Ben Burns burnsie7

Step 1: Enable log collection:

Open /etc/datadog-agent/datadog.yaml in your editor of choice
Set log_enabled: true
Save and close /etc/datadog-agent/datadog.yaml

Step 2: Enable the apache integration and add the log file settings to apache.d/conf.yaml

The datadog agent runs under the dd-agent user and dd-agent group. This prevents dd-agent accessing the logs in /var/log as they are only accessible by root (or a sudo admin).

In order to allow read only access for dd-agent only, create ACL's and modify logrotate to persist the permissions changes.

You can manually set the ACL with:

sudo setfacl -m g:dd-agent:rx /var/log/<application-name>

This will not persist as logrotate will not re-apply the ACL setting so for a more permanent solution you can add a rule to logrotate to reset the ACL. You will need to create a new file:

Log Monitoring Notes

Important - A valid log config needs to be found in an integration yaml file before logs-agent will start.

Support training exercise:

https://docs.google.com/document/d/1FZWWDak01pXXB9JACUTbIzgLAaoAT4gSZ5lLBjgkJ_s/edit#heading=h.h489z9dybrci

ssh -i ~/.ssh/support-sandbox-demo.pem -l ubuntu hostname

install agent v6

sudo DD_API_KEY=REDACTED bash -c "$(curl -L https://raw.githubusercontent.com/DataDog/datadog-agent/master/cmd/agent/install_script.sh)"

copy and edit trace-agent.conf

sudo cp /etc/datadog-agent/trace-agent.conf.example /etc/datadog-agent/trace-agent.conf

Add to config:

	ac_exclude: []
	ac_include: []
	additional_checksd: c:\programdata\datadog\checks.d
	api_key: paste_api_key_here
	bind_host: localhost
	cmd_host: localhost
	cmd_port: 5001
	conf_path: .
	confd_path: c:\programdata\datadog\conf.d
	dd_url: https://app.datadoghq.com

	import random
	import time
	import requests
	import json

	'''
	Set the following options in /etc/datadog-agent/datadog.yaml:

	# Trace Agent Specific Settings
	apm_config:

	from checks import AgentCheck
	import os

	'''
	This is for demostration purposes only and not recommended for production use.
	Because of the nature of ping, it could result in a very long running check if there are excessive timeouts.
	Long running checks could potentially result in other metrics and checks being skipped.

	Example conf.yaml:

	import json
	import os
	import requests
	import pprint

	from datadog import initialize, api


	DD_API_KEY = os.getenv('DD_API_KEY', '<YOUR_API_KEY>')
	DD_APP_KEY = os.getenv('DD_APP_KEY', '<YOUR_APP_KEY>')

	mkdir /etc/datadog-agent/conf.d/logtest.d
	touch /etc/datadog-agent/conf.d/logtest.d/conf.yaml

	##Log section
	logs:

	- type: file
	path: /home/vagrant/logout.log
	service: cawa-server
	source: cawa

	import datetime
	import time
	import requests
	import simplejson

	from datadog import initialize, api

	"""
	This script gives a real time report on ec2 and Datadog agent host usage
	from multiple organizations and reports them up to the 'main' parent account.