{"id":6299,"date":"2019-07-26T18:59:25","date_gmt":"2019-07-26T18:59:25","guid":{"rendered":"https:\/\/www.devopsschool.com\/blog\/?p=6299"},"modified":"2021-11-16T11:11:21","modified_gmt":"2021-11-16T11:11:21","slug":"working-with-kubernetes-cluster-using-kubectl-part-11-probes","status":"publish","type":"post","link":"https:\/\/www.devopsschool.com\/blog\/working-with-kubernetes-cluster-using-kubectl-part-11-probes\/","title":{"rendered":"Working with Kubernetes Cluster using Kubectl Part – 11 – Probes"},"content":{"rendered":"\n

container-probes.yaml<\/strong><\/p>\n\n\n

apiVersion: apps\/v1\nkind: Deployment\nmetadata:\n  name: hello-world\nspec:\n  replicas: 1\n  selector:\n    matchLabels:\n      app: hello-world\n  template:\n    metadata:\n      labels:\n        app: hello-world\n    spec:\n      containers:\n      - name: hello-world\n        image: gcr.io\/google-samples\/hello-app:1.0\n        ports:\n        - containerPort: 8080\n        livenessProbe:\n          tcpSocket:\n            port: 8080\n          initialDelaySeconds: 10\n          periodSeconds: 5\n        readinessProbe:\n          httpGet:\n            path: \/\n            port: 8080\n          initialDelaySeconds: 10\n          periodSeconds: 5\n<\/code><\/span><\/pre>\n\n\n
Commands<\/strong><\/p>\n\n\n
\n\n#Start a watch to see the events associated with our probes.<\/span>\nkubectl get events --watch &\nclear\n\n#We have a single container pod app, in a Deployment that has both a liveness probe and a readiness probe<\/span>\nmore container-probes.yaml\n\n#Send in our deployment, after 10 seconds, our liveness and readiness probes will fail.<\/span>\n#The liveness probe will kill the current pod, and recreate one.<\/span>\nkubectl apply -f container-probes.yaml\n\n#kill our watch<\/span>\nfg\nctrl+c\n\n#We can see that our container isn't ready 0\/1 and it's Restarts are increasing.<\/span>\nkubectl get pods\n\n#Let's figure out what's wrong<\/span>\n#1. We can see in the events. The Liveness and Readiness probe failures.<\/span>\n#2. Under Containers, Liveness and Readiness, we can see the current configuration. And the current probe configuration. Both are pointing to 8081.<\/span>\n#3. Under Containers, Ready and Container Contidtions, we can see that the container isn't ready.<\/span>\n#3. Our Container Port is 8080, that's what we want our probes, probings. <\/span>\nkubectl describe pods\n\n#So let's go ahead and change the probes to 8080<\/span>\nvi container-probes.yaml\n\n#And send that change into the API Server for this deployment.<\/span>\nkubectl apply -f container-probes.yaml\n\n#Confirm our probes are pointing to the correct container port now, which is 8080.<\/span>\n \n#Let's check our status, a couple of things happened there.<\/span>\n#1. Our Deployment ReplicaSet created a NEW Pod, when we pushed in the new deployment configuration.<\/span>\n#2. It's not immediately ready because of our initialDelaySeconds which is 10 seconds.<\/span>\n#3. If we wait long enough, the livenessProbe will kill the original Pod and it will go away.<\/span>\n#4. Leaving us with the one pod in our Deployment's ReplicaSet<\/span>\nkubectl get pods \n\nkubectl delete deployment hello-world\n<\/code><\/span>Code language:<\/span> PHP<\/span> (<\/span>php<\/span>)<\/span><\/small><\/pre>\n\n