various fixes

README.md

@@ -70,7 +70,7 @@ metrics:
 
 As shown in the above example, under `hosts` you can specify login information for individual hosts via their IP address, otherwise the exporter will attempt to use the login information under `default`. The login user only needs read-only permissions. Under `metrics` you can select what kind of metrics that should be returned, as described in more detail below.
 
-For a detailed description of the configuration, please see the [sample-config.yml](sample-config.yml) file.
+**For a detailed description of the configuration, please see the [sample-config.yml](sample-config.yml) file.**
 
 Because the metrics are collected on-demand it can take several minutes to scrape the metrics endpoint, depending on how many metrics groups are selected in the configuration file. For this reason you should carefully select the metrics of interest and make sure Prometheus is configured with a sufficiently high scrape timeout value.
 
@@ -125,7 +125,7 @@ idrac_power_control_avg_consumed_watts{id="0",name="System Power Control"} 166
 idrac_power_control_interval_in_minutes{id="0",name="System Power Control"} 1
 ```
 
-### Event Log
+### System Event Log
 This is not exactly an ordinary metric, but it is often convenient to be informed about new entries in the event log. The value of this metric is the unix timestamp for when the entry was created.
 
 ```text

charts/idrac-exporter/values.yaml

@@ -86,7 +86,6 @@ idracConfig: |
   address: 0.0.0.0   # Running in a container, this makes sense
   port: 9348         # Listen port
   timeout: 60        # HTTP timeout (in seconds) for Redfish API calls
-  retries: 10        # Number of retries before a target is marked as unreachable
   hosts:
     default:
       username: IDRAC_USERNAME
@@ -95,7 +94,7 @@ idracConfig: |
     system: true
     sensors: true
     power: true
-    sel: true        # iDRAC only
+    events: true
     storage: true
     memory: true
     network: true

go.mod

@@ -5,6 +5,7 @@ go 1.20
 require (
 	github.com/prometheus/client_golang v1.19.0
 	github.com/prometheus/common v0.48.0
+	github.com/xhit/go-str2duration/v2 v2.1.0
 	gopkg.in/yaml.v2 v2.4.0
 )
 

go.sum

@@ -16,6 +16,8 @@ github.com/prometheus/common v0.48.0/go.mod h1:0/KsvlIEfPQCQ5I2iNSAWKPZziNCvRs5E
 github.com/prometheus/procfs v0.12.0 h1:jluTpSng7V9hY0O2R9DzzJHYb2xULk9VTR1V1R/k6Bo=
 github.com/prometheus/procfs v0.12.0/go.mod h1:pcuDEFsWDnvcgNzo4EEweacyhjeA9Zk3cnaOZAZEfOo=
 github.com/rogpeppe/go-internal v1.10.0 h1:TMyTOH3F/DB16zRVcYyreMH6GnZZrwQVAoYjRBZyWFQ=
+github.com/xhit/go-str2duration/v2 v2.1.0 h1:lxklc02Drh6ynqX+DdPyp5pCKLUQpRT8bp8Ydu2Bstc=
+github.com/xhit/go-str2duration/v2 v2.1.0/go.mod h1:ohY8p+0f07DiV6Em5LKB0s2YpLtXVyJfNt1+BlmyAsU=
 golang.org/x/sys v0.18.0 h1:DBdB3niSjOA/O0blCZBqDefyWNYveAYMNF1Wum0DYQ4=
 golang.org/x/sys v0.18.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
 google.golang.org/protobuf v1.33.0 h1:uNO2rsAINq/JlFpSdYEKIZ0uKD/R9cpdv0T+yoGwGmI=

idrac.yml.template

@@ -1,7 +1,6 @@
 address: 0.0.0.0   # Running in a container, this makes sense
 port: 9348         # Listen port
 timeout: 60        # HTTP timeout (in seconds) for Redfish API calls
-retries: 10        # Number of retries before a target is marked as unreachable
 hosts:
   default:
     username: "$IDRAC_USERNAME"
@@ -10,7 +9,7 @@ metrics:
   system: true
   sensors: true
   power: true
-  sel: true
+  events: true
   storage: true
   memory: true
   network: true

internal/collector/client.go

@@ -124,6 +124,16 @@ func (client *Client) findAllEndpoints() error {
 		client.vendor = H3C
 	}
 
