From d17cd1a5147ca7feb63b5d841f24527861b02912 Mon Sep 17 00:00:00 2001 From: MrHohn Date: Thu, 22 Sep 2016 18:02:30 -0700 Subject: [PATCH] Split dns healthcheck into two different urls --- cluster/addons/dns/skydns-rc.yaml.base | 23 ++++++--- cluster/addons/dns/skydns-rc.yaml.in | 22 +++++++-- cluster/addons/dns/skydns-rc.yaml.sed | 22 +++++++-- .../kube-manifests/addons/dns/skydns-rc.yaml | 46 ++++++++++++------ .../kubernetes/templates/kubedns-rc.yaml | 47 ++++++++++++------- 5 files changed, 112 insertions(+), 48 deletions(-) diff --git a/cluster/addons/dns/skydns-rc.yaml.base b/cluster/addons/dns/skydns-rc.yaml.base index cb60f0b2e6c..a8ad68e0032 100644 --- a/cluster/addons/dns/skydns-rc.yaml.base +++ b/cluster/addons/dns/skydns-rc.yaml.base @@ -35,7 +35,6 @@ spec: labels: k8s-app: kube-dns version: v19 - kubernetes.io/cluster-service: "true" annotations: scheduler.alpha.kubernetes.io/critical-pod: '' scheduler.alpha.kubernetes.io/tolerations: '[{"key":"CriticalAddonsOnly", "operator":"Exists"}]' @@ -55,7 +54,7 @@ spec: memory: 70Mi livenessProbe: httpGet: - path: /healthz + path: /healthz-kubedns port: 8080 scheme: HTTP initialDelaySeconds: 60 @@ -85,6 +84,15 @@ spec: protocol: TCP - name: dnsmasq image: gcr.io/google_containers/kube-dnsmasq-amd64:1.3 + livenessProbe: + httpGet: + path: /healthz-dnsmasq + port: 8080 + scheme: HTTP + initialDelaySeconds: 60 + timeoutSeconds: 5 + successThreshold: 1 + failureThreshold: 5 args: - --cache-size=1000 - --no-resolv @@ -98,7 +106,7 @@ spec: name: dns-tcp protocol: TCP - name: healthz - image: gcr.io/google_containers/exechealthz-amd64:1.1 + image: gcr.io/google_containers/exechealthz-amd64:1.2 resources: limits: memory: 50Mi @@ -110,9 +118,12 @@ spec: # net memory requested by the pod constant. memory: 50Mi args: - - -cmd=nslookup kubernetes.default.svc.__PILLAR__DNS__DOMAIN__ 127.0.0.1 >/dev/null && nslookup kubernetes.default.svc.__PILLAR__DNS__DOMAIN__ 127.0.0.1:10053 >/dev/null - - -port=8080 - - -quiet + - --cmd=nslookup kubernetes.default.svc.__PILLAR__DNS__DOMAIN__ 127.0.0.1 >/dev/null + - --url=/healthz-dnsmasq + - --cmd=nslookup kubernetes.default.svc.__PILLAR__DNS__DOMAIN__ 127.0.0.1:10053 >/dev/null + - --url=/healthz-kubedns + - --port=8080 + - --quiet ports: - containerPort: 8080 protocol: TCP diff --git a/cluster/addons/dns/skydns-rc.yaml.in b/cluster/addons/dns/skydns-rc.yaml.in index a047ebab003..d0a18a41822 100644 --- a/cluster/addons/dns/skydns-rc.yaml.in +++ b/cluster/addons/dns/skydns-rc.yaml.in @@ -55,7 +55,7 @@ spec: memory: 70Mi livenessProbe: httpGet: - path: /healthz + path: /healthz-kubedns port: 8080 scheme: HTTP initialDelaySeconds: 60 @@ -97,8 +97,17 @@ spec: - containerPort: 53 name: dns-tcp protocol: TCP + livenessProbe: + httpGet: + path: /healthz-dnsmasq + port: 8080 + scheme: HTTP + initialDelaySeconds: 60 + timeoutSeconds: 5 + successThreshold: 1 + failureThreshold: 5 - name: healthz - image: gcr.io/google_containers/exechealthz-amd64:1.1 + image: gcr.io/google_containers/exechealthz-amd64:1.2 resources: limits: memory: 50Mi @@ -110,9 +119,12 @@ spec: # net memory requested by the pod constant. memory: 50Mi args: - - -cmd=nslookup kubernetes.default.svc.{{ pillar['dns_domain'] }} 127.0.0.1 >/dev/null && nslookup kubernetes.default.svc.{{ pillar['dns_domain'] }} 127.0.0.1:10053 >/dev/null - - -port=8080 - - -quiet + - --cmd=nslookup kubernetes.default.svc.{{ pillar['dns_domain'] }} 127.0.0.1 >/dev/null + - --url=/healthz-dnsmasq + - --cmd=nslookup kubernetes.default.svc.