InfluxDB API reference

The InfluxDB API provides a simple way to interact with the database.It uses HTTP response codes, HTTP authentication, JWT Tokens, and basic authentication, and responses are returned in JSON.

The following sections assume your InfluxDB instance is running on localhostport 8086 and HTTPS is not enabled.Those settings are configurable.

InfluxDB HTTP endpoints

/debug/pprof HTTP endpoint

InfluxDB supports the Go net/http/pprof HTTP endpoints, which are useful for troubleshooting. The pprof package serves runtime profiling data in the format expected by the pprof visualization tool.

Definition

curl http://localhost:8086/debug/pprof/

The /debug/pprof/ endpoint generates an HTML page with a list of built-in Go profiles and hyperlinks for each.

To access one of the /debug/pprof/ profiles listed above, use the following cURL request, substituting <profile> with the name of the profile. The resulting profile is output to a file specified in <path/to/output-file>.

curl -o <path/to/output-file>  http://localhost:8086/debug/pprof/<profile>

In the following example, the cURL command outputs the resulting heap profile to a file:

curl -o <path/to/output-file> http://localhost:/8086/debug/pprof/heap

You can also use the Go pprof interactive tool to access the InfluxDB /debug/pprof/ profiles.For example, to look at the heap profile of a InfluxDB instance using this tool, you would use a command like this:

go tool pprof http://localhost:8086/debug/pprof/heap

For more information about the Go /net/http/pprof package and the interactive pprof analysis and visualization tool, see:

• Package pprof (net/http/pprof)
• pprof analysis and visualization tool
• Profiling Go programs
• Diagnostics - Profiling

/debug/pprof/all HTTP endpoint

The /debug/pprof/all endpoint is a custom /debug/pprof profile intended primarily for use by InfluxData support. This endpoint generates a profile.tar.gz archive containing text files with the standard Go profiling information and additional debugging data. An optional CPU profile is generated when using the cpu=true option (default is false).

To create a profile.tar.gz archive, use the following cURL command to generate a profile.tar.gz file for sharing with InfluxData support.

curl -o profiles.tar.gz "http://localhost:8086/debug/pprof/all?cpu=true"

Note: When the cpu=true option is included, a CPU profile is generated for 30+ seconds.If you’re concerned about running a CPU profile (which only has a small, temporary impact on performance), then you can set ?cpu=false or omit ?cpu=true altogether.

As the following example shows, the cURL output includes “Time Spent,” the time elapsed (in seconds). After 30 seconds of data has been collected, the results are output to a file.

➜  ~ curl -o profiles.tar.gz "http://localhost:8086/debug/pprof/all?cpu=true"
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  237k    0  237k    0     0   8025      0 --:--:--  0:00:30 --:--:-- 79588

/debug/requests HTTP endpoint

Use this endpoint to track HTTP client requests to the /write and /query endpoints.The /debug/requests endpoint returns the number of writes and queries to InfluxDB per username and IP address.

Definition

curl http://localhost:8086/debug/requests

Query string parameters

Examples

Track requests over a ten-second interval

$ curl http://localhost:8086/debug/requests
{
"user1:123.45.678.91": {"writes":1,"queries":0},
}

The response shows that, over the past ten seconds, the user1 user sent one request to the /write endpoint and no requests to the /query endpoint from the 123.45.678.91 IP address.

Track requests over a one-minute interval

$ curl http://localhost:8086/debug/requests?seconds=60
{
"user1:123.45.678.91": {"writes":3,"queries":0},
"user1:000.0.0.0": {"writes":0,"queries":16},
"user2:xx.xx.xxx.xxx": {"writes":4,"queries":0}
}

The response shows that, over the past minute, user1 sent three requests to the /write endpoint from 123.45.678.91, user1 sent 16 requests to the /query endpoint from 000.0.0.0, and user2 sent four requests to the /write endpoint from xx.xx.xxx.xxx.

