Uptime and ICMP metrics

New: - Include uptime metrics - Include hosts ICMP metrics
monofox · Apr 14, 2024 · 52cbf7e · 52cbf7e
1 parent 32cce24
commit 52cbf7e
Show file tree

Hide file tree

Showing 4 changed files with 82 additions and 14 deletions.
diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml
@@ -6,6 +6,8 @@ name: Release monit_exporter
 on:
   push:
     tags: [ "*" ]
+  release:
+    types: [created]
 
 jobs:
 
@@ -22,7 +24,7 @@ jobs:
       - name: Set up Go
         uses: actions/setup-go@v3
         with:
-          go-version: 1.21.3
+          go-version: 1.21.9
 
       - name: Run GoReleaser
         uses: goreleaser/goreleaser-action@master

diff --git a/CHANGES.md b/CHANGES.md
@@ -10,7 +10,7 @@ The headers are:
 - Enhancements
 - Features
 
-## 0.3.0 (2024-04-xx)
+## 0.3.0 (2024-04-14)
 
 ### Bugs
 - Catching scrape errors
@@ -35,12 +35,15 @@ The headers are:
 - Added extraction of:
   - port response times
   - unix socket response times
+  - ICMP response times (hosts)
   - CPU usage
   - Memory usage
   - Disk write metrics
   - Disk read metrics
   - I/O service times
   - Network link metrics
+  - Uptime metrics
+  - Monit version information
 - Added option in order to ignore TLS certificate validation (restricted and not recommended)
 
 ## 0.2.2 (2023-10-22)

diff --git a/README.md b/README.md
@@ -34,16 +34,17 @@ These metrics are exported by `monit_exporter`:
 
 | name                                   | description                                                                                                                                                                                                                                        |
 |----------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
-| monit_service_check                    | Monit service check info with following labels provided:<br><dl><dt>`check_name`</dt><dd>Name of monit check</dd><dt>`monitored`</dt><dd>Specifies, if the service is monitored or not, whereas `0` means no and `1` means yes.</dd><dt>`type`</dt><dd>Specifies the type of service.</dd></dl>
+| monit_service_check                    | Monit service check info with following labels provided:<br><dl><dt>`check_name`</dt><dd>Name of monit check</dd><dt>`monitored`</dt><dd>Specifies, if the service is monitored or not, whereas `0` means no, `1` means yes, `2` means init and `4` means waiting. A combination is possible. E.g. `5` means, that the service is monitored, but currently waiting for e.g. the right time range.</dd><dt>`type`</dt><dd>Specifies the type of service.</dd></dl><br>Value of this metric is representing "error". Means: `0` is "service is fine" whereas `> 0` means, there is an error.
 | monit_service_cpu_perc                 | Monit service CPU info with following labels:<br><dl><dt>`check_name`</dt><dd>Name of monit check</dd><dt>`type`</dt><dd>Specifies value type whereas value can be `percentage` or `percentage_total`</dd></dl>
 | monit_service_mem_bytes                | Monit service mem info with following labels:<br><dl><dt>`check_name`</dt><dd>Name of monit check</dd><dt>`type`</dt><dd>Specifies value type whereas value can be `kilobyte` or `kilobyte_total`</dd></dl>
 | monit_service_network_link_state       | Monit service link states<br><dl><dt>`check_name`</dt><dd>Name of monit check</dd></dl><br>Value can be either `-1` = Not available, `0` = down and `1` = up
 | monit_service_network_link_statistics  | Monit service link statistics<br><dl><dt>`check_name`</dt><dd>Name of monit check</dd><dt>`direction`</dt><dd>Specifies link direction (upload / download)</dd><dt>`unit`</dt><dd>Spcifies unit of metrics (bytes, errors, packets)</dd><dt>`type`</dt><dd>Specifies the type with either now or total. Whereas now means "per second"</dd></dl>
-| monit_service_port_response_times      | Monit service port and unix socket checks response times<br><dl><dt>`check_name`</dt><dd>Name of monit check</dd><dt>`hostname`</dt><dd>Specifies hostname checked</dd><dt>`path`</dt><dd>Specifies a unix socket path</dd><dt>`port`</dt><dd>Specifies port to check</dd><dt>`protocol`</dt><dd>Specifies protocol used for checking service (e.g. POP, IMAP, REDIS, etc.). Default is a RAW check.</dd><dt>`type`</dt><dd>Specifies protocol type (e.g. TCP, UDP, UNIX)</dd><dt>`uri`</dt><dd>Gives full URI for the service check including type, host and port or path.</dd></dl>
+| monit_service_port_response_times      | Monit service port, unix socket and icmp checks response times<br><dl><dt>`check_name`</dt><dd>Name of monit check</dd><dt>`hostname`</dt><dd>Specifies hostname checked</dd><dt>`path`</dt><dd>Specifies a unix socket path</dd><dt>`port`</dt><dd>Specifies port to check</dd><dt>`protocol`</dt><dd>Specifies protocol used for checking service (e.g. POP, IMAP, REDIS, etc.). Default is a RAW check.</dd><dt>`type`</dt><dd>Specifies protocol type (e.g. TCP, UDP, UNIX, ICMP)</dd><dt>`uri`</dt><dd>Gives full URI for the service check including type, host and port or path.</dd></dl>
 | monit_service_read_bytes               | Monit service Disk Read Bytes<dl><dt>`check_name`</dt><dd>Name of monit check</dd><dt>`type`</dt><dd>Specifies type of read / write. Possible values: read_count, read_count_total. Value is given in bytes.</dd></dl>
+| monit_service_uptime                   | Service and server uptime in seconds<br><dl><dt>`check_name`</dt><dd>Name of monit check</dd><dt>`type`</dt><dd>Type of the uptime service check (Possible values: `system` / `server`)</dd></dl>
 | monit_service_write_bytes              | Monit service Disk Writes Bytes<dl><dt>`check_name`</dt><dd>Name of monit check</dd><dt>`type`</dt><dd>Specifies type of read / write. Possible values: write_count, write_count_total. Value is given in bytes.</dd></dl>
 | monit_up                               | Monit status availability. `0` = not available and `1` = available
-
+| monit_version                          | Monit current version as label - the value is const 1.
 
