Kubernetes安装ThingsBoard (on Google Cloud Platform)
This guide will help you to deploy ThingsBoard into a cluster on Google Cloud Platform using Kubernetes.
本章描述如何使用Kubernetes在谷歌云平台上将ThingsBoard部署到到一个集群中。
安装步骤
- Install Cloud SDK
- Make sure that you have Google Cloud Platform account activated and you have sufficient funds to create clusters and provision instances.
- Download the following files from thingsboard repo:
- **common.yaml - Kubernetes config file for common resources (StorageClass etc.).
- **cassandra.yaml - Kubernetes config file for Cassandra Service. By Default start 2 nodes.
- **zookeeper.yaml - Kubernetes config file for ZK Service. By Default start 3 nodes.
- **tb.yaml - Kubernetes config file for ThingsBoard Service. By Default starts 2 nodes.
- **cassandra-setup.yaml - Kubernetes config file for Pod that creates ThingsBoard keyspace and tables inside Cassandra storage.
- 安装Cloud SDK
- 确保您的Google Cloud Platform帐户被激活，并且您有足够的资金来创建集群以及置备实例。
- 从ThingsBoard repo中下载以下文件：
- common.yaml - 用于公共资源的Kubernetes配置文件(StorageClass等)。
- cassandra.yaml - 用于Cassandra服务的Kubernetes配置文件，默认启动2个节点。
- zookeeper.yaml - 用于ZK服务的Kubernetes配置文件，默认启动3个节点。
- tb.yaml - 用于ThingsBoard服务的Kubernetes配置文件，默认启动2个节点。
- cassandra-setup.yaml - 用于Pod（在Cassandr存储空间中创建ThingBoard密匙和表）的配置文件。
curl -L https://raw.githubusercontent.com/thingsboard/thingsboard/release-1.3/docker/k8s/common.yaml > common.yaml
curl -L https://raw.githubusercontent.com/thingsboard/thingsboard/release-1.3/docker/k8s/cassandra.yaml > cassandra.yaml
curl -L https://raw.githubusercontent.com/thingsboard/thingsboard/release-1.3/docker/k8s/zookeeper.yaml > zookeeper.yaml
curl -L https://raw.githubusercontent.com/thingsboard/thingsboard/release-1.3/docker/k8s/tb.yaml > tb.yaml
curl -L https://raw.githubusercontent.com/thingsboard/thingsboard/release-1.3/docker/k8s/cassandra-setup.yaml > cassandra-setup.yaml
- Check server config to get available kubernetes cluster versions:
检查服务器配置，获取可用的kubernetes集群版本:
gcloud container get-server-config
- The output will be similar to:
输出结果类似如下显示内容：
Fetching server config for us-east1-b
defaultClusterVersion: 1.6.4
defaultImageType: COS
validImageTypes:
- COS
- CONTAINER_VM
validMasterVersions:
- 1.7.1
- 1.6.7
- 1.6.4
validNodeVersions:
- 1.7.1
- 1.7.0
- 1.6.7
- 1.6.6
- 1.6.4
- 1.5.7
- 1.4.9
- Create cluster using gcloud cli (for demo purpose 6 standard nodes are going to be deployed for cluster pods). Select from the list above validMasterVersions that is higher than 1.6.2 (VALID_MASTER_VERSION) and create cluster:
使用gcloud cli创建集群（建议在集群pods中部署6个标准节点用作演示）。使用以上列表中validMasterVersions版本高于1.6.2的(VALID_MASTER_VERSION)并创建集群：
gcloud container clusters create YOUR_CLUSTER_NAME --cluster-version=VALID_MASTER_VERSION --node-labels=machinetype=tb --num-nodes=2
- Create additional node pool for Cassandra and Zookeeper PODs:
为Cassandra and Zookeeper PODs创建额外的节点池。
gcloud container node-pools create cassandra-pool --cluster=YOUR_CLUSTER_NAME --node-labels=machinetype=other --num-nodes=3 --disk-size=10
- Create the default credentials file on your local machine:
为本地机器创建默认凭证文件：
gcloud auth application-default login
- Create common resources that are used by other resources:
创建其他资源使用的公共资源:
kubectl create -f common.yaml
- Provision cassandra cluster:
配置cassandra集群：
kubectl create -f cassandra.yaml
- Monitor provisioning of Cassandra Pods by executing following command:
执行以下命令监控Cassandra Pods配置：
kubectl get pods -w -l app=cassandra
Please wait until all 2 Pods of Cassandra service become Running:
请等待Cassandra服务的2个Pods运行：
NAME READY STATUS RESTARTS AGE
cassandra-0 1/1 Running 0 5s
cassandra-1 1/1 Running 0 3s
- Once Cassandra Pods are running please provision Pod that will create ThingsBoard schema and tables inside Cassandra storage:
Pods运行时，请配置在Cassandra存储空间内创建ThingBoard shcema和表的pod：
kubectl create -f cassandra-setup.yaml
- Check logs of the cassandra setup node:
查阅cassandra设置节点日志：
kubectl logs -f cassandra-setup
- Wait for successful installation message:
等待安装成功信息:
Adding group `thingsboard' (GID 102) ...
