Operation Manager, ebay Agenda About ebay Our fleet Kubernetes makes magic at ebay Model + Controller How we model our datacenter Operation in large scale Q&A About ebay 177M Active buyers worldwide should play magic ...- 2010 2015- Now 2010- Now 2012- Now Bare metals Way to Kubernetes Search Grid Hadoop PoP Database Frontend VM Kubernetes plays magic api etcd kind: metadata: spec: control Rolling Update Quota Docker Image Volume RBAC Kubernetes models applications How about Kubernetes itself? How about the fleet Kubernetes running on? Onboard •Rack •NetworkDevice •ComputeAsset •Region/AZ

Using Kubernetes for handling second screen experience of european tv show About us Thomas Hector Head of IT Operations Key Qualifications ● 12+ years’ experience with development and operations Extensive hand-on experience with Kubernetes and Containerization @peytzco @Peytz & Co @Peytz & Co About Peytz & Co ● 5 offices ● +100 employees ● OpenSource ● Infrastructure projects ● Custom built nodes KOPS: Validate cluster Monitoring Prometheus autodiscovery Grafana dashboards Graylogs search and dashboards Performance test Performancetest: LOCUST ● Test as code ● Highly scalable

Gartner has long talked about the "80% rule": that 80 percent of IT budgets get spent simply "keeping the lights on” – The average data center cpu utilization is about 10% ��������� �� ������� segregating prod and non-prod workloads ����������� �������� ����������� �������� Retail spark search adds MR flink Sigma Fuxi Kubernetes YARN Node ���������� �� ��������� ������� ��� �����������

Research findings and the future of Dashboard 3. Where to learn more and how to get involved About Kubernetes Dashboard Dashboard 1.0 launched in 2016 github.com/kubernetes/dashboard SIG-UI Team upcoming features https://unsplash.com/ Features since 2016 KubeCon update ● Exec into pod ● Global search ● Login mechanism ● Settings page ● Support for Cron Jobs ● Redesigned resource creation ● Dashboard How do people use Dashboard today, and want to use it tomorrow? https://unsplash.com/ About the survey 163 Responses How would you rate your knowledge of Kubernetes? 您如何评价您对于Kubernetes的了解程度？

charts into Kubernetes, creating a new release for each installation. And to find new charts, you can search Helm chart repositories. Apache OpenWhisk Source event Trigger Rule Action Result stateful applications. Manages the deployment and scaling of a set of pods, and provides guarantees about the ordering and uniqueness of these Pods • A DaemonSet ensures that all (or some) Nodes run

Store • Leader based clustering • Can be clustered across Master Nodes • Contains all state known about cluster • Kubernetes Front-end Control Plane • Provides RESTful interface • Returns state objects Unstructured Data Logs Messages VMware vRealize Log Insight Log analytics, aggregation, and search Virtual Applications vRealize Ops, Log Insight For Comprehensive Visibility 32 UI and API Backend

Helm 安装 emqx 集群 Step1. 下载 emqx helm chart helm repo add emqx https://repos.emqx.io/charts helm search repo emqx helm fetch emqx/emqx tar xvf emqx-4.2.0.tgz Step2. 修改 StatefulSet.yaml vi e Step1. 下载 redis helm chart helm repo add bitnami https://charts.bitnami.com/bitnami helm search repo bitnami/redis helm fetch bitnami/redis-cluster tar zxvf redis-cluster-3.2.5.tgz Step2 Step1. 下载 mongodb helm chart helm repo add bitnami https://charts.bitnami.com/bitnami helm search repo mongodb helm fetch aliyun/mongodb-replicaset tar zxvf mongodb-replicaset-2.3.1.tgz

er 都需对docker daemon进行DNS配置。 "--dns=10.0.0.10 --dns=192.168.0.1 --dns-search \ default.svc.cluster.local --dns-search \ svc.cluster.local --dns-opt ndots:2 --dns-opt \ timeout:2 --dns-opt attempts:2

e的port 去访问目标服务 名称解析默认后缀为集群初始化时--cluster-domain指定的域名后缀，所以如果 pod里的/etc/resolv.conf的search后得指定默认查询的域，如果不对，则域名解 析会失败 search default.svc.cluster.local svc.cluster.local cluster.local ②NodePort类型 # vi com/charts # helm repo list NAME URL test-repo h�p://cof-lee.com/charts # helm search repo test-chart NAME CHART VERSION APP VERSION DESCRIPTION test-repo/test-chart

SLIs in a period of time. SLA = SLO + Punishment. SLA/SLO/SLI What we concern about Large k8s Cluster What happened about the cluster 1 Is there something unexpected happened in the cluster 2 What end Warning and Fatal are the possible value. 2. Success rate A rate value indicates the rate of success about creating/upgrading pod. 3. Number of Terminating Pod A number value indicates the count of pods that unexpected happened. Trace system： Collect and analyze logs in cluster. So we can known what happened about the cluster. Increase of SLO： Get the weakness of the cluster by analyzing the failure reasons and