+	// Path for event log
+	switch client.vendor {
+	case DELL:
+		client.eventPath = "/redfish/v1/Managers/iDRAC.Embedded.1/Logs/Sel"
+	case LENOVO:
+		client.eventPath = "/redfish/v1/Systems/1/LogServices/PlatformLog/Entries"
+	case HPE:
+		client.eventPath = "/redfish/v1/Systems/1/LogServices/IML/Entries"
+	}
+
 	// Fix for Inspur bug
 	if client.vendor == INSPUR {
 		client.storagePath = strings.ReplaceAll(client.storagePath, "Storages", "Storage")
@@ -134,20 +144,11 @@ func (client *Client) findAllEndpoints() error {
 		if strings.Contains(root.Name, "HP RESTful") {
 			client.memoryPath = "/redfish/v1/Systems/1/Memory/"
 			client.storagePath = "/redfish/v1/Systems/1/SmartStorage/ArrayControllers/"
+			client.eventPath = ""
 			client.version = 4
 		}
 	}
 
-	// Path for event log
-	switch client.vendor {
-	case DELL:
-		client.eventPath = "/redfish/v1/Managers/iDRAC.Embedded.1/Logs/Sel"
-	case LENOVO:
-		client.eventPath = "/redfish/v1/Systems/1/LogServices/PlatformLog/Entries"
-	case HPE:
-		client.eventPath = "/redfish/v1/Systems/1/LogServices/IML/Entries"
-	}
-
 	return nil
 }
 

internal/config/config.go

@@ -4,9 +4,9 @@ import (
 	"math"
 	"os"
 	"strings"
-	"time"
 
 	"github.com/mrlhansen/idrac_exporter/internal/log"
+	"github.com/xhit/go-str2duration/v2"
 	"gopkg.in/yaml.v2"
 )
 
@@ -105,10 +105,10 @@ func ReadConfig(filename string) {
 	}
 
 	if Config.Event.MaxAge == "" {
-		Config.Event.MaxAge = "168h"
+		Config.Event.MaxAge = "7d"
 	}
 
-	t, err := time.ParseDuration(Config.Event.MaxAge)
+	t, err := str2duration.ParseDuration(Config.Event.MaxAge)
 	if err != nil {
 		log.Fatal("Invalid configuration: unable to parse duration: %v", err)
 	}

internal/config/env.go

@@ -49,8 +49,8 @@ func readConfigEnv() {
 	getEnvString("CONFIG_METRICS_PREFIX", &Config.MetricsPrefix)
 	getEnvString("CONFIG_DEFAULT_USERNAME", &username)
 	getEnvString("CONFIG_DEFAULT_PASSWORD", &password)
-	getEnvString("CONFIG_EVENT_SEVERITY", &Config.Event.Severity)
-	getEnvString("CONFIG_EVENT_MAXAGE", &Config.Event.MaxAge)
+	getEnvString("CONFIG_EVENTS_SEVERITY", &Config.Event.Severity)
+	getEnvString("CONFIG_EVENTS_MAXAGE", &Config.Event.MaxAge)
 
 	getEnvUint("CONFIG_PORT", &Config.Port)
 	getEnvUint("CONFIG_TIMEOUT", &Config.Timeout)

sample-config.yml

@@ -60,10 +60,9 @@ metrics:
   memory: false   # CONFIG_METRICS_MEMORY=true
   network: false  # CONFIG_METRICS_NETWORK=true
 
-# The events section is used for filtering events and it is only needed when
-# the "events" metrics are enabled. Events can be filtered based on minimum
-# severity and maximum age.
+# The events section is used for filtering events when the "events" metrics group
+# is enabled. Events can be filtered based on minimum severity and maximum age.
 # Severity must be one of "ok", "warning", "critical"
 events:
   severity: warning  # CONFIG_EVENTS_SEVERITY=warning
-  maxage: 168h       # CONFIG_EVENTS_MAXAGE=168h
+  maxage: 7d         # CONFIG_EVENTS_MAXAGE=7d