/debug/vars HTTP endpoint

InfluxDB exposes statistics and information about its runtime through the /debug/vars endpoint, which can be accessed using the following cURL command:

curl http://localhost:8086/debug/vars

Server statistics and information are displayed in JSON format.

Note: The InfluxDB input plugin is available to collect metrics (using the /debug/vars endpoint) from specified Kapacitor instances. For a list of the measurements and fields, see the InfluxDB input plugin README.

/ping HTTP endpoint

The ping endpoint accepts both GET and HEAD HTTP requests.Use this endpoint to check the status of your InfluxDB instance and your versionof InfluxDB.

Definition

GET http://localhost:8086/ping

HEAD http://localhost:8086/ping

verbose option

By default, the /ping HTTP endpoint returns a simple HTTP 204 status response to let the client know that the server is running. Default value is false.When verbose option is set to true (/ping?verbose=true), an HTTP 200 status is returned.The verbose=true option is required for Google Cloud Load Balancing health checks.

Example

You can use the /ping endpoint to find the build and version of an InfluxDB instance.The X-Influxdb-Build header field displays the InfluxDB build type, either OSS (open source) or ENT (Enterprise).The X-Influxdb-Version header field displays the InfluxDB version.

~ curl -sl -I http://localhost:8086/ping
HTTP/1.1 204 No Content
Content-Type: application/json
Request-Id: 9c353b0e-aadc-11e8-8023-000000000000
X-Influxdb-Build: OSS
X-Influxdb-Version: v1.7.0
X-Request-Id: 9c353b0e-aadc-11e8-8023-000000000000
Date: Tue, 05 Nov 2018 16:08:32 GMT

Status Codes and Responses

The response body is empty.

/query HTTP endpoint

The /query endpoint accepts GET and POST HTTP requests.Use this endpoint to query data and manage databases, retention policies,and users.

Definition

GET http://localhost:8086/query

POST http://localhost:8086/query

Verb usage

• The only exceptions are SELECT queries that include an INTO clause.Those SELECT queries require a POST request.

Examples

Query data with a SELECT statement

$ curl -G 'http://localhost:8086/query?db=mydb' --data-urlencode 'q=SELECT * FROM "mymeas"'
{"results":[{"statement_id":0,"series":[{"name":"mymeas","columns":["time","myfield","mytag1","mytag2"],"values":[["2017-03-01T00:16:18Z",33.1,null,null],["2017-03-01T00:17:18Z",12.4,"12","14"]]}]}]}

The mymeas measurement has two points.The first point has the timestamp 2017-03-01T00:16:18Z, a myfield value of 33.1, and no tag values for the mytag1 and mytag2 tag keys.The second point has the timestamp 2017-03-01T00:17:18Z, a myfield value of 12.4, a mytag1 value of 12, and a mytag2 value of 14.

The same query in the InfluxDB Command Line Interface (CLI) returns the following table:

name: mymeas
time                  myfield  mytag1  mytag2
----                  -------  ------  ------
2017-03-01T00:16:18Z  33.1
2017-03-01T00:17:18Z  12.4     12      14

Query data with a SELECT statement and an INTO clause

$ curl -XPOST 'http://localhost:8086/query?db=mydb' --data-urlencode 'q=SELECT * INTO "newmeas" FROM "mymeas"'
{"results":[{"statement_id":0,"series":[{"name":"result","columns":["time","written"],"values":[["1970-01-01T00:00:00Z",2]]}]}]}

SELECT queries that include and INTO clause require a POST request.

The response shows that InfluxDB writes two points to the newmeas measurement.Note that the system uses epoch 0 (1970-01-01T00:00:00Z) as a null timestamp equivalent.

Create a database

$ curl -XPOST 'http://localhost:8086/query' --data-urlencode 'q=CREATE DATABASE "mydb"'
{"results":[{"statement_id":0}]}

A successful CREATE DATABASE query returns no additional information.

