release: bump version to 0.66.1

docs/technical/en/earth-satellite-footprint-policy.md

@@ -17,7 +17,13 @@ Related context:
 
 ## Current Local Categories
 
-The CelesTrak collector now downloads the complete active satellite catalog from `GROUP=active&FORMAT=json` instead of fetching several smaller groups and merging them. This prevents one failed CelesTrak group request from being saved as a successful but incomplete batch. The collector only proceeds when the downloaded JSON is a parseable array and records include `NORAD_CAT_ID`; network, resume, or parsing failures are retried, and final failure preserves the previous current dataset.
+The CelesTrak collector now prefers the complete active satellite catalog from `GROUP=active&FORMAT=json`. This prevents one failed CelesTrak group request from being saved as a successful but incomplete batch. The collector only proceeds when the downloaded JSON is a parseable array and records include `NORAD_CAT_ID`; network, resume, or parsing failures are retried, and final failure preserves the previous current dataset.
+
+CelesTrak applies a repeat-download window to large groups such as `active`. To make database resets recoverable, Planet stores raw downloads under `$PLANET_CACHE_DIR/downloads`. When `active` returns the CelesTrak "GP data has not updated" HTTP 403:
+
+- If an `active` cache exists, the collector reuses it to repopulate the database.
+- If no `active` cache exists, the collector switches to fallback group mode and downloads `starlink`, `gps-ops`, `galileo`, `glonass`, `beidou`, `leo`, `geo`, and `iridium-next`.
+- Fallback group mode requires every group to succeed or have a reusable cache; if any group is missing, the whole collection fails and no partial dataset is saved.
 
 The collector still provides `metadata.constellation_group` to the frontend, but the value now comes from executable inference:
 
@@ -25,7 +31,7 @@ The collector still provides `metadata.constellation_group` to the frontend, but
 - `OBJECT_NAME` starting with `IRIDIUM` is marked as `iridium-next`
 - Other active satellites are not forced into the old CelesTrak small-group labels, because a broad source group is not an exact constellation
 
-The product policy therefore still discusses GNSS/RNSS, GEO, generic LEO, and Iridium NEXT semantics, but code should no longer assume that saved CelesTrak rows carry the old `gps-ops`, `galileo`, `glonass`, `beidou`, `leo`, or `geo` group labels.
+The product policy therefore still discusses GNSS/RNSS, GEO, generic LEO, and Iridium NEXT semantics. However, only fallback group mode stores the old `gps-ops`, `galileo`, `glonass`, `beidou`, `leo`, or `geo` labels; the active primary path does not force every satellite into those labels.
 
 ## Research Conclusions
 

docs/technical/en/ops-runbook.md

@@ -73,9 +73,9 @@ Cleanup order and boundaries:
 
 - If `planet_postgres` is running, the script first clears the `public` schema in `planet_db`. This prevents old `collected_data.is_current = true` rows from making Earth OOBE report `ready=true` if Docker volume removal later fails.
 - Docker cleanup targets resources whose Compose project is `planet`, plus the explicit volumes `planet_postgres_data`, `planet_redis_data`, `postgres_data`, and `redis_data`; do not delete unlabeled volumes by a broad `planet_*` pattern, because another local project could own them.
-- Local build state removes `.venv`, frontend `node_modules` / `dist`, Planet state/cache, and scattered Python / Vite cache directories.
+- Local build state removes `.venv`, frontend `node_modules` / `dist`, Planet state, and scattered Python / Vite cache directories. `$PLANET_CACHE_DIR/downloads` is preserved so upstream raw downloads such as CelesTrak can survive database resets and local rebuild cleanup.
 
-After the reset, run `./planet.sh init` again to recreate tables and default seed data. Old collected records are not restored, and Earth OOBE is evaluated from the backend's real collection state on the next visit.
+After the reset, run `./planet.sh init` again to recreate tables and default seed data. Old collected records are not restored, and Earth OOBE is evaluated from the backend's real collection state on the next visit. When CelesTrak later returns its "GP data has not updated" HTTP 403, the backend first reuses the preserved download cache to repopulate the database; if no cache exists, wait for the next CelesTrak update window or use Space-Track as a fallback.
 
 ## Health Check