{{ pillar['dns_domain'] }} 127.0.0.1:10053 >/dev/null + - --url=/healthz-kubedns + - --port=8080 + - --quiet ports: - containerPort: 8080 protocol: TCP diff --git a/cluster/addons/dns/skydns-rc.yaml.sed b/cluster/addons/dns/skydns-rc.yaml.sed index eb95574a3ea..55767766eeb 100644 --- a/cluster/addons/dns/skydns-rc.yaml.sed +++ b/cluster/addons/dns/skydns-rc.yaml.sed @@ -55,7 +55,7 @@ spec: memory: 70Mi livenessProbe: httpGet: - path: /healthz + path: /healthz-kubedns port: 8080 scheme: HTTP initialDelaySeconds: 60 @@ -96,8 +96,17 @@ spec: - containerPort: 53 name: dns-tcp protocol: TCP + livenessProbe: + httpGet: + path: /healthz-dnsmasq + port: 8080 + scheme: HTTP + initialDelaySeconds: 60 + timeoutSeconds: 5 + successThreshold: 1 + failureThreshold: 5 - name: healthz - image: gcr.io/google_containers/exechealthz-amd64:1.1 + image: gcr.io/google_containers/exechealthz-amd64:1.2 resources: limits: memory: 50Mi @@ -109,9 +118,12 @@ spec: # net memory requested by the pod constant. memory: 50Mi args: - - -cmd=nslookup kubernetes.default.svc.$DNS_DOMAIN 127.0.0.1 >/dev/null && nslookup kubernetes.default.svc.$DNS_DOMAIN 127.0.0.1:10053 >/dev/null - - -port=8080 - - -quiet + - --cmd=nslookup kubernetes.default.svc.$DNS_DOMAIN 127.0.0.1 >/dev/null + - --url=/healthz-dnsmasq + - --cmd=nslookup kubernetes.default.svc.$DNS_DOMAIN 127.0.0.1:10053 >/dev/null + - --url=/healthz-kubedns + - --port=8080 + - --quiet ports: - containerPort: 8080 protocol: TCP diff --git a/cluster/gce/coreos/kube-manifests/addons/dns/skydns-rc.yaml b/cluster/gce/coreos/kube-manifests/addons/dns/skydns-rc.yaml index a0c1b8c5655..cfa2f355f8b 100644 --- a/cluster/gce/coreos/kube-manifests/addons/dns/skydns-rc.yaml +++ b/cluster/gce/coreos/kube-manifests/addons/dns/skydns-rc.yaml @@ -1,23 +1,25 @@ apiVersion: v1 kind: ReplicationController metadata: - name: kube-dns-v15 + name: kube-dns-v19 namespace: kube-system labels: k8s-app: kube-dns - version: v15 + version: v19 kubernetes.io/cluster-service: "true" spec: replicas: ${DNS_REPLICAS} selector: k8s-app: kube-dns - version: v15 + version: v19 template: metadata: labels: k8s-app: kube-dns - version: v15 - kubernetes.io/cluster-service: "true" + version: v19 + annotations: + scheduler.alpha.kubernetes.io/critical-pod: '' + scheduler.alpha.kubernetes.io/tolerations: '[{"key":"CriticalAddonsOnly", "operator":"Exists"}]' spec: containers: - name: kubedns @@ -28,14 +30,13 @@ spec: # guaranteed class. Currently, this container falls into the # "burstable" category so the kubelet doesn't backoff from restarting it. limits: - cpu: 100m memory: 200Mi requests: cpu: 100m memory: 100Mi livenessProbe: httpGet: - path: /healthz + path: /healthz-kubedns port: 8080 scheme: HTTP initialDelaySeconds: 60 @@ -63,7 +64,16 @@ spec: name: dns-tcp-local protocol: TCP - name: dnsmasq - image: gcr.io/google_containers/dnsmasq:1.1 + image: gcr.io/google_containers/dnsmasq:1.3 + livenessProbe: + httpGet: + path: /healthz-dnsmasq + port: 8080 + scheme: HTTP + initialDelaySeconds: 60 + timeoutSeconds: 5 + successThreshold: 1 + failureThreshold: 5 args: - --cache-size=1000 - --no-resolv @@ -77,18 +87,24 @@ spec: name: dns-tcp protocol: TCP - name: healthz - image: gcr.io/google_containers/exechealthz-amd64:1.0 + image: gcr.io/google_containers/exechealthz-amd64:1.2 resources: - # keep request = limit to keep this container in guaranteed class limits: - cpu: 10m - memory: 20Mi + memory: 50Mi requests: cpu: 10m - memory: 20Mi + # Note that this container shouldn't really need 50Mi of memory. The + # limits are set higher than expected pending investigation on #29688. + # The extra memory was stolen from the kubedns container to keep the + # net memory requested by the pod constant. + memory: 50Mi args: - - -cmd=nslookup kubernetes.default.svc.