Query string parameters

• InfluxDB does not truncate the number of rows returned for requests without the chunked parameter.That behavior is configurable; see the max-row-limit configuration option for more information.

** The InfluxDB API also supports basic authentication.Use basic authentication if you’ve enabled authenticationand aren’t using the query string parameters u and p.See below for an example of basic authentication.

Examples

Query data with a SELECT statement and return pretty-printed JSON

$ curl -G 'http://localhost:8086/query?db=mydb&pretty=true' --data-urlencode 'q=SELECT * FROM "mymeas"'
{
    "results": [
        {
            "statement_id": 0,
            "series": [
                {
                    "name": "mymeas",
                    "columns": [
                        "time",
                        "myfield",
                        "mytag1",
                        "mytag2"
                    ],
                    "values": [
                        [
                            "2017-03-01T00:16:18Z",
                            33.1,
                            null,
                            null
                        ],
                        [
                            "2017-03-01T00:17:18Z",
                            12.4,
                            "12",
                            "14"
                        ]
                    ]
                }
            ]
        }
    ]
}

Query data with a SELECT statement and return second precision epoch timestamps

$ curl -G 'http://localhost:8086/query?db=mydb&epoch=s' --data-urlencode 'q=SELECT * FROM "mymeas"'
{"results":[{"statement_id":0,"series":[{"name":"mymeas","columns":["time","myfield","mytag1","mytag2"],"values":[[1488327378,33.1,null,null],[1488327438,12.4,"12","14"]]}]}]}

Create a database using HTTP authentication

Valid credentials:

$ curl -XPOST 'http://localhost:8086/query?u=myusername&p=mypassword' --data-urlencode 'q=CREATE DATABASE "mydb"'
{"results":[{"statement_id":0}]}

A successful CREATE DATABASE query returns no additional information.

Invalid credentials:

$ curl -XPOST 'http://localhost:8086/query?u=myusername&p=notmypassword' --data-urlencode 'q=CREATE DATABASE "mydb"'
{"error":"authorization failed"}

Create a database using basic authentication

The following example uses valid credentials.

$ curl -XPOST -u myusername:mypassword 'http://localhost:8086/query' --data-urlencode 'q=CREATE DATABASE "mydb"'
{"results":[{"statement_id":0}]}

A successful CREATE DATABASE query returns no additional information.

The following example uses invalid credentials.

$ curl -XPOST -u myusername:notmypassword 'http://localhost:8086/query' --data-urlencode 'q=CREATE DATABASE "mydb"'
{"error":"authorization failed"}

Request body

--data-urlencode "q=<InfluxQL query>"

All queries must be URL encoded and followInfluxQL syntax.Our example shows the —data-urlencode parameter from curl, which we use in all examples on this page.

Options

Request multiple queries

Delimit multiple queries with a semicolon ;.

Submit queries from a file

The API supports submitting queries from a file using a multipart POSTrequest.The queries in the file must be separated a semicolon (;).

Syntax:

curl -F "q=@<path_to_file>" -F "async=true" http://localhost:8086/query

Request query results in CSV format

Syntax:

curl -H "Accept: application/csv" -G 'http://localhost:8086/query [...]

Note that when the request includes -H "Accept: application/csv", the system returns timestamps in epoch format, not RFC3339 format.

Bind parameters

The API supports binding parameters to particular field values or tag values inthe WHERE clause.Use the syntax $<placeholder_key> as a placeholder in the query, and URLencode the map of placeholder keys to placeholder values in the request body:

Query syntax:

--data-urlencode 'q= SELECT [...] WHERE [ <field_key> | <tag_key> ] = $<placeholder_key>'

Map syntax:

--data-urlencode 'params={"<placeholder_key>":[ <placeholder_float_field_value> | <placeholder_integer_field_value> | "<placeholder_string_field_value>" | <placeholder_boolean_field_value> | "<placeholder_tag_value>" ]}'

Delimit multiple placeholder key-value pairs with comma ,.

