开源时序服务器influxdb使用

文档

https://influxdb.com/docs/v0.9/introduction/overview.html

配置文件

/etc/opt/influxdb/influxdb.conf

reporting-disabled = false

[meta]
  dir = "/var/opt/influxdb/meta"
  hostname = "localhost"
  bind-address = ":8088"
  retention-autocreate = true
  election-timeout = "1s"
  heartbeat-timeout = "1s"
  leader-lease-timeout = "500ms"
  commit-timeout = "50ms"

[data]
  dir = "/var/opt/influxdb/data"
  MaxWALSize = 104857600 # Maximum size the WAL can reach before a flush. Defaults to 100MB.
  WALFlushInterval = "10m" # Maximum time data can sit in WAL before a flush.
  WALPartitionFlushDelay = "2s" # The delay time between each WAL partition being flushed.

[cluster]
  shard-writer-timeout = "5s"

[retention]
  enabled = true
  check-interval = "10m"

[admin]
  enabled = true
  bind-address = ":8083"

[http]
  enabled = true
  bind-address = ":8086"
  auth-enabled = false
  log-enabled = true
  write-tracing = false
  pprof-enabled = false

[[graphite]]
  enabled = false
  # bind-address = ":2003"
  # protocol = "tcp"
  # consistency-level = "one"
  # name-separator = "."
  ## e.g. "type.host.measurement.device" will parse "server.localhost.cpu.cpu0" as
  ## {
  ##     measurement: "cpu",
  ##     tags: {
  ##         "type": "server",
  ##         "host": "localhost,
  ##         "device": "cpu0"
  ##     }
  ## }
  # name-schema = "type.host.measurement.device"
  # ignore-unnamed = true

[collectd]
  enabled = false
  # bind-address = ""
  # database = ""
  # typesdb = ""

[opentsdb]
  enabled = false
  # bind-address = ""
  # database = ""
  # retention-policy = ""

[udp]
  enabled = false
  # bind-address = ""
  # database = ""
  # batch-size = 0
  # batch-timeout = "0"

[monitoring]
  enabled = true
  write-interval = "24h"

[continuous_queries]
  enabled = true
  recompute-previous-n = 2
  recompute-no-older-than = "10m"
  compute-runs-per-interval = 10
  compute-no-more-than = "2m"

[hinted-handoff]
  enabled = true
  dir = "/var/opt/influxdb/hh"
  max-size = 1073741824
  max-age = "168h"
  retry-rate-limit = 0
  retry-interval = "

协议

1. influxdb的数据由时间序列(time series)组织，包含度量值(measurement),像cpu_load, temperature等等。2
2. 时间序列有0个或多个记录点(points)，每个指标一个离散样本。
3. points包含timestamp和measurement，至少一个key-value域(fields)，例如 value=0.64 or 15min=0.78，0或多个key-value标签(tags)，包含元数据(metadata)，例如host=server01，region=EMEA，dc=Frankfurt。
4. 从概念上讲，可以把measurement想象是一个sql表，row行是主索引，tags和fields是表中的列，tag是索引，fileds不是。

points example

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

cpu,host=serverA,region=us_west value=0.64 payment,device=mobile,product=Notepad,method=credit billed=33,licenses=3 1434067467100293230 stock,symbol=AAPL bid=127.46,ask=127.48  temperature,machine=unit42,type=assembly external=25,internal=37 1434067467000000000

接口

CLI

# 建立数据库
create database <name>

# 使用数据库
use <name>

# 显示所有存在数据库
show databases;
name: databases
---------------
name
collectd
test

# 建立保留策略（retention policy）
create retention policy rp_collectd on collectd duration 1w replication 1 default
duration: 1h, 90m, 12h, 7d, 4w, INF 最小1h, 最大INF

# 显示保留策略
show retention policies on dashboard
name		duration	replicaN	default
default		0		1		false
rp_dashboard	168h0m0s	1		true

# 修改保留策略
alter retention policy rp_collectd on collectd duration 2d

# 当前数据库插入数据，measure是cpu, tag是host..., 值是0.64
insert cpu,host=serverA,region=us_west value=0.64

# measure查询cpu
select * from cpu
name: cpu
tags: host=serverA, region=us_west
time				value
----				-----
2015-07-20T04:27:25.960705835Z	0.64

