From c2529c1ea81958d8b87d5d7c8ebe9a66eec7b991 Mon Sep 17 00:00:00 2001
From: Carlos Rodriguez-Fernandez <carlosrodrifernandez@gmail.com>
Date: Thu, 28 Mar 2024 21:56:22 -0700
Subject: [PATCH] set up `healthy` and `reload` endpoints

* `/-/healthy` indicates that the http server successfully started, and
  consequentially the config was successfully parsed.
* `/-/reload` allows to attempt to reload the config, which is expected
  to be called by a sidecar that watches for config file changes

Signed-off-by: Carlos Rodriguez-Fernandez <carlosrodrifernandez@gmail.com>
---
 README.md |  7 ++++++-
 main.go   | 38 +++++++++++++++++++++++++++++++++++++-
 2 files changed, 43 insertions(+), 2 deletions(-)

diff --git a/README.md b/README.md
index 1ab61ad..f323b6a 100644
--- a/README.md
+++ b/README.md
@@ -61,7 +61,7 @@ In each host group the `interval`, `network`, and `protocol` are optional.
 
 The interval Duration is in [Go time.ParseDuration()](https://golang.org/pkg/time/#ParseDuration) syntax.
 
-The config is read on startup, and can be reloaded with the SIGHUP signal.
+The config is read on startup, and can be reloaded with the SIGHUP signal, or with an HTTP POST to the URI path `/-/reload`.
 
 ## Building and running
 
@@ -100,3 +100,8 @@ The Smokeping Prober supports TLS and basic authentication.
 To use TLS and/or basic authentication, you need to pass a configuration file
 using the `--web.config.file` parameter. The format of the file is described
 [in the exporter-toolkit repository](https://github.com/prometheus/exporter-toolkit/blob/master/docs/web-configuration.md).
+
+
+### Health check
+
+A health check can be requested in the URI path `/-/healthy`.
diff --git a/main.go b/main.go
index a358384..b8d23d6 100644
--- a/main.go
+++ b/main.go
@@ -264,20 +264,37 @@ func main() {
 
 	hup := make(chan os.Signal, 1)
 	signal.Notify(hup, syscall.SIGHUP)
+	reloadCh := make(chan chan error)
 	go func() {
 		for {
-			<-hup
+			var errCallback func(e error)
+			var successCallback func()
+			select {
+			case <-hup:
+				errCallback = func(e error) {}
+				successCallback = func() {}
+			case rc := <-reloadCh:
+				errCallback = func(e error) {
+					rc <- e
+				}
+				successCallback = func() {
+					rc <- nil
+				}
+			}
 			if err := sc.ReloadConfig(*configFile); err != nil {
 				level.Error(logger).Log("msg", "Error reloading config", "err", err)
+				errCallback(err)
 				continue
 			}
 			err = smokePingers.prepare(hosts, interval, privileged, sizeBytes)
 			if err != nil {
 				level.Error(logger).Log("msg", "Unable to create ping from config", "err", err)
+				errCallback(err)
 				continue
 			}
 			if smokePingers.sizeOfPrepared() == 0 {
 				level.Error(logger).Log("msg", "No targets specified on command line or in config file")
+				errCallback(fmt.Errorf("no targets specified"))
 				continue
 			}
 
@@ -285,10 +302,29 @@ func main() {
 			smokepingCollector.updatePingers(smokePingers.started, *pingResponseSeconds)
 
 			level.Info(logger).Log("msg", "Reloaded config file")
+			successCallback()
 		}
 	}()
 
 	http.Handle(*metricsPath, promhttp.Handler())
+	http.HandleFunc("/-/healthy", func(w http.ResponseWriter, r *http.Request) {
+		w.WriteHeader(http.StatusOK)
+		w.Write([]byte("Healthy"))
+	})
+	http.HandleFunc("/-/reload",
+		func(w http.ResponseWriter, r *http.Request) {
+			if r.Method != "POST" {
+				w.WriteHeader(http.StatusMethodNotAllowed)
+				fmt.Fprintf(w, "This endpoint requires a POST request.\n")
+				return
+			}
+
+			rc := make(chan error)
+			reloadCh <- rc
+			if err := <-rc; err != nil {
+				http.Error(w, fmt.Sprintf("Failed to reload config: %s", err), http.StatusInternalServerError)
+			}
+		})
 	if *metricsPath != "/" && *metricsPath != "" {
 		landingConfig := web.LandingConfig{
 			Name:        "Smokeping Prober",