recent flights only — avoids full-lake rglob."""
+    limit = flight_limit or flight_window()
+    out: list[Path] = []
+    for ds in ("telemetry", "detections", "state"):
+        for flight in recent_flights(data_dir, ds, limit):
+            out.extend(flight.rglob("*.parquet"))
+    return out

simulator/monitoring/prometheus.yml

@@ -1,7 +1,10 @@
 global:
-  scrape_interval: 5s
+  scrape_interval: 15s
 
 scrape_configs:
   - job_name: swarm
     static_configs:
       - targets: ["exporter:9105"]
+  - job_name: explorer
+    static_configs:
+      - targets: ["explorer:8088"]

simulator/tests/test_broadcast.py

@@ -9,7 +9,7 @@ from virtual_drone.flight import Pose
 
 
 def test_frame_size_matches_wire_spec() -> None:
-    # Documented as 46 B in docs/04; struct packs to FRAME.size (45 B on this layout).
+    # Keep docs/04, the journey, and prototype/src/sim.ts POSE_BYTES in lockstep.
     assert FRAME.size == 45
 
 

simulator/tests/test_lake.py

@@ -0,0 +1,40 @@
+"""Lake scan helpers — bounded to recent flight partitions."""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+from monitoring.lake import flight_window, iter_parquet_files, parquet_reader, recent_flights
+
+
+def test_parquet_reader_limits_to_recent_flights(tmp_path: Path) -> None:
+    for i, name in enumerate(("flight=aaa", "flight=bbb", "flight=ccc")):
+        hour = tmp_path / f"dataset=telemetry/{name}/drone=dr-01/sensor=imu/year=2026/month=07/day=08/hour=10"
+        hour.mkdir(parents=True)
+        f = hour / "data.parquet"
+        f.write_bytes(b"x" * (i + 1))
+        # Make later names newer
+        import os
+        import time
+        os.utime(f, (time.time() + i, time.time() + i))
+
+    reader = parquet_reader(tmp_path, "telemetry", limit=1)
+    assert "flight=ccc" in reader
+    assert "flight=bbb" not in reader
+
+
+def test_iter_parquet_files_skips_old_flights(tmp_path: Path) -> None:
+    old = tmp_path / "dataset=state/flight=old/drone=dr-01/year=2026/month=07/day=08/hour=09"
+    old.mkdir(parents=True)
+    (old / "data.parquet").write_bytes(b"old")
+    new = tmp_path / "dataset=state/flight=new/drone=dr-01/year=2026/month=07/day=08/hour=10"
+    new.mkdir(parents=True)
+    new_file = new / "data.parquet"
+    new_file.write_bytes(b"new")
+    import os
+    import time
+    os.utime(new_file, (time.time() + 10, time.time() + 10))
+
+    files = iter_parquet_files(tmp_path, flight_limit=1)
+    assert len(files) == 1
+    assert "flight=new" in str(files[0])

simulator/virtual_drone/broadcast.py

@@ -1,12 +1,13 @@
 """State broadcast over UDP: the compact pose frame from the sync design.
 
-Frame layout (little-endian, 46 bytes):
+Frame layout (little-endian, 45 bytes) — source of truth for docs and the
+prototype bandwidth estimate:
 
     magic     2s   b"SH"
     version   B
-    drone_id  8s   zero-padded ascii
+    drone_id  8s   zero-padded ascii (PoC; production may switch to uint16 registry)
     ts_ns     q    epoch nanoseconds
-    pos_mm    3i   position, millimeters (quantized on the wire only)
+    pos_mm    3i   position in the mission frame, millimeters (wire quantization only)
     att_cdeg  3h   roll/pitch/yaw, centi-degrees
     vel_cms   3h   velocity, cm/s
     frame_ref B

simulator/virtual_drone/main.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+import os
 import random
 import select
 import time
@@ -108,6 +109,10 @@ def run(cfg: Config) -> None:
         writer.seal()
     print(f"[{cfg.drone_id}] done: {frames_sent} state frames sent, "
           f"{len(peers_seen)} peers seen {sorted(peers_seen)}; sealed to {root}")
+    if os.environ.get("KEEP_ALIVE", "0") == "1":
+        print(f"[{cfg.drone_id}] KEEP_ALIVE=1 — idle after seal (no pod restart churn)")
+        while True:
+            time.sleep(3600)
 
 
 if __name__ == "__main__":