# 插入2个值的measure temperature
insert temperature,machine=unit42,type=assembly external=25,internal=37

# measure查询temperature
select * from temperature
name: temperature
tags: machine=unit42, type=assembly
time				external	internal
----				--------	--------
2015-07-20T04:29:17.253782361Z	25		37
(max values 255)

# select * from /.*/ LIMIT 1
# select * from cpu_load_short
# select * from cpu_load_short WHERE value > 0.9

# 显示所有measurements
show measurements;
name: measurements
------------------
name
cpu
temperature

# 显示所有时间序列
show series
name: cpu
---------
_key				host	region
cpu,host=serverA,region=us_west	serverA	us_west


name: temperature
-----------------
_key						machine	type
temperature,machine=unit42,type=assembly	unit42	assembly

HTTP

POST to the endpoint /write
# 建立数据库 
curl -G http://localhost:8086/query --data-urlencode "q=CREATE DATABASE mydb"  

# 写数据
curl -i -XPOST 'http://localhost:8086/write?db=mydb' --data-binary 'cpu_load_short,host=server01,region=us-west value=0.64 1434055562000000000'

timestamp可选，如果没有，使用服务器当前时间。
retention policy query parameter (rp)，数据保留时间，默认是系统配置文件中的。

# 写多条数据
curl -i -XPOST 'http://localhost:8086/write?db=mydb' --data-binary 'cpu_load_short,host=server01,region=us-west value=0.64 cpu_load_short,host=server02,region=us-west value=0.55 1422568543702900257 cpu_load_short,direction=in,host=server01,region=us-west value=23422.0 1422568543702900257'

# 时间格式
curl -i -XPOST 'http://localhost:8086/write?db=mydb&precision=s' --data-binary 'temperature,machine=unit42,type=assembly external=25,internal=37 1434059627' 
precision=s 不指定，系统默认使用ns，可用的n, u, ms, s, m, h

GET to the endpoint /query
# 读数据
curl -G 'http://localhost:8086/query' --data-urlencode "db=mydb" --data-urlencode "epoch=s" --data-urlencode "q=SELECT value FROM cpu_load_short WHERE region='us-west'"
epoch= 时间格式，不指定，系统默认使用ns，n, u, ms, s, m, h
q=请求命令

返回格式
{
    "results": [
        {
            "series": [{}],
            "error": "...."
        }
    ],
    "error": "...."
}
 
# 读多条数据
curl -G 'http://localhost:8086/query' --data-urlencode "db=mydb" --data-urlencode "q=SELECT * FROM cpu_load_short WHERE region=us-west;SELECT * FROM temperature"
用分号间隔

# GET请求需要认证
/query?u=bob&p=mypass

Pretty Printing
curl -G 'http://localhost:8086/query?pretty=true' --data-urlencode "db=mydb" --data-urlencode "q=SELECT * FROM cpu_load_short"

数据库是唯一存在的，是包含users、retention policies、continuous queries的容器。
Retention Policy是建库时默认指定的，确定数据的保存时间。
Duration是Retention Policy不存在时，确定数据的保存时间，不能小于１小时，单位minutes, hours, days, or weeks, with INF（infinite无限）。

Aggregations
COUNT, MEAN, SUM, MEDIAN, PERCENTILE, MIN, MAX, SPREAD, STDDEV, DERIVATIVE, NON_NEGATIVE_DERIVATIVE, DISTINCT, FIRST, and LAST.

show measurements
show measurements with host='server1'

show tag keys
show tag keys from cpu
show tag values from cpu with key='region'

show series
show series with serice='redis'

select * from some_series where time > now() - 1h
# aggregates
select percentile(90, value) from cpu where time > now() - 1d group by time(10m), region
# regex
select value from some_series where value =~ /.*ERROR.*/ and time > '2014-03-01' and time < '2014-4-1'

select value from some_series where host =~ /.*asdf.*/ and time > '2014-03-01' and time < '2014-4-1' group by host