centos7安装与使用influxdb1.7.8

1 安装

# wget https://dl.influxdata.com/influxdb/releases/influxdb-1.7.8.x86_64.rpm
# sudo yum localinstall influxdb-1.7.8.x86_64.rpm 

2 启动

#启动
sudo systemctl start influxdb

#配置开机启动
sudo systemctl enable influxdb

3 防火墙配置

#开放端口
# firewall-cmd --add-port=8086/tcp --permanent

#重载防火墙配置
# firewall-cmd --reload

4 启动cli命令

[[e-mail] grafana]# influx
Connected to http://localhost:8086 version 1.7.8
InfluxDB shell version: 1.7.8
>

5 配置用户和数据库

创建用户
> CREATE USER dolphin WITH PASSWORD 'dolphininfluxpwd' WITH ALL PRIVILEGES

查询用户
> SHOW USERS
user    admin
----    -----
influx  true
dolphin true
>

创建数据库
> CREATE DATABASE dolphin

查询数据库
> SHOW DATABASES
name: databases
name
----
_internal
dolphin
>

6 插入数据与查询数据

1. database
   数据库，和mysql的数据库相比，没有太大的歧义

2. measurement
   对比的是mysql中的table，从实际体验来看，两个之间最明显的区别在于没有单独的创建measurement的方法，直接新增一条数据时，若measurement不存在，则直接创建并插入一条数据

3. Point
   这个对比的是mysql中的record，在influxDB中，表示每个表中，某个时刻，满足某个条件的filed数据（简单来说就是 timestamp + tag + filed)的组成一个point

timestamp : 时间戳，ns单位，每个记录都必然有这个属性，没有显示添加时，默认给一个。（纳秒，1s=1000ms 1ms=1000us 1us=1000ns）
tag: 标签，kv结构，在database中， tag + measurement 一起构建索引
参与索引创建，因此适合作为查询的过滤条件
tag的数据量不要太多，最好能有典型的辨别性（和mysql的建立索引的原则差不多）
value为String类型
tag是可选的，在measurement不设置tag也是ok的
field：存储数据，kv结构
数据类型为: long, String, boolean, float
4. Series
Series: tag key 与tag value的唯一组合

在insert执行语句中，tag与tag、field与field之间用都好进行分割，tag与field之间用空格分割
tag的value都是，String类型，不需要加双引号
field的String类型数据，需要放在双引号中，否则会报错
如果需要显示添加时间戳，在filed后添加空格，再添加时间戳

格式

<measurement>[,<tag-key>=<tag-value>...] <field-key>=<field-value>[,<field2-key>=<field2-value>...] [unix-nano-timestamp]

插入

> USE dolphin
Using database dolphin

插入数据（host是tag,load,usage是filed）
> INSERT cpu,host=192.168.1.1 load=0.1,usage=0.2
> INSERT cpu,host=192.168.1.2 load=0.2,usage=0.3
> INSERT cpu,host=192.168.1.3 load=0.6,usage=0.7

插入相同时间数据（tagkey1，tagkey2是tag,fieldkey1,fieldkey2是filed；tag的value都是，String类型，不需要加双引号
field的String类型数据，需要放在双引号中，否则会报错）
> insert table2,tagkey1=1,tagkey2=a fieldkey1=11,fieldkey2="aa" 1581771014614699790
> insert table2,tagkey1=2,tagkey2=a fieldkey1=11,fieldkey2="aa" 1581771014614699790
> select * from table2
name: table2
time                fieldkey1 fieldkey2 tagkey1 tagkey2
----                --------- --------- ------- -------
1581771014614699790 11        aa        1       a
1581771014614699790 11        aa        2       a

查询数据全部数据
> SELECT * FROM "cpu"
name: cpu
time                host        load usage
----                ----        ---- -----
1581756324805070467 192.168.1.1 0.1  0.2
1581756335749318117 192.168.1.2 0.2  0.3
1581756346893987133 192.168.1.3 0.6  0.7

# 筛选条件
> SELECT "host","load","usage" FROM "cpu" WHERE "host" = '192.168.1.1'
name: cpu
time                host        load usage
----                ----        ---- -----
1581756324805070467 192.168.1.1 0.1  0.2

# 范围
> SELECT "host","load","usage" FROM "cpu" WHERE "usage" > 0.3
name: cpu
time                host        load usage
----                ----        ---- -----
1581756346893987133 192.168.1.3 0.6  0.7
>