Examples

Send multiple queries

$ curl -G 'http://localhost:8086/query?db=mydb&epoch=s' --data-urlencode 'q=SELECT * FROM "mymeas";SELECT mean("myfield") FROM "mymeas"'
{"results":[{"statement_id":0,"series":[{"name":"mymeas","columns":["time","myfield","mytag1","mytag2"],"values":[[1488327378,33.1,null,null],[1488327438,12.4,"12","14"]]}]},{"statement_id":1,"series":[{"name":"mymeas","columns":["time","mean"],"values":[[0,22.75]]}]}]}

The request includes two queries: SELECT * FROM "mymeas" and SELECT mean("myfield") FROM "mymeas"'.In the results, the system assigns a statement identifier to each query return.The first query’s result has a statement_id of 0 and the second query’s result has a statement_id of 1.

Request query results in CSV format

$ curl -H "Accept: application/csv" -G 'http://localhost:8086/query?db=mydb' --data-urlencode 'q=SELECT * FROM "mymeas"'
name,tags,time,myfield,mytag1,mytag2
mymeas,,1488327378000000000,33.1,mytag1,mytag2
mymeas,,1488327438000000000,12.4,12,14

The first point has no tag values for the mytag1 and mytag2 tag keys.

Submit queries from a file

curl -F "q=@queries.txt" -F "async=true" 'http://localhost:8086/query'

A sample of the queries in queries.txt:

CREATE DATABASE mydb;
CREATE RETENTION POLICY four_weeks ON mydb DURATION 4w REPLICATION 1;

Bind a parameter in the WHERE clause to specific tag value

$ curl -G 'http://localhost:8086/query?db=mydb' --data-urlencode 'q=SELECT * FROM "mymeas" WHERE "mytag1" = $tag_value' --data-urlencode 'params={"tag_value":"12"}'
{"results":[{"statement_id":0,"series":[{"name":"mymeas","columns":["time","myfield","mytag1","mytag2"],"values":[["2017-03-01T00:17:18Z",12.4,"12","14"]]}]}]}

The request maps $tag_value to 12.InfluxDB stores tag values as strings they and must be double quoted in the request.

Bind a parameter in the WHERE clause to a numerical field value

$ curl -G 'http://localhost:8086/query?db=mydb' --data-urlencode 'q=SELECT * FROM "mymeas" WHERE "myfield" > $field_value' --data-urlencode 'params={"field_value":30}'
{"results":[{"statement_id":0,"series":[{"name":"mymeas","columns":["time","myfield","mytag1","mytag2"],"values":[["2017-03-01T00:16:18Z",33.1,null,null]]}]}]}

The request maps $field_value to 30.The value 30 does not require double quotes because myfield stores numerical field values.

Bind two parameters in the WHERE clause to a specific tag value and numerical field value

$ curl -G 'http://localhost:8086/query?db=mydb' --data-urlencode 'q=SELECT * FROM "mymeas" WHERE "mytag1" = $tag_value AND  "myfield" < $field_value' --data-urlencode 'params={"tag_value":"12","field_value":30}'
{"results":[{"statement_id":0,"series":[{"name":"mymeas","columns":["time","myfield","mytag1","mytag2"],"values":[["2017-03-01T00:17:18Z",12.4,"12","14"]]}]}]}

The request maps $tag_value to 12 and $field_value to 30.

Status codes and responses

Responses are returned in JSON.Include the query string parameter pretty=trueto enable pretty-print JSON.

