Telescope
Telescope settings
Pause recording, sample, set thresholds, ignore paths and types, redact keys and monitor tags — from the console, without a deploy.
Telescope → Settings is per environment. SDKs receive these settings in the response to each delivery and apply them to the next journeys.
| Setting | Default | Effect |
|---|---|---|
| Recording | on | Pause to stop storing journeys everywhere at once. |
| Retention | 48 hours | Up to 14 days. Older journeys are deleted. |
| Sample rate | 100% | Share of journeys kept. Failed, slow and monitored journeys are always kept. A per-service sample_rate in code wins. |
| Slow request | 1000 ms | Root duration that marks a journey slow. |
| Slow query | 100 ms | Query duration that marks a query slow. |
| N+1 threshold | 5 | Repeats of one query shape in a batch that count as N+1. |
| Ignored paths | — | Globs (/health*) never recorded. Merged with each SDK's own list. |
| Ignored types | — | Entry types never recorded (cache, view, …). |
| Redaction keys | built-in list | Extra keys to redact, on top of authorization, cookie, set-cookie, x-api-key, CSRF tokens, passwords and tokens. |
| Monitored tags | — | Journeys with these tags are always kept — e.g. Auth:42 while debugging one customer. |
| Forward exceptions | on | File exceptions in Errors. |
| Forward logs | on | Copy log entries to Logs. |
Viewing journeys needs the telescope.view permission; changing settings or
clearing entries needs telescope.manage. Developers, admins and owners have
both.