Telemetry

Tripl can send one small, anonymous ping a day, so the people who build it know how many instances keep running and on which warehouses. It is off unless you turn it on, and nothing in it names a person, a team, a host or your data.

Turning it on or off

• tripl install --telemetry writes TELEMETRY_ENABLED=true into the .env it creates; without the flag it writes false.
• Anywhere else, set TELEMETRY_ENABLED=true (or false) in the server's environment and restart. TELEMETRY_ENDPOINT says where the ping goes (https://telemetry.tripl.io/v1/ping by default); empty sends nothing.

A public demo instance (PUBLIC_DEMO) never sends it, whatever these say.

What it sends

Once a day, at 07:13 UTC, the worker POSTs this JSON document — and nothing else:

Field	Example	What it is
schema	1	The document's version.
instance_id	"5f0c…"	A random id, made the first time and kept in the database. It identifies the instance, not anyone using it.
version	"0.2.3"	The server's version.
edition	"community"	community or enterprise.
deployment_mode	"self_hosted"	self_hosted or hosted.
warehouse_engines	["clickhouse"]	Which engines your data sources use; never their hosts or names.
projects	"1-10"	How many projects, as a range: 0, 1-10, 11-100, 101-1000 or 1000+.
event_types	"11-100"	How many event types, as a range.
users	"1-10"	How many accounts, as a range.
scans_last_day	"0"	How many scans ran in the last 24 hours, as a range.
days_since_install	42	Days since the instance made its id.

Never sent: names, slugs, email addresses, hosts, URLs, queries, event or field names, or anything read from your warehouse.

Seeing what was sent

Settings → Platform → Runtime shows, to platform admins, whether telemetry is on, where it goes, and the exact document it last sent (GET /api/v1/platform/settings/telemetry).

When the receiver is down

The request has a five-second timeout and follows no redirect. A failure is logged at INFO and forgotten; nothing retries, and nothing else is affected.