Skip to main content
Video tutorial · 2:44

Certificate discovery

Find the certs you don't know about. Scan IP/CIDR ranges, watch CT logs for new issuances, detect drift, and transition discovered certs to managed lifecycle.

Discovery sources​

A discovery source is a recurring scan definition. Six types:

  • CIDR / IP — IP ranges + ports. The worker opens TLS connections and captures the certificate chain. SNI-aware, so certs behind SNI-routed frontends are found too (see below).
  • Hostname — explicit FQDN targets, each probed with its own name as SNI. Useful for names that don't appear in any CIDR you control.
  • DNS Zone Transfer (AXFR) — transfers your internal DNS zones and probes each host with the correct SNI. The authoritative way to enumerate every certificate behind an ingress / content switch. See AXFR discovery.
  • Cloud DNS Provider — for DNS you host in the cloud (Cloudflare, Google Cloud DNS), where AXFR isn't available: lists the zone's records through the provider API and probes each host with its own name as SNI; a Cloudflare-proxied record is scanned at the Cloudflare edge (what the public sees) and at the origin behind it. See Cloud DNS discovery.
  • Targets — reads the certificates stored on the SSH (Linux) and WinRM, IIS and Exchange (Windows) targets you already deploy to: files on every local disk and the Windows certificate stores, served or not. Private keys never leave the server. See Server certificate scan.
  • CT log monitor — subscribes to Certificate Transparency logs and matches issuances against domains you own. It depends on an external aggregator (crt.sh) that is often rate-limited or unavailable, so it is currently hidden from the source-type picker in the UI; for SNI-routed internal infrastructure prefer AXFR, which is authoritative and needs no external service.

SNI-aware discovery​

A single IP:port behind an ingress, content switch, or virtual-host load balancer can serve many certificates, chosen by the SNI in the handshake. Network scans run deep SNI discovery — probing each IP with names mined from cert SANs, reverse DNS, and a cross-source candidate pool — so hidden certs surface instead of only the default one. For authoritative enumeration, use an AXFR source. See Discovery sources.

Servers​

Discovery keeps one row per source, address, port and server name. The screens group them into servers — one per address and port, with the names it answers for and the sources that saw it inside — and count servers wherever they count places. Discovery → Servers lists them and opens a page per server with its certificates, TLS settings, names, findings and history. See Servers.

Overview and insights​

Discovery → Overview (the first tab) shows the discovered estate at a glance — when certificates expire, how much is managed, which CAs, key types and TLS versions are in use, what changed this week, the most common risk reasons and the certificates served on the most servers. Discovery → Insights groups it by CA, domain or network, team, risk reason and reach. Every number opens the list it counts. See Overview page and Insights.

Scheduling​

Each source has its own cadence — either an interval (every N hours, minimum 24) or a day+time recurrence (selected weekdays at a fixed time in a chosen timezone, for off-peak scans). An operator can also trigger a scan manually. See Discovery sources.

Security findings​

Every scan produces zero or more findings per certificate or endpoint. Severity ranges from info to critical. A selection:

FindingSeverity
Expired certificatecritical
Expires within 7 dayshigh
Weak key (RSA < 2048-bit)high
SHA-1 signaturehigh
Accepts TLS 1.0high
Weak cipher (RC4, 3DES, EXPORT)high
Accepts TLS 1.1medium
Self-signedmedium
Not managed and expiring within 30 daysmedium
SNI required, no certificate discoveredmedium
Ingress/default placeholder certificatelow
PQC-vulnerable (RSA < 2048-bit)low

The full list is in Findings.

RSA-2048 is not flagged (still acceptable today); only keys smaller than 2048 bits are. The SNI required finding marks a visible blind spot — an IP:port that rejects SNI-less handshakes where no certificate could be enumerated (consider an AXFR source).

Drift detection​

On every recurring scan, CertAutoPilot diffs the captured certificate against the previous snapshot. A drift event fires when:

  • The fingerprint changes (cert was replaced).
  • The issuer changes (CA migration).
  • The public key changes (key rotation).
  • The SAN set changes.

Each change is recorded per endpoint (old and new fingerprint plus what changed), so you can see the full history on the discovered certificate's detail page and on the server's page.

PQC classification​

Each discovered cert is classified for post-quantum readiness:

  • vulnerable — already-weak keys (RSA below 2048 bits).
  • transition — every other classical algorithm (RSA 2048+, ECDSA, Ed25519): sound today, to be migrated once post-quantum certificates are available.
  • ready — a post-quantum signature key (ML-DSA).
  • unknown — the key algorithm could not be classified.

The inventory shows the split, and Discovery → Quantum readiness puts it next to the key exchange every server negotiates — the part that can be made quantum-safe today. Most environments are almost entirely transition today; the value is knowing where you'd start migrating. See PQC readiness.

Tags, owners and tag rules​

Each discovered certificate can carry tags and an owner, typed on its page or given by tag rules — conditions on its names, issuer, key, trust, source, address or port that are applied after a scan (at most once every 5 minutes per project; a scan inside that window is picked up by the next pass), every hour, and right after a rule is created, changed or deleted. The inventory, findings and exports filter by both.

Managed transition​

Discovered certificates are read-only by default. To bring one under management:

  1. From Discovery → Inventory, locate the cert marked Not managed.
  2. Click Manage on that row (or open the cert detail page and use Manage with CertAutoPilot) — opens the Manage wizard.
  3. Choose where to issue from — ACME (an ACME account and a validation method) or Microsoft AD CS (a connection and a certificate template) — then the domains and key type.
  4. Confirm — a new managed certificate for the same names is issued straight away and linked to the discovered record.

From then on the certificate follows the standard lifecycle. Attach distributions to deploy it to the endpoints that still serve the old cert — Target candidates lists the places discovery found certificates and turns one into a distribution target in a few clicks.

See also​

Video tutorial

Short, hands-on videos showing this feature working for real. English audio, English and Turkish subtitles. All video tutorials →

Discover Certificates by Scanning Networks and HostnamesYouTube ↗