Summary table

Examples

A successful request that returns data

$ curl -i -G 'http://localhost:8086/query?db=mydb' --data-urlencode 'q=SELECT * FROM "mymeas"'
HTTP/1.1 200 OK
Connection: close
Content-Type: application/json
Request-Id: [...]
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 19:22:54 GMT
Transfer-Encoding: chunked
{"results":[{"statement_id":0,"series":[{"name":"mymeas","columns":["time","myfield","mytag1","mytag2"],"values":[["2017-03-01T00:16:18Z",33.1,null,null],["2017-03-01T00:17:18Z",12.4,"12","14"]]}]}]}

A successful request that returns an error

$ curl -i -G 'http://localhost:8086/query?db=mydb1' --data-urlencode 'q=SELECT * FROM "mymeas"'
HTTP/1.1 200 OK
Connection: close
Content-Type: application/json
Request-Id: [...]
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 19:23:48 GMT
Transfer-Encoding: chunked
{"results":[{"statement_id":0,"error":"database not found: mydb1"}]}

An incorrectly formatted query

$ curl -i -G 'http://localhost:8086/query?db=mydb' --data-urlencode 'q=SELECT *'
HTTP/1.1 400 Bad Request
Content-Type: application/json
Request-Id: [...]
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 19:24:25 GMT
Content-Length: 76
{"error":"error parsing query: found EOF, expected FROM at line 1, char 9"}

Query data with invalid authentication credentials

$ curl -i  -XPOST 'http://localhost:8086/query?u=myusername&p=notmypassword' --data-urlencode 'q=CREATE DATABASE "mydb"'
HTTP/1.1 401 Unauthorized
Content-Type: application/json
Request-Id: [...]
Www-Authenticate: Basic realm="InfluxDB"
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 19:11:26 GMT
Content-Length: 33
{"error":"authorization failed"}

/write HTTP endpoint

The /write endpoint accepts POST HTTP requests.Use this endpoint to write data to a pre-existing database.

Definition

POST http://localhost:8086/write

Query string parameters

• The InfluxDB API also supports basic authentication.Use basic authentication if you’ve enabled authenticationand aren’t using the query string parameters u and p.See below for an example of basic authentication.

** We recommend using the least precise precision possible as this can resultin significant improvements in compression.

Examples

Write a point to the database mydb with a timestamp in seconds

$ curl -i -XPOST "http://localhost:8086/write?db=mydb&precision=s" --data-binary 'mymeas,mytag=1 myfield=90 1463683075'
HTTP/1.1 204 No Content
Content-Type: application/json
Request-Id: [...]
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 17:33:23 GMT

Write a point to the database mydb and the retention policy myrp

$ curl -i -XPOST "http://localhost:8086/write?db=mydb&rp=myrp" --data-binary 'mymeas,mytag=1 myfield=90'
HTTP/1.1 204 No Content
Content-Type: application/json
Request-Id: [...]
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 17:34:31 GMT

Write a point to the database mydb using HTTP authentication

Valid credentials:

$ curl -i -XPOST "http://localhost:8086/write?db=mydb&u=myusername&p=mypassword" --data-binary 'mymeas,mytag=1 myfield=91'
HTTP/1.1 204 No Content
Content-Type: application/json
Request-Id: [...]
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 17:34:56 GMT

Invalid credentials:

$ curl -i -XPOST "http://localhost:8086/write?db=mydb&u=myusername&p=notmypassword" --data-binary 'mymeas,mytag=1 myfield=91'
HTTP/1.1 401 Unauthorized
Content-Type: application/json
Request-Id: [...]
Www-Authenticate: Basic realm="InfluxDB"
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 17:40:30 GMT
Content-Length: 33
{"error":"authorization failed"}

Write a point to the database mydb using basic authentication

Valid credentials:

$ curl -i -XPOST -u myusername:mypassword "http://localhost:8086/write?db=mydb" --data-binary 'mymeas,mytag=1 myfield=91'
HTTP/1.1 204 No Content
Content-Type: application/json
Request-Id: [...]
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 17:36:40 GMT

Invalid credentials:

