0%

Influx集群解决方案(Influx Proxy篇)_influxdb集群方案-CSDN博客

Excerpt

文章浏览阅读2.1k次,点赞14次,收藏25次。github地址Influx Proxy 是一个基于高可用、一致性哈希的 InfluxDB 集群代理服务,实现了 InfluxDB 高可用集群的部署方案,具有动态扩/缩容、故障恢复、数据同步等能力。连接到 Influx Proxy 和连接原生的 InfluxDB Server 没有显著区别(支持的查询语句列表),对上层客户端是透明的,上层应用可以像使用单机的 InfluxDB 一样使用,Influx Proxy会处理请求的转发,并对各个 InfluxDB 集群节点进行管理。_influxdb集群方案


InFluxDB 集群搭建

本次搭建使用influx proxy

介绍

github地址:https://github.com/chengshiwen/influx-proxy/

Influx Proxy 是一个基于高可用、一致性哈希的 InfluxDB 集群代理服务,实现了 InfluxDB 高可用集群的部署方案,
具有动态扩/缩容、故障恢复、数据同步等能力。连接到 Influx Proxy 和连接原生的 InfluxDB Server 没有显著区别
(支持的查询语句列表),对上层客户端是透明的,上层应用可以像使用单机的 InfluxDB 一样使用,Influx Proxy
会处理请求的转发,并对各个 InfluxDB 集群节点进行管理。Influx Proxy 基于饿了么开源的 Influx-Proxy,
并进一步开发和优化,支持了更多的特性,移除了 Python、Redis 依赖,解决了受限于一个数据库、需要额外配置
KEYMAPS 、数据负载不均衡的问题。

架构说明

  • 在改造我们的系统中我们相当于要实现以下步骤

    在这里插入图片描述

实现步骤

Influx1.8环境

Influx1.8+Influx Proxy +SpringBoot +Ngnix

SpringBoot搭建
  • 引入依赖

    1
    <dependency> <groupId>org.influxdb</groupId> <artifactId>influxdb-java</artifactId> <version>2.6</version> </dependency>
  • java代码(参考https://github.com/influxdata/influxdb-java)

    1
    // Create an object to handle the communication with InfluxDB. // (best practice tip: reuse the 'influxDB' instance when possible) final String serverURL = "http://127.0.0.1:8086", username = "root", password = "root"; final InfluxDB influxDB = InfluxDBFactory.connect(serverURL, username, password); // Create a database... // https://docs.influxdata.com/influxdb/v1.7/query_language/database_management/ String databaseName = "NOAA_water_database"; influxDB.query(new Query("CREATE DATABASE " + databaseName)); influxDB.setDatabase(databaseName); // ... and a retention policy, if necessary. // https://docs.influxdata.com/influxdb/v1.7/query_language/database_management/ String retentionPolicyName = "one_day_only"; influxDB.query(new Query("CREATE RETENTION POLICY " + retentionPolicyName + " ON " + databaseName + " DURATION 1d REPLICATION 1 DEFAULT")); influxDB.setRetentionPolicy(retentionPolicyName); // Enable batch writes to get better performance. influxDB.enableBatch( BatchOptions.DEFAULTS .threadFactory(runnable -> { Thread thread = new Thread(runnable); thread.setDaemon(true); return thread; }) ); // Close it if your application is terminating or you are not using it anymore. Runtime.getRuntime().addShutdownHook(new Thread(influxDB::close)); // Write points to InfluxDB. influxDB.write(Point.measurement("h2o_feet") .time(System.currentTimeMillis(), TimeUnit.MILLISECONDS) .tag("location", "santa_monica") .addField("level description", "below 3 feet") .addField("water_level", 2.064d) .build()); influxDB.write(Point.measurement("h2o_feet") .time(System.currentTimeMillis(), TimeUnit.MILLISECONDS) .tag("location", "coyote_creek") .addField("level description", "between 6 and 9 feet") .addField("water_level", 8.12d) .build()); // Wait a few seconds in order to let the InfluxDB client // write your points asynchronously (note: you can adjust the // internal time interval if you need via 'enableBatch' call). Thread.sleep(5_000L); // Query your data using InfluxQL. // https://docs.influxdata.com/influxdb/v1.7/query_language/data_exploration/#the-basic-select-statement QueryResult queryResult = influxDB.query(new Query("SELECT * FROM h2o_feet")); System.out.println(queryResult); // It will print something like: // QueryResult [results=[Result [series=[Series [name=h2o_feet, tags=null, // columns=[time, level description, location, water_level], // values=[ // [2020-03-22T20:50:12.929Z, below 3 feet, santa_monica, 2.064], // [2020-03-22T20:50:12.929Z, between 6 and 9 feet, coyote_creek, 8.12] // ]]], error=null]], error=null]
Ngnix搭建(搭建中)

