Add documentation

Author: MalloZup

collector/sbd/sbd.go

@@ -35,7 +35,7 @@ func NewCollector(sbdPath string, sbdConfigPath string) (*sbdCollector, error) {
 	}
 
 	c.SetDescriptor("devices", "SBD devices; one line per device", []string{"device", "status"})
-	c.SetDescriptor("sbd_timeout", "sbd watchdog timeout", []string{"device", "type"})
+	c.SetDescriptor("timeouts", "SBD timeouts for each device and type", []string{"device", "type"})
 
 	return c, nil
 }
@@ -73,11 +73,11 @@ func (c *sbdCollector) CollectWithError(ch chan<- prometheus.Metric) error {
 
 	sbdWatchdogs, sbdMsgWaits := c.getSbdTimeouts(sbdDevices)
 	for sbdDev, sbdWatchdog := range sbdWatchdogs {
-		ch <- c.MakeGaugeMetric("sbd_timeout", sbdWatchdog, sbdDev, "watchdog_timeout")
+		ch <- c.MakeGaugeMetric("timeouts", sbdWatchdog, sbdDev, "watchdog")
 	}
 
 	for sbdDev, sbdMsgWait := range sbdMsgWaits {
-		ch <- c.MakeGaugeMetric("sbd_timeout", sbdMsgWait, sbdDev, "msgwait_timeout")
+		ch <- c.MakeGaugeMetric("timeouts", sbdMsgWait, sbdDev, "msgwait")
 	}
 
 	return nil

doc/metrics.md

@@ -198,7 +198,8 @@ The status of each Corosync ring; `1` means healthy, `0` means faulty.
 The SBD subsystems collect devices stats by parsing its configuration and the output of `sbd --dump`.
 
 0. [Sample](../test/sbd.metrics)
-2. [`ha_cluster_sbd_devices`](#ha_cluster_sbd_devices)
+1. [`ha_cluster_sbd_devices`](#ha_cluster_sbd_devices)
+2. [`ha_cluster_sbd_timeouts`](#ha_cluster_sbd_timeouts)
 
 ### `ha_cluster_sbd_devices`
 
@@ -214,6 +215,18 @@ Either the value is `1`, or the line is absent altogether.
 
 The total number of lines for this metric will be the cardinality of `device`.
 
+### `ha_cluster_sbd_timeouts`
+
+#### Description
+
+The SBD timeouts pro SBD device
+Value is an integer expessing the timeout
+
+#### Labels
+
+- `device`: the path of the SBD device
+- `type`:  either `watchdog` or `msgwait`
+
 
 ## DRBD
 

test/sbd.metrics

@@ -2,9 +2,9 @@
 # TYPE ha_cluster_sbd_devices gauge
 ha_cluster_sbd_devices{device="/dev/vdc",status="healthy"} 1
 ha_cluster_sbd_devices{device="/dev/vdd",status="healthy"} 1
-# HELP ha_cluster_sbd_sbd_timeout sbd watchdog timeout
-# TYPE ha_cluster_sbd_sbd_timeout gauge
-ha_cluster_sbd_sbd_timeout{device="/dev/vdc",type="msgwait_timeout"} 10
-ha_cluster_sbd_sbd_timeout{device="/dev/vdc",type="watchdog_timeout"} 9
-ha_cluster_sbd_sbd_timeout{device="/dev/vdd",type="msgwait_timeout"} 10
-ha_cluster_sbd_sbd_timeout{device="/dev/vdd",type="watchdog_timeout"} 9
+# HELP ha_cluster_sbd_timeouts SBD timeouts for each device and type
+# TYPE ha_cluster_sbd_timeouts gauge
+ha_cluster_sbd_timeouts{device="/dev/vdc",type="msgwait"} 10
+ha_cluster_sbd_timeouts{device="/dev/vdc",type="watchdog"} 9
+ha_cluster_sbd_timeouts{device="/dev/vdd",type="msgwait"} 10
+ha_cluster_sbd_timeouts{device="/dev/vdd",type="watchdog"} 9