$ curl -i -XPOST -u myusername:notmypassword "http://localhost:8086/write?db=mydb" --data-binary 'mymeas,mytag=1 myfield=91'
HTTP/1.1 401 Unauthorized
Content-Type: application/json
Request-Id: [...]
Www-Authenticate: Basic realm="InfluxDB"
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 17:46:40 GMT
Content-Length: 33
{"error":"authorization failed"}

Request body

--data-binary '<Data in InfluxDB line protocol format>'

All data must be binary encoded and in theInfluxDB line protocol format.Our example shows the —data-binary parameter from curl, which we will use inall examples on this page.Using any encoding method other than —data-binary will likely lead to issues;-d, —data-urlencode, and —data-ascii may strip out newlines orintroduce new, unintended formatting.

Options:

• Write several points to the database with one request by separating each pointby a new line.
• Write points from a file with the @ flag.The file should contain a batch of points in the InfluxDB line protocol format.Individual points must be on their own line and separated by newline characters(\n).Files containing carriage returns will cause parser errors.

We recommend writing points in batches of 5,000 to 10,000 points.Smaller batches, and more HTTP requests, will result in sub-optimal performance.

Examples

Write a point to the database mydb with a nanosecond timestamp

$ curl -i -XPOST "http://localhost:8086/write?db=mydb" --data-binary 'mymeas,mytag=1 myfield=90 1463683075000000000'
HTTP/1.1 204 No Content
Content-Type: application/json
Request-Id: [...]
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 18:02:57 GMT

Write a point to the database mydb with the local server’s nanosecond timestamp

$ curl -i -XPOST "http://localhost:8086/write?db=mydb" --data-binary 'mymeas,mytag=1 myfield=90'
HTTP/1.1 204 No Content
Content-Type: application/json
Request-Id: [...]
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 18:03:44 GMT

Write several points to the database mydb by separating points with a new line

$ curl -i -XPOST "http://localhost:8086/write?db=mydb" --data-binary 'mymeas,mytag=3 myfield=89 1463689152000000000
mymeas,mytag=2 myfield=34 1463689152000000000'
HTTP/1.1 204 No Content
Content-Type: application/json
Request-Id: [...]
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 18:04:02 GMT

Write several points to the database mydb from the file data.txt

$ curl -i -XPOST "http://localhost:8086/write?db=mydb" --data-binary @data.txt
HTTP/1.1 204 No Content
Content-Type: application/json
Request-Id: [...]
X-Influxdb-Version: 1.4.x
Date: Wed, 08 Nov 2017 18:08:11 GMT

A sample of the data in data.txt:

mymeas,mytag1=1 value=21 1463689680000000000
mymeas,mytag1=1 value=34 1463689690000000000
mymeas,mytag2=8 value=78 1463689700000000000
mymeas,mytag3=9 value=89 1463689710000000000

Status codes and responses

In general, status codes of the form 2xx indicate success, 4xx indicatethat InfluxDB could not understand the request, and 5xx indicate that thesystem is overloaded or significantly impaired.Errors are returned in JSON.

Summary table

Examples

A successful write

HTTP/1.1 204 No Content

Write a point with an incorrect timestamp

HTTP/1.1 400 Bad Request
[...]
{"error":"unable to parse 'mymeas,mytag=1 myfield=91 abc123': bad timestamp"}

Write an integer to a field that previously accepted a float

HTTP/1.1 400 Bad Request
[...]
{"error":"field type conflict: input field \"myfield\" on measurement \"mymeas\" is type int64, already exists as type float"}

Write a point with invalid authentication credentials

HTTP/1.1 401 Unauthorized
[...]
{"error":"authorization failed"}

Write a point to a database that doesn’t exist

HTTP/1.1 404 Not Found
[...]
{"error":"database not found: \"mydb1\""}

Send a request body that is too large

HTTP/2 413 Request Entity Too Large
[...]
{"error":"Request Entity Too Large"}

Write a point to a retention policy that doesn’t exist

HTTP/1.1 500 Internal Server Error
[...]
{"error":"retention policy not found: myrp"}