fix Could'nt read record from /sys/class

.gitignore

@@ -35,4 +35,4 @@ packaging/*/BUILDROOT/*
 packaging/*/BUILDROOT/*
 
 .vscode
-.vscode/*
+.vscode/*

docs_src/SUMMARY.md

@@ -32,6 +32,7 @@
 
 - [JSON exporter](references/exporter-json.md)
 - [Prometheus exporter](references/exporter-prometheus.md)
+- [Prometheus-push exporter](references/exporter-prometheuspush.md)
 - [Qemu exporter](references/exporter-qemu.md)
 - [Riemann exporter](references/exporter-riemann.md)
 - [Stdout exporter](references/exporter-stdout.md)

docs_src/references/exporter-prometheuspush.md

@@ -2,7 +2,7 @@
 
 ## Usage
 
-You can launch the prometheus exporter this way (running the default powercap_rapl sensor):
+You can launch the prometheus exporter this way:
 
 	scaphandre prometheus-push
 

docs_src/tutorials/installation-windows.md

@@ -4,7 +4,7 @@
 
 ## Using the installer
 
-Download the [package](https://scaphandre.s3.fr-par.scw.cloud/x86_64/scaphandre_0.5.0_installer.exe) and install it **as an administrator**.
+Download the latest exe installer [from the release page](https://github.com/hubblo-org/scaphandre/releases) and install it **as an administrator**.
 
 ### Configuring a Windows service to run Scaphandre in the background
 

docs_src/tutorials/kubernetes.md

@@ -21,6 +21,14 @@ to be installed from the source code.
 | `serviceMonitor.namespace`         | The namespace in which the ServiceMonitor will be created   (if not set, default to namespace on which this chart is installed)                                                    | `""`                      |
 | `serviceMonitor.interval`          | The interval at which metrics should be scraped                                                                 | `1m`                     |
 
+
+#### Security Parameters
+
+| Name                                       | Description                                                                                                     | Value                     |
+| ------------------------------------------ | --------------------------------------------------------------------------------------------------------------- | ------------------------- |
+| `securityContext.privileged`           | Gives full host access, bypassing container isolation (can be needed to run on physical server)                                    | `false`                    |
+| `securityContext.runAsUser`         | Run as root user to get proper permissions                                                     | `0`                      |
+| `securityContext.runAsGroup`          | Run as root group to get proper permissions                                                                  | `0`                     | 
 ## Install Prometheus
 
 Next we will install Prometheus which will scrape the metrics generated by Scaphandre.

helm/scaphandre/templates/daemonset.yaml

@@ -43,6 +43,8 @@ spec:
         ports:
         - name: metrics
           containerPort: {{ .Values.port }}
+        securityContext:
+          privileged: {{ .Values.securityContext.privileged }}
         resources:
 {{ toYaml .Values.resources | indent 10 }}
         volumeMounts:
@@ -53,8 +55,8 @@ spec:
           name: powercap
           readOnly: false
       securityContext:
-        runAsUser: {{ .Values.userID }}
-        runAsGroup: {{ .Values.userGroup }}
+        runAsUser: {{ .Values.securityContext.userID }}
+        runAsGroup: {{ .Values.securityContext.userGroup }}
       serviceAccountName: {{ template "scaphandre.name" . }}
       tolerations:
       # Tolerate all taints for observability

helm/scaphandre/values.yaml

@@ -18,9 +18,13 @@ scaphandre:
     containers:
 #  rustBacktrace: '1'
 
+
+securityContext:
+  # Gives full host access, bypassing container isolation. true or false
+  privileged: false
 # Run as root user to get proper permissions
-userID: 0
-groupID: 0
+  userID: 0
+  groupID: 0
 
 serviceMonitor:
   # Specifies whether ServiceMonitor for Prometheus operator should be created