Probes

The Probes group takes you from a probe ID, an ASN or a location to everything Atlas knows about the matching probes — current state, network, history, and the measurements they take part in.

Probe lookup

/probes · public

Enter a probe ID and press Look up. The result is read live from RIPE, so it reflects the probe as it is right now rather than the daily archive. Two extra controls sit alongside:

• Refresh from RIPE — force a re-read, bypassing any cached response.
• Clear — reset the page.

Three badges can appear next to the result and each changes how you should read the rest of the page:

• Anchor — this probe is also an anchor; it has a whole second life as a measurement target.
• Private — the owner has restricted visibility, so parts of the record are simply unavailable.
• Blocked — the probe is blocked upstream.

The tabs

Details carries the core record, split into cards:

• Network — address, ASN and prefix, per address family. A probe can sit in different ASNs for IPv4 and IPv6, and this is where you notice.
• Timeline — status and how long the probe has been in it. "Disconnected for 43 days" is the number worth reading, not "Disconnected".
• Tags — RIPE-assigned and host-assigned tags together. Treat them as hints; hosts set their own.

There is also a map of the nearest other probes, which answers "if this one is down, what else is close?" — the practical question when you are choosing vantage points.

Measurements lists what the probe participates in.

Stats charts its measurement participation and connectivity — the shape of what it actually contributes, rather than just whether it is up.

Connections shows the latest 25 connect/disconnect events from measurement 7000, plus a per-family rollup of how its connecting prefix and ASN have changed. This is the fastest way to see a probe that keeps flapping, or one that has quietly moved networks.

Archive history walks the daily archive snapshots — the probe as it looked on each day, paginated.

Change history is the distilled version: only the fields that actually changed, rendered as field: old → new, with tag additions and removals as green and red chips. When you want "what happened to this probe" rather than "what did it look like every day", this is the tab.

Health shows the evaluated issues for the probe and their recent transitions, with a Re-check now control. Requires staff.

Probe list

/probes/list · public

The fleet, filterable by country, city, ASN, tag, status, firmware and status since, with a configurable page size and a cluster control for the map.

Status since is the filter that earns its place: combined with status = Disconnected it answers "what has been down for more than a month?", which is the question behind most fleet clean-up work.

Probe health

/probes/health · staff

Fleet-wide health rather than per-probe: every probe evaluated against the issue rules, with current state and transitions. Use it to find systematic problems — a firmware version misbehaving, an AS whose probes all degraded at once.

Probe result health

/probes/result-health · staff

The harder question: is a probe producing usable results for the measurements it participates in? A probe can be connected, green, and contributing nothing measurable. Results are graded, and this view is where a silently useless probe shows up.

---

Background reading: Probes · Connection events and uptime