Done.
Unpacking thingsboard (1.3.0-1) ...
Setting up thingsboard (1.3.0-1) ...
9042/tcp open unknown
Creating 'ThingsBoard' schema and system data...
plus demo data...
===================================================
:: ThingsBoard :: (v1.3.0)
===================================================
log4j:WARN No appenders could be found for logger (hsqldb.db.HSQLDB5D8345905E.ENGINE).
log4j:WARN Please initialize the log4j system properly.
log4j:WARN See http://logging.apache.org/log4j/1.2/faq.html#noconfig for more info.
Starting ThingsBoard Installation...
Installing DataBase schema...
Installing SQL DataBase schema...
Loading system data...
Loading demo data...
Installation finished successfully!
ThingsBoard installed successfully!
- Provision zookeeper Service:
kubectl create -f zookeeper.yaml
- Provision ThingsBoard Service:
kubectl create -f tb.yaml
- Execute the next command to see the status of ThingsBoard Pods:
执行如下命令查看ThingsBoard Pods状态：
kubectl get pods -w -l app=tb
ThingsBoard Pods should become into Running status:
ThingsBoard Pods应该变为运行状态：
NAME READY STATUS RESTARTS AGE
tb-0 1/1 Running 0 5s
tb-1 1/1 Running 0 3s
- Once ThingsBoard Pods are in ‘Running’ status check logs of the first Pod to make sure that everything is OK:
ThingsBoard Pods为运行状态时，请检查第一个Pod日志确保未出现任何异常：
kubectl logs -f tb-0
- ThingsBoard container is successfully started once similar line appears in the log:
如果日志中出现类似如下信息则表示ThingsBoard启动成功：
2016-12-13 13:44:52,407 [main] INFO o.t.s.ThingsboardServerApplication - Started ThingsboardServerApplication in 113.64 seconds (JVM running for 118.624)
To obtain external IP of ThingsBoard Service execute:
获得ThingsBoard服务的外部IP执行:
kubectl get services tb-service
- In the response something similar should appear:
回应信息应显示如下信息（类似）：
NAME CLUSTER-IP EXTERNAL-IP PORT(S) AGE
tb-service 10.3.251.137 35.185.81.65 8080:31099/TCP,1883:32314/TCP,5683:30062/TCP 1m
Now you are able to open Web UI using following link:
现在可以使用如下链接打开Web页面了：
http://EXTERNAL-IP:8080/
- To delete the entire cluster execute the following command:
执行以下命令删除整个集群：
gcloud container clusters delete YOUR_CLUSTER_NAME
高级应用
common.yaml file
Common Kubernetes config file common.yaml contains next set of cloud resources:
公共Kubernetes配置文件common.yaml包含下一组云资源：
- StorageClass. These resources are used by Cassandra and ZK Services as volume storage definitions
- StorageClass.
cassandra.yaml file
Cassandra Kubernetes config file cassandra.yaml contains next set of cloud resources:
- Headless Service. Service exposes 9042 port for the Cassandra Pods and guarantees network identity for them.
- StatefulSet. Set is responsible for provisioning Cassandra Pods onto Cloud Nodes.
- podAntiAffinity. This property guarantees that Cassandra Pods are deployed to different Nodes.
zookeeper.yaml file
Zookeeper Kubernetes config file zookeeper.yaml contains next set of cloud resources:
- Headless Service. Service exposes 2888 and 3888 ports for the ZK Pods and guarantees network identity for them.
- ConfigMap. Contains set of configurations options that are pushed into ZK images.
- PodDisruptionBudget. Guarantees minimum instances of the cluster that must be up and running.
- StatefulSet. Set is responsible for provisioning ZK Pods onto Cloud Nodes.
- podAntiAffinity. This property guarantees that ZK Pods are deployed to different Nodes.
tb.yaml file
ThingsBoard Kubernetes config file tb.yaml contains next set of cloud resources:
- Loadbalancer Service. Service that exposes 8080, 1883 and 5683 ports of ThingsBoard cluster to external world using external IP.
- ConfigMap. Contains set of configurations options that are pushed into ThingsBoard images.
- PodDisruptionBudget. Guarantees minimum instances of the cluster that must be up and running.
- StatefulSet. Set is responsible for provisioning ThingsBoard Pods onto Cloud Nodes.
- podAntiAffinity. This property guarantees that ThingsBoard Pods are deployed to different Nodes.
cassandra-setup.yaml
ThingsBoard Kubernetes config file cassandra-setup.yaml contains next set of cloud resources:
- Pod that runs once and creates ThingsBoard keyspace and tables inside Cassandra storage