Metrics

Set metrics_addr (for example "127.0.0.1:9100") to serve Prometheus metrics at /metrics on a separate listener. It is off when empty. The endpoint has no authentication, so bind it to localhost or a private network, never to the public interface. In Docker, bind 0.0.0.0:9100 and only publish it on the network Prometheus uses.

scrape_configs:
  - job_name: verifyngo
    static_configs:
      - targets: ["verifyngo:9100"]

Metric	Labels	Meaning
`verifyngo_decisions_total`	`decision`	Every request and how it was handled
`verifyngo_rule_hits_total`	`rule`, `action`	Requests decided by each rule; `default` when none matched
`verifyngo_challenges_total`	`provider`, `result`	Challenges `issued`, and verify attempts `passed`, `failed`, `automated`, `throttled`, or `error`
`verifyngo_bans_total`	`reason`	Bans by `walk-aways`, `failed attempts`, or `fingerprint`
`verifyngo_passive_total`	`result`	Progressive mode outcomes
`verifyngo_browser_mismatch_total`	`kind`	User agents that do not match their headers
`verifyngo_upstream_responses_total`	`status`	Upstream responses by status class, or `error`
`verifyngo_store_errors_total`	`op`	Failed Redis operations
`verifyngo_rule_reloads_total`	`result`	Rule loads and reloads
`verifyngo_rules_loaded`		`0` while rules are loading at startup, then `1`
`verifyngo_blocklist_entries`		Entries in the AbuseIPDB blocklist
`verifyngo_tor_requests_total`	`kind`	Requests over Tor, `onion` or `exit`
`verifyngo_tor_exits`		Addresses in the Tor exit list

Go runtime and process metrics (go_*, process_*) are included. Policy rules are labeled by their name, and unnamed rules as rule-N by position.

Get started

Rules

Protection

Operations