Use configmap for node problem detector (#10548)

* use configmap

Signed-off-by: Hai Hui Wang <haihwang@cn.ibm.com>

* update version

Signed-off-by: Hai Hui Wang <haihwang@cn.ibm.com>

stable/node-problem-detector/Chart.yaml

@@ -1,5 +1,5 @@
 name: node-problem-detector
-version: "1.1.1"
+version: "1.1.2"
 appVersion: v0.6.1
 home: https://github.com/kubernetes/node-problem-detector
 description: Installs the node-problem-detector daemonset for monitoring extra attributes on nodes

stable/node-problem-detector/templates/configmap.yaml

@@ -0,0 +1,94 @@
+apiVersion: v1
+data:
+  kernel-monitor.json: |
+    {
+        "plugin": "kmsg",
+        "logPath": "/dev/kmsg",
+        "lookback": "5m",
+        "bufferSize": 10,
+        "source": "kernel-monitor",
+        "conditions": [
+            {
+                "type": "KernelDeadlock",
+                "reason": "KernelHasNoDeadlock",
+                "message": "kernel has no deadlock"
+            },
+            {
+                "type": "ReadonlyFilesystem",
+                "reason": "FilesystemIsReadOnly",
+                "message": "Filesystem is read-only"
+            }
+        ],
+        "rules": [
+            {
+                "type": "temporary",
+                "reason": "OOMKilling",
+                "pattern": "Kill process \\d+ (.+) score \\d+ or sacrifice child\\nKilled process \\d+ (.+) total-vm:\\d+kB, anon-rss:\\d+kB, file-rss:\\d+kB.*"
+            },
+            {
+                "type": "temporary",
+                "reason": "TaskHung",
+                "pattern": "task \\S+:\\w+ blocked for more than \\w+ seconds\\."
+            },
+            {
+                "type": "temporary",
+                "reason": "UnregisterNetDevice",
+                "pattern": "unregister_netdevice: waiting for \\w+ to become free. Usage count = \\d+"
+            },
+            {
+                "type": "temporary",
+                "reason": "KernelOops",
+                "pattern": "BUG: unable to handle kernel NULL pointer dereference at .*"
+            },
+            {
+                "type": "temporary",
+                "reason": "KernelOops",
+                "pattern": "divide error: 0000 \\[#\\d+\\] SMP"
+            },
+            {
+                "type": "permanent",
+                "condition": "KernelDeadlock",
+                "reason": "AUFSUmountHung",
+                "pattern": "task umount\\.aufs:\\w+ blocked for more than \\w+ seconds\\."
+            },
+            {
+                "type": "permanent",
+                "condition": "KernelDeadlock",
+                "reason": "DockerHung",
+                "pattern": "task docker:\\w+ blocked for more than \\w+ seconds\\."
+            },
+            {
+                "type": "permanent",
+                "condition": "ReadonlyFilesystem",
+                "reason": "FilesystemIsReadOnly",
+                "pattern": "Remounting filesystem read-only"
+            }
+        ]
+    }
+  docker-monitor.json: |
+    {
+        "plugin": "journald",
+        "pluginConfig": {
+            "source": "dockerd"
+        },
+        "logPath": "/var/log/journal",
+        "lookback": "5m",
+        "bufferSize": 10,
+        "source": "docker-monitor",
+        "conditions": [],
+        "rules": [
+            {
+                "type": "temporary",
+                "reason": "CorruptDockerImage",
+                "pattern": "Error trying v2 registry: failed to register layer: rename /var/lib/docker/image/(.+) /var/lib/docker/image/(.+): directory not empty.*"
+            }
+        ]
+    }
+kind: ConfigMap
+metadata:
+  name: {{ include "node-problem-detector.fullname" . }}
+  labels:
+    app.kubernetes.io/name: {{ include "node-problem-detector.name" . }}
+    helm.sh/chart: {{ include "node-problem-detector.chart" . }}
+    app.kubernetes.io/instance: {{ .Release.Name }}
+    app.kubernetes.io/managed-by: {{ .Release.Service }}

stable/node-problem-detector/templates/daemonset.yaml

@@ -31,7 +31,7 @@ spec:
           command:
             - "/bin/sh"
             - "-c"
-            - "exec /node-problem-detector --logtostderr --system-log-monitors={{- range $index, $monitor := .Values.settings.log_monitors }}{{if ne $index 0}},{{end}}{{ $monitor }}{{- end }}"
+            - "exec /node-problem-detector --logtostderr --system-log-monitors={{- range $index, $monitor := .Values.settings.log_monitors }}{{if ne $index 0}},{{end}}/config/{{ $monitor }}{{- end }}"
           securityContext:
             privileged: true
           env:
@@ -45,6 +45,9 @@ spec:
             - name: localtime
               mountPath: /etc/localtime
               readOnly: true
+            - name: config
+              mountPath: /config
+              readOnly: true
           terminationGracePeriodSeconds: 30
           resources:
 {{ toYaml .Values.resources | indent 12 }}
@@ -64,3 +67,11 @@ spec:
           hostPath:
             path: /etc/localtime
             type: "FileOrCreate"
+        - name: config
+          configMap:
+            name: {{ include "node-problem-detector.fullname" . }}
+            items:
+            {{- range $index, $monitor := .Values.settings.log_monitors }}
+            - key: {{ $monitor }}
+              path: {{ $monitor }}
+            {{- end }}

stable/node-problem-detector/values.yaml

@@ -1,7 +1,7 @@
 settings:
   log_monitors:
-    - /config/kernel-monitor.json
-    - /config/docker-monitor.json
+    - kernel-monitor.json
+    - docker-monitor.json
 
 hostpath:
   logdir: /var/log/