lib/promscrape: properly set Host header when sending requests via http proxy

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/2794

docs/CHANGELOG.md

@@ -61,6 +61,7 @@ scrape_configs:
 * BUGFIX: [vmagent](https://docs.victoriametrics.com/vmagent.html): properly add service-level labels (`__meta_kubernetes_service_*`) to discovered targets for `role: endpointslice` in [kubernetes_sd_config](https://prometheus.io/docs/prometheus/latest/configuration/configuration/#kubernetes_sd_config). Previously these labels were missing. See [this issue](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/2823).
 * BUGFIX: [vmagent](https://docs.victoriametrics.com/vmagent.html): make sure that [stale markers](https://docs.victoriametrics.com/vmagent.html#prometheus-staleness-markers) are generated with the actual timestamp when unsuccessful scrape occurs. This should prevent from possible time series overlap on scrape target restart in dynmaic envirnoments such as Kubernetes.
 * BUGFIX: [vmagent](https://docs.victoriametrics.com/vmagent.html): properly reload changed `-promscrape.config` file when `-promscrape.configCheckInterval` option is set. The changed config file wasn't reloaded in this case since [v1.69.0](#v1690). See [this pull request](https://github.com/VictoriaMetrics/VictoriaMetrics/pull/2786). Thanks to @ttyv for the fix.
+* BUGFIX: [vmagent](https://docs.victoriametrics.com/vmagent.html): properly set `Host` header during target scraping when `proxy_url` is set to http proxy. Previously the `Host` header was set to the proxy hostname instead of the target hostname. See [this issue](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/2794).
 * BUGFIX: [VictoriaMetrics cluster](https://docs.victoriametrics.com/Cluster-VictoriaMetrics.html): assume that the response is complete if `-search.denyPartialResponse` is enabled and up to `-replicationFactor - 1` `vmstorage` nodes are unavailable. See [this issue](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1767).
 * BUGFIX: [vmselect](https://docs.victoriametrics.com/#vmselect): update `vm_partial_results_total` metric labels to be consistent with `vm_requests_total` labels.
 * FEATURE: accept tags without values when reading data in [DataDog format](https://docs.victoriametrics.com/Single-server-VictoriaMetrics.html#how-to-send-data-from-datadog-agent). Thanks to @PerGon for the [pull request](https://github.com/VictoriaMetrics/VictoriaMetrics/pull/2839).

lib/promscrape/client.go

@@ -46,7 +46,7 @@ type client struct {
 
 	scrapeURL               string
 	scrapeTimeoutSecondsStr string
-	host                    string
+	hostPort                string
 	requestURI              string
 	setHeaders              func(req *http.Request)
 	setProxyHeaders         func(req *http.Request)
@@ -57,10 +57,21 @@ type client struct {
 	disableKeepAlive        bool
 }
 
+func addMissingPort(addr string, isTLS bool) string {
+	if strings.Contains(addr, ":") {
+		return addr
+	}
+	if isTLS {
+		return addr + ":443"
+	}
+	return addr + ":80"
+}
+
 func newClient(sw *ScrapeWork) *client {
 	var u fasthttp.URI
 	u.Update(sw.ScrapeURL)
-	host := string(u.Host())
+	hostPort := string(u.Host())
+	dialAddr := hostPort
 	requestURI := string(u.RequestURI())
 	isTLS := string(u.Scheme()) == "https"
 	var tlsCfg *tls.Config
@@ -75,7 +86,7 @@ func newClient(sw *ScrapeWork) *client {
 		// like net/http package from Go does.
 		// See https://en.wikipedia.org/wiki/Proxy_server#Web_proxy_servers
 		pu := proxyURL.GetURL()
-		host = pu.Host
+		dialAddr = pu.Host
 		requestURI = sw.ScrapeURL
 		isTLS = pu.Scheme == "https"
 		if isTLS {
@@ -90,19 +101,14 @@ func newClient(sw *ScrapeWork) *client {
 		}
 		proxyURL = &proxy.URL{}
 	}
-	if !strings.Contains(host, ":") {
+	hostPort = addMissingPort(hostPort, isTLS)
-		if !isTLS {
+	dialAddr = addMissingPort(dialAddr, isTLS)
-			host += ":80"
-		} else {
-			host += ":443"
-		}
-	}
 	dialFunc, err := newStatDialFunc(proxyURL, sw.ProxyAuthConfig)
 	if err != nil {
 		logger.Fatalf("cannot create dial func: %s", err)
 	}
 	hc := &fasthttp.HostClient{
-		Addr:                         host,
+		Addr:                         dialAddr,
 		Name:                         "vm_promscrape",
 		Dial:                         dialFunc,
 		IsTLS:                        isTLS,
@@ -152,7 +158,7 @@ func newClient(sw *ScrapeWork) *client {
 		sc:                      sc,
 		scrapeURL:               sw.ScrapeURL,
 		scrapeTimeoutSecondsStr: fmt.Sprintf("%.3f", sw.ScrapeTimeout.Seconds()),
-		host:                    host,
+		hostPort:                hostPort,
 		requestURI:              requestURI,
 		setHeaders:              func(req *http.Request) { sw.AuthConfig.SetHeaders(req, true) },
 		setProxyHeaders:         setProxyHeaders,
@@ -218,7 +224,7 @@ func (c *client) ReadData(dst []byte) ([]byte, error) {
 	deadline := time.Now().Add(c.hc.ReadTimeout)
 	req := fasthttp.AcquireRequest()
 	req.SetRequestURI(c.requestURI)
-	req.Header.SetHost(c.host)
+	req.Header.SetHost(c.hostPort)
 	// The following `Accept` header has been copied from Prometheus sources.
 	// See https://github.com/prometheus/prometheus/blob/f9d21f10ecd2a343a381044f131ea4e46381ce09/scrape/scrape.go#L532 .
 	// This is needed as a workaround for scraping stupid Java-based servers such as Spring Boot.

