The New Relic Query Language (NRQL) is an SQL-flavored query language for making calls against an event database. This document explains how to use NRQL to pull data from the events database and organize it to view on your dashboard.
New Relic Query Language (NRQL) enables you to query data collected from your application and transform that data into dynamic charts. From there, you can interpret your data to better understand how your application is used in a variety of ways. If you are new to New Relic Insights or don't know much about query languages, this tutorial will give you a quick overview of NRQL query syntax and composition, along with some real examples.
Use the query tool
The command line query tool appears at the top of your dashboards and the Query history page. Use it to build out commands with SQL-style syntax. As you type, the auto-completion feature suggests attributes, and it flags syntax errors with a red underline. For more information about NRQL syntax, see NRQL reference.
The parser can identify syntax errors, but logic errors will normally not be apparent until you try to execute the statement. If the statement cannot be run, an error report will indicate the location of the first identified error. New Relic Insights also ignores
NULL results entirely, unless your query explicitly requests
SELECT function(attribute) [AS 'label'][, ...] FROM event [WHERE attribute [comparison] [AND|OR ...]][AS 'label'][, ...] [FACET attribute | function(attribute)] [LIMIT number] [SINCE time] [UNTIL time] [WITH TIMEZONE timezone] [COMPARE WITH time] [TIMESERIES time]
New Relic refers to NRQL statements as event queries. The syntax of the NRQL event query is similar to standard SQL queries. Basic syntax rules include:
NRQL uses single quotes
''to designate strings.
You must use backticks
``to quote a custom attribute name with a space in it (for example:
Insights does not support "duck typing." This means that a float stored as a string is treated as a string and cannot be operated on by functions expecting float values.
Insights does not have the equivalent of the SQL JOIN function, but you can simulate a JOIN with custom attributes.
- Basic NRQL query
SELECT uniqueCount(user) FROM PageView WHERE userAgentOS = 'Mac' FACET countryCode SINCE 1 day ago LIMIT 20
- NRQL query on attribute with space in name
SELECT count(*) FROM Transaction FACET `Logged-in user`
- Query multiple events or columns
Use a comma (
,) to merge data from two event types or to return multiple columns from a dataset. To return multiple event types, join the types with a comma:
SELECT ... FROM event, event ...
SELECT count(*) FROM Transaction, PageView SINCE 3 days ago
To return multiple columns from a dataset, separate the aggregator arguments with a comma:
SELECT function(attribute), function(attribute) ... FROM ...
This query returns the minimum, average, and maximum duration for Browser events over the last week:
SELECT min(duration), max(duration), average(duration) FROM PageView SINCE 1 week ago
Send custom data to Insights
Adding a high number of custom attributes and events may cause degraded query performance and run up against insertion rate limits for events. For optimal performance, first think about what data you want to analyze, and then create only the custom attributes required for the events.
|Source||How to send custom data|
|APM agent||Use agent API to send custom events and custom attributes|
|Insights API||Insert custom events and custom attributes with the Insights API.|
|Mobile agent||Use mobile agent API to send custom events and custom attributes|
You can use NRQL to query your data several ways. NRQL can query the application data collected by New Relic APM, New Relic Browser, and New Relic Mobile. You can then use this data to answer a variety of questions. For example:
- Unique users: How many unique user sessions did you have in the last week?
- Unique user trends: Were your unique user sessions up or down last week compared to the week before?
SELECT uniqueCount(session) FROM PageView SINCE 1 week ago COMPARE WITH 1 week ago
- Pageview trend graphs: How can I graph the number of unique users yesterday compared with the day before?
SELECT count(*) FROM PageView SINCE 1 day ago COMPARE WITH 1 day ago TIMESERIES AUTO
- OS version: How many of your mobile users are on the latest OS version?
SELECT uniqueCount(uuid) FROM MobileSession FACET osVersion SINCE 7 days ago
- Key account Apdex: Monitor Apdex score for particularly important customer
If you have defined some custom attributes you can query Insights like this to keep an eye on how this customer experiences your app from a performance standpoint: