In DC/OS Admin Router wraps an instance of Nginx.
As a first step for collecting metrics for Admin Router we want to collect advanced metrics from Nginx. Nginx can be instrumented by adding the 3rd party module Virtual Host Traffic Status Module (VTS). https://github.com/vozlt/nginx-module-vts
Telegraf provides a native Plugin for scraping data from the VTS module endpoint then exposed by Admin Router. Therefore, dcos-telegraf must be configured to scrape data from Admin Router using that plugin.
This ticket is only complete with a corresponding DC/OS integration test that makes sure that metrics are collected by dcos-telegraf. One can use the dcos-telegraf output endpoint at localhost:61091/metrics where collected metrics are populated at least once every 20 seconds.
The collected metrics must be tagged accordingly so that they are identifiable as stemming from the Admin Router instances running on DC/OS master nodes.
This ticket is part of the metrics collection effort of the Security Team tracked in:
Please refer to the metrics design doc for further information: