telemetry: Make http_user_agent a normalized field

This way we store a short 8-byte hash of the UA instead of the full
string; exactly the same way we store TLS ClientHello info.
This commit is contained in:
Matthew Holt
2018-05-10 08:57:25 -06:00
parent df7cdc3fae
commit 9160789b42
4 changed files with 17 additions and 3 deletions

View File

@@ -65,9 +65,10 @@ func (h *tlsHandler) ServeHTTP(w http.ResponseWriter, r *http.Request) {
h.listener.helloInfosMu.RUnlock()
ua := r.Header.Get("User-Agent")
uaHash := telemetry.FastHash([]byte(ua))
// report this request's UA in connection with this ClientHello
go telemetry.AppendUnique("tls_client_hello_ua:"+caddytls.ClientHelloInfo(info).Key(), ua)
go telemetry.AppendUnique("tls_client_hello_ua:"+caddytls.ClientHelloInfo(info).Key(), uaHash)
var checked, mitm bool
if r.Header.Get("X-BlueCoat-Via") != "" || // Blue Coat (masks User-Agent header to generic values)

View File

@@ -354,7 +354,9 @@ func (s *Server) ServeHTTP(w http.ResponseWriter, r *http.Request) {
if len(ua) > 512 {
ua = ua[:512]
}
go telemetry.AppendUnique("http_user_agent", ua)
uaHash := telemetry.FastHash([]byte(ua)) // this is a normalized field
go telemetry.SetNested("http_user_agent", uaHash, ua)
go telemetry.AppendUnique("http_user_agent_count", uaHash)
go telemetry.Increment("http_request_count")
// copy the original, unchanged URL into the context