Kubernetes deployment strategies
In Kubernetes there is few different way to release an application, you haveto carefully choose the right strategy to make your infrastructure resilient.
- recreate: terminate the old version and release the new one
- ramped: release a new version on a rolling update fashion, oneafter the other
- blue/green: release a new version alongside the old versionthen switch traffic
- canary: release a new version to a subset of users, then proceedto a full rollout
- a/b testing: release a new version to a subset of users in aprecise way (HTTP headers, cookie, weight, etc.). This doesn’t come out of thebox with Kubernetes, it imply extra work to setup a smarterloadbalancing system (Istio, Linkerd, Traeffik, custom nginx/haproxy, etc).
- shadow: release a new version alongside the old version. Incomingtraffic is mirrored to the new version and doesn't impact theresponse.
Before experimenting, checkout the following resources:
- CNCF prensentation
- CNCF prensentation slides
- Kubernetes deployment strategies
- Six Strategies for Application Deployment.
- Canary deployment using Istio and Helm
- Automated rollback of Helm releases based on logs or metrics
Getting started
These examples were created and tested on Minikuberunning with Kubernetes v1.10.0.
$ minikube start --kubernetes-version v1.10.0 --memory 8192 --cpus 2
Visualizing using Prometheus and Grafana
The following steps describe how to setup Prometheus and Grafana to visualizethe progress and performance of a deployment.
Install Helm
To install Helm, follow the instructions provided on theirwebsite.
$ helm init
Install Prometheus
$ helm install \
--namespace=monitoring \
--name=prometheus \
--version=7.0.0 \
stable/prometheus
Install Grafana
$ helm install \
--namespace=monitoring \
--name=grafana \
--version=1.12.0 \
--set=adminUser=admin \
--set=adminPassword=admin \
--set=service.type=NodePort \
stable/grafana
Setup Grafana
Now that Prometheus and Grafana are up and running, you can access Grafana:
$ minikube service grafana
To login, username: admin, password: admin.
Then you need to connect Grafana to Prometheus, to do so, add a DataSource:
Name: prometheus
Type: Prometheus
Url: http://prometheus-server
Access: Server
Create a dashboard with a Graph. Use the following query:
sum(rate(http_requests_total{app="my-app"}[5m])) by (version)
To have a better overview of the version, add {{version}} in the legend field.
Example graph
Recreate:
Ramped:
Blue/Green:
Canary:
A/B testing:
Shadow:
-
你能讲清楚污点和容忍度吗？ 当Kubernetes集群中的节点出现故障或需要维护时，管理员可能需要将工作负载从一个节点移到另一个节点。但是，将工作负载迁移到另一个节点可能会导致节点过载或资源不足，从而影响其他工作负载的性能。因此，Kubernetes提供了污点和容忍度的概念，用于控制工作负载可以在哪些节点上运行。 污点（Taint）是指在节点上设置一个标记，表示该节点“有问题”，即不适合运行某些工
-
k8s解决pod调度不均衡的问题 2020-07-31阅读 1.8K评论 1 问题及原因 k8s是通过sceduler来调度pod的，在调度过程中，由于一些原因，会出现调度不均衡的问题，例如： 节点故障新节点被加到集群中节点资源利用不足 这些都会导致pod在调度过程中分配不均，例如会造成节点负载过高，引发pod触发OOM等操作造成服务不可用 其中，节点资源利用不足时是最容易出现问题的，例如，设置的
-
Deployment 简述 Deployment 为 Pod 和 ReplicaSet 提供了一个声明式定义 (declarative) 方法，用来替代以前的 ReplicationController 更方便的管理应用。 作为最常用的 Kubernetes 对象，Deployment 经常会用来创建 ReplicaSet 和 Pod，我们往往不会直接在集群中使用 ReplicaSet 部署一个新
-
--- apiVersion: apps/v1 kind: Deployment metadata: name: ops-nginx-api # Deployment 对象的名称，与应用名称保持一致 namespace: default #命名空间 labels: appName: ops-nginx-api # 应用名称 spec: selector:
-
简述 Deployment 为 Pod 和 ReplicaSet 提供了一个声明式定义(declarative)方法，用来替代以前的ReplicationController 来方便的管理应用。典型的应用场景包括： 定义Deployment来创建Pod和ReplicaSet 滚动升级和回滚应用 扩容和缩容 暂停和继续Deployment 比如一个简单的nginx应用可以定义为 apiVersion
-
deployment 是一个使用 BASH 编写的一个软件发布程序，实现自动备份，增量发布，失败回撤，在线备份等等功能。
-
Deployment的使用 前面的课程中我们学习了Replication Controller和Replica Set两种资源对象，RC和RS的功能基本上是差不多的，唯一的区别就是RS支持集合的selector。我们也学习到了用RC/RS来控制Pod副本的数量，也实现了滚动升级Pod的功能。现在看上去似乎一切都比较完美的运行着，但是我们上节课最后也提到了现在我们推荐使用Deployment这种控制
-
通过使用Spring Boot应用程序，我们可以创建一个war文件以部署到Web服务器中。 在本章中，您将学习如何创建WAR文件并在Tomcat Web服务器中部署Spring Boot应用程序。 Spring Boot Servlet初始化程序 传统的部署方式是使Spring Boot Application @SpringBootApplication类扩展SpringBootServletI
-
This section covers running your application in production on a web server such as Apache or lighttpd. CGI Creating a .cgi file Server Setup mod_wsgi (Apache) Installing mod_wsgi Creating a .wsgi file
-
Depends on the Ingress Controller’s monitoring scope(cluster wide, single namespace base on ingress class), there are 3 types deployment architectures, BIG-IP used as external ingress controler for Ng