 #### Service types
 

diff --git a/monit_exporter.go b/monit_exporter.go
@@ -45,23 +45,32 @@ var serviceTypes = map[int]string{
 }
 
 type monitXML struct {
+	MonitServer   monitServer    `xml:"server"`
 	MonitServices []monitService `xml:"service"`
 }
 
+type monitServer struct {
+	Hostname string `xml:"localhostname"`
+	Uptime   int64  `xml:"uptime"`
+	Version  string `xml:"version"`
+}
+
 // Simplified structure of monit check.
 type monitService struct {
-	Type         int                `xml:"type,attr"`
-	Name         string             `xml:"name"`
-	Status       int                `xml:"status"`
-	Monitored    string             `xml:"monitor"`
-	Memory       monitServiceMem    `xml:"memory"`
 	CPU          monitServiceCPU    `xml:"cpu"`
-	DiskWrite    monitServiceDisk   `xml:"write"`
 	DiskRead     monitServiceDisk   `xml:"read"`
-	ServiceTimes monitServiceTime   `xml:"servicetime"`
+	DiskWrite    monitServiceDisk   `xml:"write"`
+	Link         monitServiceLink   `xml:"link"`
+	Memory       monitServiceMem    `xml:"memory"`
+	Monitored    string             `xml:"monitor"`
+	Name         string             `xml:"name"`
 	Ports        []monitServicePort `xml:"port"`
+	ServiceTimes monitServiceTime   `xml:"servicetime"`
+	Status       int                `xml:"status"`
+	Type         int                `xml:"type,attr"`
 	UnixSockets  []monitServicePort `xml:"unix"`
-	Link         monitServiceLink   `xml:"link"`
+	Uptime       int64              `xml:"uptime"`
+	Icmp         monitServiceIcmp   `xml:"icmp"`
 }
 
 type monitServiceMem struct {
@@ -110,6 +119,11 @@ type monitServiceLinkDirection struct {
 	Errors  monitNetworkCount `xml:"errors"`
 }
 
