Kubernetes 1.5.0 was released just about a month ago! Key theme for the release are:
- StatefulSets (ex-PetSets)
- StatefulSets are beta now (fixes and stabilization)
- Improved Federation Support
- New command:
kubefed - DaemonSets
- Deployments
- ConfigMaps
- New command:
- Simplified Cluster Deployment
- Improvements to
kubeadm - HA Setup for Master
- Improvements to
- Node Robustness and Extensibility
- Windows Server Container support
- CRI for pluggable container runtimes
kubeletAPI supports authentication and authorization
Ler CHANGELOG for complete details. Up until 1.5.0, starting up a Kubernetes cluster on Amazon Web Services was pretty straight forward.
|
1 |
NUM_NODES=2 NODE_SIZE=m3.medium KUBERNETES_PROVIDER=aws ./aglomerado/kube–Para cima.sh |
But with 1.5.0 and 1.5.1, the command fails with the error:
|
1 2 3 4 5 |
... Iniciando aglomerado em us–west–2a usando provedor aws ... ligando verificar–pré-requisitos ... ligando kube–Para cima Iniciando aglomerado usando os distro: jessie !!! Cannot find Kubernetes–server–Linux–amd64.alcatrão.parabéns |
What happened? Basically, Kubernetes binaries was getting bigger than 1GB. The binary was broken into a basic install bundle and client and server binaries. The updated installation process requires to download the basic install bundle of of 4.57 MB (yes, MB instead of GB). It includes cluster scripts like kubectl, kube-up.sh e kube-down.sh, examples, docs and other scripts. This then downloads client and server binaries. Server binary is the base image that is used to start EC2 instances. But instead of automating the download of binaries, somebody decided to add a README in the server directory.
This was a big user experience change, and no links in the README bundled with the release or the release blog. Ouch!
Anyway, this was filed as #38728 and fixed promptly. But it missed the 1.5.1 release and now finally showed up in the 1.5.2 release today.
So, how do you run a Kubernetes 1.5.2 cluster on AWS? It is more seamlessly integrated now but you need to hit Enter key a couple of times to accept the default value:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 |
NUM_NODES=2 NODE_SIZE=m3.medium KUBERNETES_PROVIDER=aws ./aglomerado/kube–Para cima.sh ... Iniciando aglomerado em us–west–2a usando provedor aws ... ligando verificar–pré-requisitos ... ligando verificar–kube–binaries !!! kubectl appears para ser broken ou missing !!! Cannot find Kubernetes–server–Linux–amd64.alcatrão.parabéns Required binaries appear para ser missing. Fazer você wish para download them? [Y/n] Kubernetes lançamento: v1.5.2 Servidor: Linux/amd64 (para override, definir KUBERNETES_SERVER_ARCH) Client: darwin/amd64 (autodetected) Will download Kubernetes–server–Linux–amd64.alcatrão.parabéns de https://storage.googleapis.com/kubernetes-release/release/v1.5.2 Will download e extract Kubernetes–client–darwin–amd64.alcatrão.parabéns de https://storage.googleapis.com/kubernetes-release/release/v1.5.2 Is this ok? [Y]/n Warning: Keep–alive functionality somewhat crippled due para missing support em Warning: your operating system! % Total % Received % Xferd Average Speed Tempo Tempo Tempo Current Dload Upload Total Spent Left Speed 100 299M 100 299M 0 0 2132k 0 0:02:23 0:02:23 —:—:— 2439k md5sum(Kubernetes–server–Linux–amd64.alcatrão.parabéns)=7947bd430c4ffc358a6784e51c1d2b0f sha1sum(Kubernetes–server–Linux–amd64.alcatrão.parabéns)=4dbdcfa623412dac6be8fd5a4209a1f1423e8d30 Warning: Keep–alive functionality somewhat crippled due para missing support em Warning: your operating system! % Total % Received % Xferd Average Speed Tempo Tempo Tempo Current Dload Upload Total Spent Left Speed 100 22.0M 100 22.0M 0 0 1810k 0 0:00:12 0:00:12 —:—:— 2296k