Skip to content

Anchors ​

The Anchors group covers the anchor fleet: which anchors exist and where, what each one measures and receives, the DNS monitoring set, and — for staff — the health and mesh views built on anchor-to-anchor measurements.

Anchor list ​

/anchors · public

The fleet, filterable by hostname, country, ASN, tag, firmware, probe status and status since, with a clustered map and configurable page size.

Anchors are also probes, so the probe-side filters here mean exactly what they mean on the probe list — an anchor with a disconnected probe is an anchor that is not currently measuring.

Anchor detail ​

/anchors/detail · public

Look up a single anchor by ID. Four tabs:

Details — the anchor record and its probe record side by side, which is the quickest way to see the two identities an anchor has.

Mesh — this anchor's place in the anchoring mesh: who it measures and who measures it.

Health — the evaluated issues for this anchor and their recent transitions. Because the health log is written only on change, this reads as a history of what went wrong and when rather than a per-tick dump.

Change history — what has changed about the anchor over time.

There is also a neighbours map. "Neighbour" here means geographically near, which is the useful sense when you are picking a fallback vantage point or sanity-checking a claimed location.

Anchor measurements ​

/anchors/measurements · public

The catalogue of measurements involving anchors, filterable by anchor ID, type, description, and a mesh / non-mesh toggle. The filters are reflected in the URL, so a filtered view is a shareable link.

The mesh toggle is the one to reach for first: it separates the standing all-pairs mesh from everything else anchors are involved in.

DNSMON anchors ​

/anchors/dnsmon · public

The curated anchor set used for DNSMON, refreshed daily from RIPE. It is around 50 rows and loads in one go — no pagination.

Membership rarely changes, so recent additions and removals are shown in a side panel from an append-only history. That panel is the interesting part: a DNSMON anchor being dropped is a meaningful event, and it is easy to miss in a list of 50 otherwise identical rows.

The staff views ​

All of the following require staff.

Anchor health ​

/anchors/health — fleet-wide anchor health: every active anchor evaluated against the issue rules, current problems plus the transition log. Anchors that get disabled have their open issues closed out and are marked inactive, but are never deleted.

Mesh health ​

/anchors/mesh-health — where the all-pairs invariant breaks down, reported per cell and from both directions (inbound gaps on the target, outbound gaps on the source). See The anchoring mesh for what those two views mean and why both exist.

Mesh matrix ​

/anchors/mesh-matrix — the latency matrix itself. Empty cells are a coverage question, so pair this with mesh health rather than reading a gap as "no route".

Mesh topology ​

/anchors/mesh-topology — the mesh as a graph. Adjacency is network distance, not geography; don't read the layout as a map.


Background reading: Anchors · The anchoring mesh

Atlas — built on RIPE Atlas data.