Skip to main content

New Relic query interface

Project description

Code style: black pre-commit pre-commit.ci status

Python package codecov

Query interface for the telemetry from New Relic SaaS platform.

Requirements

Description

This module, designed to be used ether from Jypyter or from application providing you a query access to your data stored in New Relic platform, so you can perform custom analysis.

Installation

You can download this module from GitHub repository. Then run

make init dependencies

This will install all dependencies. Then running the pip install as

pip install .

from the module directory source and nrquery module will be installed

How to query New Relic platform ?

nrquery module uses New Relic advanced GraphQL capabilities, but currently only supports NRQL queries. GraphQL queries will be implemented at some point in the future. Module consists of two classes:

Query class

nrquery.Query class provides an interface for sending NRQL queries to New Relic GraphQL API endpoint. Constructor of that class takes two values:

New Relic account (or None) New Relic API User key (or None)

if you pass None to any of the parameters, nrquery module will try to take account information from NRACCOUNT environment variable and API key from NRAPIKEY environment variable.

If nether is specified, exception will be raised.

nrquery.Query.Run method takes NRQL query as parameter and returns nrquery.Result object.

import nrquery

res = nrquery.Query().Run("SELECT COUNT(*) FROM TransactionError")

How to send a multiple queries

You can prepare your nrquery.Query instance to send a multiple queries, then combine result in single DataFrame

import nrquery

q = nrquery.Query()
q += "SELECT * FROM TransactionError LIMIT 1"
q += "SELECT * FROM TransactionError LIMIT 1"
res = q.Run()
df = res.Dataframe()

How to detect a nodes that stop sending data (NODATA() nodes).

There is a special method for nrquery.Query.Deadnodes. This method takes a date query as a partameter. Date query could be in a form of "May 9, 2022" or "1 day ago" or "05-09-2022" and that dat is cut-off date for detection of dead nodes.

import nrquery

q = nrquery.Query()
res = q.Deadnodes("1 hour ago")

This request will return result with list of the nodes stop sending data 1 hour ago.

Get the list of alertable nodes.

nrquery.Query.Alertable() is a special method that will return the list of the nodes that is in Alertable state

import nrquery

q = nrquery.Query()
res = q.Alertable()

Result class

You shall not directly create instances of the nrquery.Result class. Method Run of the class nrquery.Query will return an instance of the Result class. There are few class variables that can pose some interest:

  • nrquery.Result.IsSuccess - True or False insicating success or failure of the query associated with result.
  • nrquery.Result.Elapsed - time tat takes New Relic platform to run that query
  • nrquery.Result.Query - query associated with that result

Following methods can be used to extract the value of the query. You must undestand the expected query outcome and use appropriate conversion methods:

  • nrquery.Result.Json - will convert result into a plain JSON
  • nrquery.Result.Series - will convert result into a single Pandas series object
  • nrquery.Result.Dataframe - will convert result into a Pandas Dataframe object
  • nrquery.Result.CSV - will convert result into a CSV string.
  • nrqauery.Result.Numpy - will convert result into a dictionary of numpy arrays
import nrquery

res = nrquery.Query().Run("SELECT * FROM TransactionError")
df = res.Dataframe()

nrquery.Result.Deadnodes

This merhod returns the list of nodes that been prepared by nrquery.Query.Deadnodes

nrquery.Result.Alertable

This method will return the list of the nodes prepared by nrquery.Query.Alertable

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

nrquery-1.6.tar.gz (7.1 kB view hashes)

Uploaded Source

Built Distribution

nrquery-1.6-py3-none-any.whl (6.9 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page