跳到主要内容
版本:2.3.8

InfluxDB

InfluxDB source connector

Description

Read external data source data through InfluxDB.

Key features

supports query SQL and can achieve projection effect.

Options

nametyperequireddefault value
urlstringyes-
sqlstringyes-
schemaconfigyes-
databasestringyes
usernamestringno-
passwordstringno-
lower_boundlongno-
upper_boundlongno-
partition_numintno-
split_columnstringno-
epochstringnon
connect_timeout_mslongno15000
query_timeout_secintno3
common-optionsconfigno-

url

the url to connect to influxDB e.g.

http://influxdb-host:8086

sql [string]

The query sql used to search data

select name,age from test

schema [config]

fields [Config]

The schema information of upstream data. e.g.

schema {
fields {
name = string
age = int
}
}

database [string]

The influxDB database

username [string]

the username of the influxDB when you select

password [string]

the password of the influxDB when you select

split_column [string]

the split_column of the influxDB when you select

Tips:

  • influxDB tags is not supported as a segmented primary key because the type of tags can only be a string
  • influxDB time is not supported as a segmented primary key because the time field cannot participate in mathematical calculation
  • Currently, split_column only supports integer data segmentation, and does not support float, string, date and other types.

upper_bound [long]

upper bound of the split_columncolumn

lower_bound [long]

lower bound of the split_column column

     split the $split_column range into $partition_num parts
if partition_num is 1, use the whole `split_column` range
if partition_num < (upper_bound - lower_bound), use (upper_bound - lower_bound) partitions

eg: lower_bound = 1, upper_bound = 10, partition_num = 2
sql = "select * from test where age > 0 and age < 10"

split result

split 1: select * from test where ($split_column >= 1 and $split_column < 6) and ( age > 0 and age < 10 )

split 2: select * from test where ($split_column >= 6 and $split_column < 11) and ( age > 0 and age < 10 )

partition_num [int]

the partition_num of the InfluxDB when you select

Tips: Ensure that upper_bound minus lower_bound is divided bypartition_num, otherwise the query results will overlap

epoch [string]

returned time precision

  • Optional values: H, m, s, MS, u, n
  • default value: n

query_timeout_sec [int]

the query_timeout of the InfluxDB when you select, in seconds

connect_timeout_ms [long]

the timeout for connecting to InfluxDB, in milliseconds

common options

Source plugin common parameters, please refer to Source Common Options for details

Examples

Example of multi parallelism and multi partition scanning

source {

InfluxDB {
url = "http://influxdb-host:8086"
sql = "select label, value, rt, time from test"
database = "test"
upper_bound = 100
lower_bound = 1
partition_num = 4
split_column = "value"
schema {
fields {
label = STRING
value = INT
rt = STRING
time = BIGINT
}
}

}

Example of not using partition scan

source {

InfluxDB {
url = "http://influxdb-host:8086"
sql = "select label, value, rt, time from test"
database = "test"
schema {
fields {
label = STRING
value = INT
rt = STRING
time = BIGINT
}
}

}

Changelog

2.2.0-beta 2022-09-26

  • Add InfluxDB Source Connector