请参考 custom.conf

服务器搭建与数据库部署

使用docker来搭建对应的influxdb信息

本地存在两台服务器(请确保influx1可以访问到influx2 同一局域网)

influx1: 192.168.137.130

influx2: 192.168.137.131

  1. docker-compose代码

    1
    version: "3.5" services: influx-proxy: image: chengshiwen/influx-proxy:latest container_name: influx-proxy ports: - 7076:7076 environment: - TZ=Asia/Shanghai volumes: - ./proxy.json:/etc/influx-proxy/proxy.json restart: unless-stopped networks: - influx_net influxdb-1: image: influxdb:1.8 container_name: influxdb-1 restart: unless-stopped networks: - influx_net volumes: - ./influxdb1/influxdb.conf:/etc/influxdb/influxdb.conf - ./influxdb1/meta:/var/lib/influxdb/meta - ./influxdb1/data:/var/lib/influxdb/data - ./influxdb1/wal:/var/lib/influxdb/wal ports: - "8086:8086" command: ["influxd", "--config", "/etc/influxdb/influxdb.conf"] influxdb-2: image: influxdb:1.8 container_name: influxdb-2 restart: unless-stopped networks: - influx_net volumes: - ./influxdb2/influxdb.conf:/etc/influxdb/influxdb.conf - ./influxdb2/meta:/var/lib/influxdb/meta - ./influxdb2/data:/var/lib/influxdb/data - ./influxdb2/wal:/var/lib/influxdb/wal ports: - "8087:8086" command: ["influxd", "--config", "/etc/influxdb/influxdb.conf"] networks: influx_net:
  2. proxy.json代码

    1
    { "circles": [ { "name": "circle-1", "backends": [ { "name": "influxdb-1-1", "url": "http://192.168.137.130:8086", "username": "", "password": "" }, { "name": "influxdb-1-2", "url": "http://192.168.137.130:8087", "username": "", "password": "" } ] }, { "name": "circle-2", "backends": [ { "name": "influxdb-2-1", "url": "http://192.168.137.131:8086", "username": "", "password": "" }, { "name": "influxdb-2-2", "url": "http://192.168.137.131:8087", "username": "", "password": "" } ] } ], "listen_addr": ":7076", "db_list": [], "data_dir": "data", "tlog_dir": "log", "hash_key": "idx", "flush_size": 10000, "flush_time": 1, "check_interval": 1, "rewrite_interval": 10, "conn_pool_size": 20, "write_timeout": 10, "idle_timeout": 10, "username": "", "password": "", "write_tracing": false, "query_tracing": false, "pprof_enabled": false, "https_enabled": false, "https_cert": "", "https_key": "" }
  3. influx.conf配置

    1
    #这里只放出几处需要修改的 其他按照默认即可 如果生产环境可以考虑把internal禁掉 # Determines whether the Flux query endpoint is enabled. flux-enabled = true (如果需要支持flux语句 请设置为true)
  4. 测试

