You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@airflow.apache.org by "Ash Berlin-Taylor (Jira)" <ji...@apache.org> on 2020/02/13 21:08:00 UTC

[jira] [Resolved] (AIRFLOW-2906) DataDog Integration for Airflow

     [ https://issues.apache.org/jira/browse/AIRFLOW-2906?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ash Berlin-Taylor resolved AIRFLOW-2906.
----------------------------------------
    Fix Version/s: 1.10.10
       Resolution: Fixed

I've marked this as fix for 1.10.10, but to pick it in we'll need to 1.

Check this metaclass implemation works with Py2 (it won't as it stands, but six might help us out), and also pull in the previous refactors of stats from airflow.settings in to its own airflow.stats module.

> DataDog Integration for Airflow
> -------------------------------
>
>                 Key: AIRFLOW-2906
>                 URL: https://issues.apache.org/jira/browse/AIRFLOW-2906
>             Project: Apache Airflow
>          Issue Type: Improvement
>          Components: utils
>    Affects Versions: 1.8.0
>            Reporter: Austin Hsu
>            Assignee: Chandu Kavar
>            Priority: Minor
>              Labels: metrics
>             Fix For: 1.10.10
>
>
> Add functionality to Airflow to enable sending of metrics to DataDog.  DataDog provides support for tags which allows us to aggregate data more easily and visualize it.  We can utilize the [Datadog python library|https://github.com/DataDog/datadogpy] python library and the [Datadog ThreadStats module|https://datadogpy.readthedocs.io/en/latest/#datadog-threadstats-module] to send metrics directly to DataDog without needing to spin up an agent to forward the metrics.  The current implementation in 1.8 uses the statsd library to send the metrics which provides us with much less control to filter our data.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)