${DNS_DOMAIN} 127.0.0.1 >/dev/null - - -port=8080 + - --cmd=nslookup kubernetes.default.svc.${DNS_DOMAIN} 127.0.0.1 >/dev/null + - --url=/healthz-dnsmasq + - --cmd=nslookup kubernetes.default.svc.${DNS_DOMAIN} 127.0.0.1:10053 >/dev/null + - --url=/healthz-kubedns + - --port=8080 + - --quiet ports: - containerPort: 8080 protocol: TCP diff --git a/cluster/juju/layers/kubernetes/templates/kubedns-rc.yaml b/cluster/juju/layers/kubernetes/templates/kubedns-rc.yaml index e05247de021..ef96d382447 100644 --- a/cluster/juju/layers/kubernetes/templates/kubedns-rc.yaml +++ b/cluster/juju/layers/kubernetes/templates/kubedns-rc.yaml @@ -12,30 +12,30 @@ # See the License for the specific language governing permissions and # limitations under the License. -# This file should be kept in sync with cluster/images/hyperkube/dns-rc.yaml - # Warning: This is a file generated from the base underscore template file: skydns-rc.yaml.base apiVersion: v1 kind: ReplicationController metadata: - name: kube-dns-v18 + name: kube-dns-v19 namespace: kube-system labels: k8s-app: kube-dns - version: v18 + version: v19 kubernetes.io/cluster-service: "true" spec: replicas: {{ pillar['dns_replicas'] }} selector: k8s-app: kube-dns - version: v18 + version: v19 template: metadata: labels: k8s-app: kube-dns - version: v18 - kubernetes.io/cluster-service: "true" + version: v19 + annotations: + scheduler.alpha.kubernetes.io/critical-pod: '' + scheduler.alpha.kubernetes.io/tolerations: '[{"key":"CriticalAddonsOnly", "operator":"Exists"}]' spec: containers: - name: kubedns @@ -46,14 +46,13 @@ spec: # guaranteed class. Currently, this container falls into the # "burstable" category so the kubelet doesn't backoff from restarting it. limits: - cpu: 100m memory: 200Mi requests: cpu: 100m memory: 100Mi livenessProbe: httpGet: - path: /healthz + path: /healthz-kubedns port: 8080 scheme: HTTP initialDelaySeconds: 60 @@ -84,6 +83,15 @@ spec: protocol: TCP - name: dnsmasq image: gcr.io/google_containers/kube-dnsmasq-{{ arch }}:1.3 + livenessProbe: + httpGet: + path: /healthz-dnsmasq + port: 8080 + scheme: HTTP + initialDelaySeconds: 60 + timeoutSeconds: 5 + successThreshold: 1 + failureThreshold: 5 args: - --cache-size=1000 - --no-resolv @@ -97,19 +105,24 @@ spec: name: dns-tcp protocol: TCP - name: healthz - image: gcr.io/google_containers/exechealthz-{{ arch }}:1.0 + image: gcr.io/google_containers/exechealthz-{{ arch }}:1.2 resources: - # keep request = limit to keep this container in guaranteed class limits: - cpu: 10m - memory: 20Mi + memory: 50Mi requests: cpu: 10m - memory: 20Mi + # Note that this container shouldn't really need 50Mi of memory. The + # limits are set higher than expected pending investigation on #29688. + # The extra memory was stolen from the kubedns container to keep the + # net memory requested by the pod constant. + memory: 50Mi args: - - -cmd=nslookup kubernetes.default.svc.{{ pillar['dns_domain'] }} 127.0.0.1 >/dev/null && nslookup kubernetes.default.svc.{{ pillar['dns_domain'] }} 127.0.0.1:10053 >/dev/null - - -port=8080 - - -quiet + - --cmd=nslookup kubernetes.default.svc.{{ pillar['dns_domain'] }} 127.0.0.1 >/dev/null + - --url=/healthz-dnsmasq + - --cmd=nslookup kubernetes.default.svc.{{ pillar['dns_domain'] }} 127.0.0.1:10053 >/dev/null + - --url=/healthz-kubedns + - --port=8080 + - --quiet ports: - containerPort: 8080 protocol: TCP