    1
    curl -XPOST 'http://127.0.0.1:7076/query' --data-urlencode 'q=CREATE DATABASE "testdb"' sudo curl -XPOST 'http://127.0.0.1:7076/api/v2/write?bucket=testdb&precision=s' --data-binary 'mem,host=host1 used_percent=25 1700469476' sudo curl -XPOST 'http://127.0.0.1:7076/api/v2/write?bucket=testdb&precision=s' --data-binary 'mem2,host=host2 used_percent=23 1700469476' sudo curl -XPOST 'http://127.0.0.1:7076/api/v2/write?bucket=testdb&precision=s' --data-binary 'mem3,host=host3 used_percent=24 1700531670' sudo curl -XPOST 'http://127.0.0.1:7076/api/v2/query' \ -H 'Accept:application/csv' \ -H 'Content-type:application/vnd.flux' \ -d 'from(bucket:"testdb") |> range(start:-5m) |> filter(fn:(r) => r._measurement == "mem")'

Influx2.5环境

SpringBoot搭建
  • 引入依赖

    1
    <dependency> <groupId>com.influxdb</groupId> <artifactId>influxdb-client-java</artifactId> <version>6.3.0</version> </dependency>
  • Java代码

    1
    public class influxProxyTest { public static void main(String[] args) { String url = "http://192.168.137.130:7076"; String token = "testinfo"; String org = "admin"; String bucket = "analyse"; InfluxDBClient client = InfluxDBClientFactory.create(url, token.toCharArray(), org, bucket); QueryApi queryApi = client.getQueryApi(); String flux = "from(bucket:\"analyse\")\n" + "|> range(start:-5d)\n" + "|> filter(fn:(r) => r._measurement == \"mem\")"; List<FluxTable> list = queryApi.query(flux, org); for (FluxTable fluxTable:list){ List<FluxRecord> records = fluxTable.getRecords(); for (FluxRecord fluxRecord:records){ System.out.println(fluxRecord.getValue()); } } }
Ngnix搭建(实现中)

请参考 custom.conf

服务器搭建与数据库部署

使用docker来搭建对应的influxdb信息

本地存在两台服务器(请确保influx1可以访问到influx2 同一局域网)

influx1: 192.168.137.130

influx2: 192.168.137.131

  1. docker-compose代码

    1
    version: "3.5" services: influx-proxy: image: chengshiwen/influx-proxy:3.0.0-preview container_name: influx-proxy ports: - 7076:7076 environment: - TZ=Asia/Shanghai volumes: - ./proxy.json:/etc/influx-proxy/proxy.json restart: unless-stopped networks: - influx_net influxdb-1: image: influxdb:2.5.1 container_name: influxdb-1 restart: unless-stopped ports: - "8086:8086" networks: - influx_net volumes: - ./influxdb1:/var/lib/influxdb2 influxdb-2: image: influxdb:2.5.1 container_name: influxdb-2 restart: unless-stopped ports: - "8087:8086" networks: - influx_net volumes: - ./influxdb2:/var/lib/influxdb2 networks: influx_net:
  2. proxy.json代码

    1
    { "circles": [ { "name": "circle-1", "backends": [ { "name": "influxdb-1-1", "url": "http://192.168.137.130:8086", "token": "" }, { "name": "influxdb-1-2", "url": "http://192.168.137.130:8087", "token": "" } ] }, { "name": "circle-2", "backends": [ { "name": "influxdb-2-1", "url": "http://192.168.137.131:8086", "token": "" }, { "name": "influxdb-2-2", "url": "http://192.168.137.131:8087", "token": "" } ] } ], "dbrp": { "separator": "/", "mapping": {"mydb": "admin/analyse", "mydb/myrp": "admin/analyse"} }, "listen_addr": ":7076", "data_dir": "data", "flush_size": 10000, "flush_time": 1, "check_interval": 1, "rewrite_interval": 10, "conn_pool_size": 20, "write_timeout": 10, "write_tracing": false, "query_tracing": false, "token": "", "pprof_enabled": false, "https_enabled": false, "https_cert": "", "https_key": "" }
  3. 测试

    1
    数据写入 sudo curl -XPOST 'http://192.168.137.130:7076/api/v2/write?org=admin&bucket=analyse&precision=s' --data-binary 'mem,host=host3 used_percent=241 1700531671'