From 70e7a10dcb5dbe5feacd066360196305e3e1890a Mon Sep 17 00:00:00 2001 From: Staffan Olsson Date: Sat, 3 Feb 2018 21:06:31 +0100 Subject: Just checking how it renders --- README.md | 101 ++++++++++++++++++++++++-------------------------------------- 1 file changed, 39 insertions(+), 62 deletions(-) diff --git a/README.md b/README.md index 0f92d64..4971a84 100644 --- a/README.md +++ b/README.md @@ -1,78 +1,55 @@ -_Manifests here require Kubernetes 1.8 now. -On earlier versions use [v2.1.0](https://github.com/Yolean/kubernetes-kafka/tree/v2.1.0)._ +# Kafka for Kubernetes -# Kafka on Kubernetes +This community seeks to provide: + * Production-worthy setup of Kafka for persistent data, domain and ops, at small scale. + * Operational knowledge, skewed towards resilience rather than throughput, encoded into Kubernetes manifest. + * A platform for streaming/event-driven microservices design using Kubernetes. -Transparent Kafka setup that you can grow with. -Good for both experiments and production. +We suggest you `apply` manifests in the following order: + * You choice of storage classes from [./configure/](./configure). + * [./rbac-namespace-default](./rbac-namespace-default/) + * [./zookeeper](./zookeeper/) + * [./kafka](./kafka/) -How to use: - * Good to know: you'll likely want to fork this repo. It prioritizes clarity over configurability, using plain manifests and .propeties files; no client side logic. - * Run a Kubernetes cluster, [minikube](https://github.com/kubernetes/minikube) or real. - * Quickstart: use the `kubectl apply`s below. - * Have a look at [addon](https://github.com/Yolean/kubernetes-kafka/labels/addon)s, or the official forks: - - [kubernetes-kafka-small](https://github.com/Reposoft/kubernetes-kafka-small) for single-node clusters like Minikube. - - [StreamingMicroservicesPlatform](https://github.com/StreamingMicroservicesPlatform/kubernetes-kafka) Like Confluent's [platform quickstart](https://docs.confluent.io/current/connect/quickstart.html) but for Kubernetes. - * Join the discussion in issues and PRs. +That'll give you client "bootstrap" `bootstrap.kafka.svc.cluster.local:9092`. -No readable readme can properly introduce both [Kafka](http://kafka.apache.org/) and [Kubernetes](https://kubernetes.io/), -but we think the combination of the two is a great backbone for microservices. -Back when we read [Newman](http://samnewman.io/books/building_microservices/) we were beginners with both. -Now we've read [Kleppmann](http://dataintensive.net/), [Confluent](https://www.confluent.io/blog/) and [SRE](https://landing.google.com/sre/book.html) and enjoy this "Streaming Platform" lock-in :smile:. +## Fork -We also think the plain-yaml approach of this project is easier to understand and evolve than [helm](https://github.com/kubernetes/helm) [chart](https://github.com/kubernetes/charts/tree/master/incubator/kafka)s. +Our only dependency is `kubectl`. Not because we hate Helm or Operators, but because we think plain manifests make it easier to collaborate. +If you begin to rely on this kafka setup we recommend you fork, for example to edit [broker config](). -## What you get +## Version history -Keep an eye on `kubectl --namespace kafka get pods -w`. +| tag | k8s >= | highlights | +|------|----|| +| v3.1 | 1.8|The painstaking path to `min.insync.replicas`=2| +| v3.0 | 1.8|| +| v2.1 | 1.5 || +| v2.0 | 1.5 |[addon](https://github.com/Yolean/kubernetes-kafka/labels/addon)s| +| v1.0 | 1 |Stateful? In Kubernetes? 2016? Yes.| -The goal is to provide [Bootstrap servers](http://kafka.apache.org/documentation/#producerconfigs): `kafka-0.broker.kafka.svc.cluster.local:9092,kafka-1.broker.kafka.svc.cluster.local:9092,kafka-2.broker.kafka.svc.cluster.local:9092` -` +## Monitoring -Zookeeper at `zookeeper.kafka.svc.cluster.local:2181`. +Have a look at: + * [./prometheus](./prometheus/) + * [./linkedin-burrow](./linkedin-burrow/) + * [or plain JMX](https://github.com/Yolean/kubernetes-kafka/pull/96) + * what's happening in the [monitoring](https://github.com/Yolean/kubernetes-kafka/labels/monitoring) label. + * Note that this repo is intentionally light on [automation](https://github.com/Yolean/kubernetes-kafka/labels/automation). We think every SRE team must build the operational knowledge first. -## Prepare storage classes +## Outside (out-of-cluster) access -For Minikube run `kubectl apply -f configure/minikube-storageclass-broker.yml; kubectl apply -f configure/minikube-storageclass-zookeeper.yml`. + * [Brokers](./outside-services/) -There's a similar setup for AKS under `configure/aks-*` and for GKE under `configure/gke-*`. You might want to tweak it before creating. +## Fewer than three nodes? -## Start Zookeeper +For minikube, youkube etc have a look at: -The [Kafka book](https://www.confluent.io/resources/kafka-definitive-guide-preview-edition/) recommends that Kafka has its own Zookeeper cluster with at least 5 instances. + * [Scale 1](https://github.com/Yolean/kubernetes-kafka/pull/44) + * [Scale 2](https://github.com/Yolean/kubernetes-kafka/pull/118) -``` -kubectl apply -f ./zookeeper/ -``` +## Stream... -To support automatic migration in the face of availability zone unavailability we mix persistent and ephemeral storage. - -## Start Kafka - -``` -kubectl apply -f ./kafka/ -``` - -You might want to verify in logs that Kafka found its own DNS name(s) correctly. Look for records like: -``` -kubectl -n kafka logs kafka-0 | grep "Registered broker" -# INFO Registered broker 0 at path /brokers/ids/0 with addresses: PLAINTEXT -> EndPoint(kafka-0.broker.kafka.svc.cluster.local,9092,PLAINTEXT) -``` - -That's it. Just add business value :wink:. - -## RBAC - -For clusters that enforce [RBAC](https://kubernetes.io/docs/admin/authorization/rbac/) there's a minimal set of policies in -``` -kubectl apply -f rbac-namespace-default/ -``` - -## Tests - -Tests are based on the [kube-test](https://github.com/Yolean/kube-test) concept. -Like the rest of this repo they have `kubectl` as the only local dependency. - -Run self-tests or not. They do generate some load, but indicate if the platform is working or not. - * To include tests, replace `apply -f` with `apply -R -f` in your `kubectl`s above. - * Anything that isn't READY in `kubectl get pods -l test-type=readiness --namespace=test-kafka` is a failed test. + * [Kubernetes events to kafka](./events-kube/) + * [Container logs to Kafka](https://github.com/Yolean/kubernetes-kafka/pull/131) + * [Heapster metrics to Kafka](https://github.com/Yolean/kubernetes-kafka/pull/120) -- cgit v1.2.3