其它

# 创建数据库
CREATE DATABASE "db_name"
# 显示所有数据库
SHOW DATABASES
# 删除数据库
DROP DATABASE "db_name"

# 使用数据库
USE mydb
# 显示该数据库中的表
SHOW MEASUREMENTS
# 删除表
DROP MEASUREMENT "t_name"

7 HTTP API

7.1 创建数据库

[[e-mail] ~]# curl -i -XPOST http://172.16.159.132:8086/query --data-urlencode "q=CREATE DATABASE cctestdb"
HTTP/1.1 200 OK
Content-Type: application/json
Request-Id: 2ac0248e-4fd0-11ea-801e-000c29c94ec8
X-Influxdb-Build: OSS
X-Influxdb-Version: 1.7.8
X-Request-Id: 2ac0248e-4fd0-11ea-801e-000c29c94ec8
Date: Sat, 15 Feb 2020 08:50:07 GMT
Transfer-Encoding: chunked

{"results":[{"statement_id":0}]}
[[e-mail] ~]#

7.2 插入数据

#写入单条
curl -i -XPOST http://172.16.159.132:8086/write?db=cctestdb --data-binary "cpu,host=192.168.1.3 load=0.1,usage=0.33"

curl -i -XPOST http://172.16.159.132:8086/write?db=cctestdb --data-binary "cpu,host=192.168.1.3 load=0.1,usage=0.33 6666666666666666666"

#写入多条
curl -i -XPOST http://172.16.159.132:8086/write?db=cctestdb --data-binary "cpu,host=192.168.1.2 load=0.1,usage=0.22 1666666666666666661 
cpu,host=192.168.1.3 load=0.1,usage=0.33 1666666666666666663 
cpu,host=192.168.1.2 load=0.2,usage=0.22 1666666666666666664 
cpu,host=192.168.1.3 load=0.2,usage=0.33 1666666666666666665"

7.3查询数据

[[e-mail] ~]# curl -G http://172.16.159.132:8086/query?db=cctestdb --data-urlencode "q=SELECT * FROM  \"cpu\""
{"results":[{"statement_id":0,"series":[{"name":"cpu","columns":["time","host","load","usage"],"values":[["2020-02-15T08:52:00.039671928Z","192.168.1.3",0.1,0.33],["2022-10-25T02:57:46.666666661Z","192.168.1.2",0.1,0.22],["2022-10-25T02:57:46.666666661Z","192.168.1.3",0.1,0.33],["2022-10-25T02:57:46.666666662Z","192.168.1.2",0.2,0.22],["2022-10-25T02:57:46.666666662Z","192.168.1.3",0.2,0.33],["2022-10-25T02:57:46.666666663Z","192.168.1.3",0.1,0.33],["2022-10-25T02:57:46.666666664Z","192.168.1.2",0.2,0.22],["2022-10-25T02:57:46.666666665Z","192.168.1.3",0.2,0.33],["2181-04-04T11:51:06.666666666Z","192.168.1.3",0.1,0.33]]}]}]}
[[e-mail] ~]#

8 配置http账号密码

1 创建账号

创建用户
> CREATE USER dolphin WITH PASSWORD 'dolphininfluxpwd' WITH ALL PRIVILEGES

查询用户
> SHOW USERS
user    admin
----    -----
influx  true
dolphin true
>

2 配置`http密码访问。 auth-enabled = true
vim /etc/influxdb/influxdb.conf

[http]
  auth-enabled = true

3 cli终端账号密码登录

[[e-mail] grafana]# influx
Connected to http://localhost:8086 version 1.7.8
InfluxDB shell version: 1.7.8
> auth
username: dolphin
password:
> show databases;
name: databases
name
----
_internal
dolphin
cctestdb
itemdata
>

3 http账号密码登录,-u dolphin:dolphininfluxpwd

# curl -i -XPOST http://172.16.159.132:8086/write?db=dolphin -u dolphin:dolphininfluxpwd --data-binary "itemdatas,value=100, 1581772819000000000"

参考文档

《CentOS 7下InfluxDB部署与使用入门》：https://cloud.tencent.com/developer/article/1333874
《influxDB基本概念小结》：https://www.cnblogs.com/yihuihui/p/9374561.html

技术交流

CleverCode是一名架构师，技术交流，咨询问题，请加CleverCode创建的qq群(架构师俱乐部)：517133582。加群和腾讯，阿里，百度，新浪等公司的架构师交流。【架构师俱乐部】宗旨：帮助你成长为架构师！