md5sum(Kubernetes–client–darwin–amd64.alcatrão.parabéns)=f55a8f9c300042e9b16e327ad2788521 sha1sum(Kubernetes–client–darwin–amd64.alcatrão.parabéns)=c29ab99e22146ba0a3da5c25de62ed13108b8ba9 Extracting /Users/arungupta/ferramentas/Kubernetes/Kubernetes–1.5.2/Kubernetes/client/Kubernetes–client–darwin–amd64.alcatrão.parabéns into /Users/arungupta/ferramentas/Kubernetes/Kubernetes–1.5.2/Kubernetes/platforms/darwin/amd64 Add ‘/Users/arungupta/tools/kubernetes/kubernetes-1.5.2/kubernetes/client/bin’ para your PATH para usar newly–instalado binaries. ... ligando kube–Para cima Iniciando aglomerado usando os distro: jessie Uploading para Amazon S3 ... |
After the usual Kubernetes cluster is created, the output is shown as:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 |
0 minions started; waiting 0 minions started; waiting 2 minions started; ready Waiting para aglomerado initialization. Este testamento continually check para see se o API para Kubernetes é reachable. Este might loop forever se there was some uncaught erro during start Para cima. .........................................................................................................................................................................Kubernetes aglomerado created. Sanity checking aglomerado... Attempt 1 para check Docker ligado nó @ 35.166.195.134 ...não working yet Attempt 2 para check Docker ligado nó @ 35.166.195.134 ...working Attempt 1 para check Docker ligado nó @ 35.166.188.211 ...não working yet Attempt 2 para check Docker ligado nó @ 35.166.188.211 ...working Kubernetes aglomerado é running. A master é running at: https://35.165.234.219 A usuário nome e senha para usar é located em /Users/arungupta/.kube/configuração. ... ligando validate–aglomerado Não resources found. Waiting para 2 ready nodes. 0 ready nodes, 0 registered. Retrying. Waiting para 2 ready nodes. 0 ready nodes, 2 registered. Retrying. Waiting para 2 ready nodes. 0 ready nodes, 2 registered. Retrying. Found 2 nó(s). NAME STATUS AGE ip–172–20–0–206.us–west–2.compute.internal Ready 45s ip–172–20–0–246.us–west–2.compute.internal Ready 42s Validate output: NAME STATUS MESSAGE ERRO controlador–manager Healthy ok scheduler Healthy ok etcd–0 Healthy {“health”: “true”} etcd–1 Healthy {“health”: “true”} Aglomerado validation succeeded Done, listing aglomerado serviços: Kubernetes master é running at https://35.165.234.219 Elasticsearch é running at https://35.165.234.219/api/v1/proxy/namespaces/kube-system/services/elasticsearch-logging Heapster é running at https://35.165.234.219/api/v1/proxy/namespaces/kube-system/services/heapster Kibana é running at https://35.165.234.219/api/v1/proxy/namespaces/kube-system/services/kibana-logging KubeDNS é running at https://35.165.234.219/api/v1/proxy/namespaces/kube-system/services/kube-dns Kubernetes–dashboard é running at https://35.165.234.219/api/v1/proxy/namespaces/kube-system/services/kubernetes-dashboard Grafana é running at https://35.165.234.219/api/v1/proxy/namespaces/kube-system/services/monitoring-grafana InfluxDB é running at https://35.165.234.219/api/v1/proxy/namespaces/kube-system/services/monitoring-influxdb To further depurar e diagnose aglomerado problems, usar ‘kubectl cluster-info dump’. |
Even though your Kubernetes cluster on AWS starts up fine, but kube-up.sh script is going to be deprecated soon. The recommended way is to use Kubernetes Cluster on Amazon using Kops. Now that your Kubernetes cluster is up, what do you do next?
- Follow the detailed steps for Kubernetes for Java Developers workshop.
- Run a Couchbase cluster in Kubernetes
- Saiba mais sobre Couchbase cluster in Containers
Deixe um comentário
Você precisa fazer o login para publicar um comentário.