Switching Data Sources

Switching Data Sources

---

Switching from Strava API to scrape mode

Use this when your Strava API subscription lapses (or when the Club Activities API is deprecated after 2026-09-01).

See Data-Source-Scrape-Mode for complete step-by-step instructions. Summary:

1. Get a _strava4_session cookie from your browser DevTools
2. Set STRAVA_MY_SOURCE="scrape" and STRAVA_SESSION_COOKIE="..." in /etc/strava-my-activities.conf (for My Activities), or STRAVA_SOURCE="scrape" and STRAVA_SESSION_COOKIE="..." in /etc/strava-leaderboard.conf (for the club leaderboard)
3. For the club leaderboard, set STRAVA_SCRAPE_START_DATE to avoid overlap duplicates at transition
4. Deploy updated scripts: scp strava-leaderboard.sh root@192.168.1.1:/usr/bin/strava-leaderboard && scp strava-lib.sh root@192.168.1.1:/usr/bin/strava-lib.sh
5. Run once to verify: strava-leaderboard

Switching is non-destructive: existing activities are matched by Strava ID and updated in place.

---

Switching from Strava to HealthSync (keeping full history)

HealthSync only keeps approximately 30 days of exports on Google Drive. If you switch cold — just change which script cron runs — the HealthSync dashboard will start from scratch and show only recent activities.

To carry over your full Strava history, use HEALTHSYNC_IMPORT_STRAVA_STORE.

Why not just run both scripts?

Both strava-my-activities.sh and healthsync-activities.sh write to the same WEB_DIR (/www/strava/me). Whichever runs last overwrites activities.json and all HTML — you'd see only one source's activities. The migration approach below is the right path: merge the Strava store into HealthSync once, then switch cron.

Migration steps

1. Configure HealthSync with Google credentials and the Strava store path:

vi /etc/healthsync-activities.conf

Add:

HEALTHSYNC_IMPORT_STRAVA_STORE="/usr/lib/strava-my-activities/activities.ndjson"

(Adjust the path if your STRAVA_MY_STATE_DIR differs — the default on this project's router is /mnt/sda5/strava-my-activities/.)

Also fill in your Google OAuth credentials and DRIVE_FOLDER_ID. See Data-Source-HealthSync for the full Google OAuth setup.

2. Run once — imports Strava history, downloads new HealthSync activities, renders HTML:

healthsync-activities

Verify the log output:

• Strava history: 1234 activities merged from …
• store: N activities total

3. Switch cron — healthsync is already scheduled by install.sh; just remove strava-my-activities:

crontab -l | grep -v 'strava-my-activities' | crontab -

How the merge works

HEALTHSYNC_IMPORT_STRAVA_STORE is idempotent — safe to leave in place permanently.

• Strava IDs are numeric (18784255013); HealthSync IDs are date strings (2026-06-22-20-01-ride) — they never collide, so there is no risk of mixing up or duplicating activities across the two sources.
• On each run, healthsync-activities reads the Strava NDJSON file and appends any records whose ID is not yet in the HealthSync store.

The Strava detail JSON files (per-activity maps and splits) stay in DETAIL_DIR (/www/strava/me/details/ by default) and are served by the HealthSync dashboard unchanged — numeric Strava IDs still link correctly from the activity list. New HealthSync activities use GPX files cached in $WEB_DIR/gpx/ instead.

Important: keep Strava cron running until your API access ends

Every daily Strava run adds activities to activities.ndjson. The more history you accumulate before switching, the more complete the HealthSync dashboard will be from day one. Do not switch cron early.

Important: do not delete the Strava NDJSON after switching

/usr/lib/strava-my-activities/activities.ndjson is the migration source that HEALTHSYNC_IMPORT_STRAVA_STORE reads on every run. It is safe to leave it in place — it will not change once the Strava API is gone, and it takes up almost no flash space. Deleting it would prevent future healthsync-activities runs from re-importing history (e.g. after a router reset).