+type monitServiceIcmp struct {
+	Type         string  `xml:"type"`
+	Responsetime float64 `xml:"responsetime"`
+}
+
 type monitBytes struct {
 	Count int `xml:"count"`
 	Total int `xml:"total"`
@@ -127,6 +141,8 @@ type Exporter struct {
 	client *http.Client
 
 	up                 prometheus.Gauge
+	version            *prometheus.GaugeVec
+	checkUptime        *prometheus.GaugeVec
 	checkStatus        *prometheus.GaugeVec
 	checkMem           *prometheus.GaugeVec
 	checkCPU           *prometheus.GaugeVec
@@ -237,6 +253,20 @@ func NewExporter(c *Config) (*Exporter, error) {
 			Name:      "up",
 			Help:      "Monit status availability",
 		}),
+		version: prometheus.NewGaugeVec(prometheus.GaugeOpts{
+			Namespace: namespace,
+			Name:      "version",
+			Help:      "Monit version",
+		},
+			[]string{"version"},
+		),
+		checkUptime: prometheus.NewGaugeVec(prometheus.GaugeOpts{
+			Namespace: namespace,
+			Name:      "service_uptime",
+			Help:      "Monit service and server uptime",
+		},
+			[]string{"check_name", "type"},
+		),
 		checkStatus: prometheus.NewGaugeVec(prometheus.GaugeOpts{
 			Namespace: namespace,
 			Name:      "service_check",
@@ -300,6 +330,8 @@ func NewExporter(c *Config) (*Exporter, error) {
 // implements prometheus.Collector.
 func (e *Exporter) Describe(ch chan<- *prometheus.Desc) {
 	e.up.Describe(ch)
+	e.version.Describe(ch)
+	e.checkUptime.Describe(ch)
 	e.checkStatus.Describe(ch)
 	e.checkCPU.Describe(ch)
 	e.checkMem.Describe(ch)
@@ -326,6 +358,15 @@ func (e *Exporter) scrape() error {
 			log.Errorf("Error parsing data from monit: %v\n%s", err, data)
 		} else {
 			e.up.Set(1)
+			e.checkUptime.With(
+				prometheus.Labels{
+					"check_name": parsedData.MonitServer.Hostname,
+					"type":       "server",
+				}).Set(float64(parsedData.MonitServer.Uptime))
+			e.version.With(
+				prometheus.Labels{
+					"version": parsedData.MonitServer.Version,
+				}).Set(1)
 			// Constructing metrics
 			for _, service := range parsedData.MonitServices {
 				e.checkStatus.With(prometheus.Labels{"check_name": service.Name, "type": serviceTypes[service.Type], "monitored": service.Monitored}).Set(float64(service.Status))
@@ -336,7 +377,7 @@ func (e *Exporter) scrape() error {
 						"monitored":  service.Monitored,
 					}).Set(float64(service.Status))
 
-				// Memory + CPU only for specifiy status types (cf. monit/xml.c)
+				// Memory + CPU + Uptime only for specifiy status types (cf. monit/xml.c)
 				if service.Type == SERVICE_TYPE_PROCESS || service.Type == SERVICE_TYPE_SYSTEM {
 					e.checkMem.With(
 						prometheus.Labels{
@@ -358,6 +399,11 @@ func (e *Exporter) scrape() error {
 							"check_name": service.Name,
 							"type":       "percentage_total",
 						}).Set(float64(service.CPU.PercentTotal))
+					e.checkUptime.With(
+						prometheus.Labels{
+							"check_name": service.Name,
+							"type":       serviceTypes[service.Type],
+						}).Set(float64(service.Uptime))
 				}
 				if service.Type == SERVICE_TYPE_PROCESS || service.Type == SERVICE_TYPE_FILESYSTEM {
 					e.checkDiskWrite.With(
@@ -392,6 +438,20 @@ func (e *Exporter) scrape() error {
 					e.addNetLinkElement(&service, "upload", &service.Link.Upload)
 				}
 
+				// ICMP checks
+				if service.Type == SERVICE_TYPE_HOST && service.Icmp.Type != "" {
+					e.checkPortRespTimes.With(
+						prometheus.Labels{
+							"check_name": service.Name,
+							"type":       "ICMP",
+							"hostname":   "",
+							"path":       "",
+							"port":       "",
+							"protocol":   strings.ToUpper(service.Icmp.Type),
+							"uri":        "",
+						}).Set(float64(service.Icmp.Responsetime))
+				}
+
 				// Port checks
 				for _, port := range service.Ports {
 					var uri = fmt.Sprintf("%s://%s:%s", strings.ToLower(port.Type), port.Hostname, port.Portnumber)
@@ -458,6 +518,8 @@ func (e *Exporter) Collect(ch chan<- prometheus.Metric) {
 	e.checkStatus.Reset()
 	if err := e.scrape(); err == nil {
 		e.up.Collect(ch)
+		e.version.Collect(ch)
+		e.checkUptime.Collect(ch)
 		e.checkStatus.Collect(ch)
 		e.checkMem.Collect(ch)
 		e.checkCPU.Collect(ch)