lib/promscrape/config.go

@@ -1143,7 +1143,7 @@ func (swc *scrapeWorkConfig) getScrapeWork(target string, extraLabels, metaLabel
 		droppedTargetsMap.Register(originalLabels)
 		return nil, nil
 	}
-	addressRelabeled = addMissingPort(schemeRelabeled, addressRelabeled)
+	addressRelabeled = addMissingPort(addressRelabeled, schemeRelabeled == "https")
 	metricsPathRelabeled := promrelabel.GetLabelValueByName(labels, "__metrics_path__")
 	if metricsPathRelabeled == "" {
 		metricsPathRelabeled = "/metrics"
@@ -1360,18 +1360,6 @@ func appendLabel(dst []prompbmarshal.Label, name, value string) []prompbmarshal.
 	})
 }
 
-func addMissingPort(scheme, target string) string {
-	if strings.Contains(target, ":") {
-		return target
-	}
-	if scheme == "https" {
-		target += ":443"
-	} else {
-		target += ":80"
-	}
-	return target
-}
-
 const (
 	defaultScrapeInterval = time.Minute
 	defaultScrapeTimeout  = 10 * time.Second

lib/promscrape/discoveryutils/client.go

@@ -48,6 +48,16 @@ type Client struct {
 	sendFullURL             bool
 }
 
+func addMissingPort(addr string, isTLS bool) string {
+	if strings.Contains(addr, ":") {
+		return addr
+	}
+	if isTLS {
+		return addr + ":443"
+	}
+	return addr + ":80"
+}
+
 // NewClient returns new Client for the given args.
 func NewClient(apiServer string, ac *promauth.Config, proxyURL *proxy.URL, proxyAC *promauth.Config) (*Client, error) {
 	var u fasthttp.URI
@@ -64,6 +74,7 @@ func NewClient(apiServer string, ac *promauth.Config, proxyURL *proxy.URL, proxy
 	}
 
 	hostPort := string(u.Host())
+	dialAddr := hostPort
 	isTLS := string(u.Scheme()) == "https"
 	var tlsCfg *tls.Config
 	if isTLS {
@@ -76,7 +87,7 @@ func NewClient(apiServer string, ac *promauth.Config, proxyURL *proxy.URL, proxy
 		// like net/http package from Go does.
 		// See https://en.wikipedia.org/wiki/Proxy_server#Web_proxy_servers
 		pu := proxyURL.GetURL()
-		hostPort = pu.Host
+		dialAddr = pu.Host
 		isTLS = pu.Scheme == "https"
 		if isTLS {
 			tlsCfg = proxyAC.NewTLSConfig()
@@ -87,13 +98,8 @@ func NewClient(apiServer string, ac *promauth.Config, proxyURL *proxy.URL, proxy
 		}
 		proxyURL = &proxy.URL{}
 	}
-	if !strings.Contains(hostPort, ":") {
+	hostPort = addMissingPort(hostPort, isTLS)
-		port := "80"
+	dialAddr = addMissingPort(dialAddr, isTLS)
-		if isTLS {
-			port = "443"
-		}
-		hostPort = net.JoinHostPort(hostPort, port)
-	}
 	if dialFunc == nil {
 		var err error
 		dialFunc, err = proxyURL.NewDialFunc(proxyAC)
@@ -102,7 +108,7 @@ func NewClient(apiServer string, ac *promauth.Config, proxyURL *proxy.URL, proxy
 		}
 	}
 	hc := &fasthttp.HostClient{
-		Addr:                hostPort,
+		Addr:                dialAddr,
 		Name:                "vm_promscrape/discovery",
 		IsTLS:               isTLS,
 		TLSConfig:           tlsCfg,
@@ -113,7 +119,7 @@ func NewClient(apiServer string, ac *promauth.Config, proxyURL *proxy.URL, proxy
 		Dial:                dialFunc,
 	}
 	blockingClient := &fasthttp.HostClient{
-		Addr:                hostPort,
+		Addr:                dialAddr,
 		Name:                "vm_promscrape/discovery",
 		IsTLS:               isTLS,
 		TLSConfig:           tlsCfg,