WebPlatform Docs WebPlatformDocs

Make OpenStack meta-data part of your grains

Add the following file into your salt-states in the _grains/ folder to get more data related to the VMs you run on OpenStack/DreamCompute.

Full blog post describing how it works has been published at https://renoirboulanger.com/blog/2015/05/add-openstack-instance-meta-data-info-salt-grains/

Get an instance UUID

Locally

Progress executive summary

Migrated to Dreamhost
Set in place a complete, separate, deployment of the site (webplatformstaging.org) and reworked deployment scripts to facilitate infrastructure work without impacting live site.
Brought on Dreamhost as steward
Dramatically increased stability/uptime. We now have about 99.8% uptime compared to frequent and lasting outages.
All servers are using up to date version
Automatic security updates
Added features:

Compatibility tables imported from MDN, data editable through github, updates table in wiki
Notes server, also loaded from a few w3.org/TR/ documents

Bootstrapping a new Salt Master for WebPlatform

Moved to webplatform/ops in salt-master/

Installing WebPlatform Docs wiki locally

This gist gives instructions to recreate the exact configuration of WebPlatform Docs wiki.

Steps

Clone MediaWiki Vagrant repo
Follow the instructions
Change the MediaWiki version branch in mediawiki/ folder to 1.25wmf15 (the version we upgraded on 2015-02-05

Monit configuration on WebPlatform infrastructure

The monit project has two components; Monit, and M/Monit. While monit is open-source and can be used stand alone along with an independent web ui, M/Monit can manage from one place every machine that runs monit.

M/Monit can control monit configuration, creates a global cluster health graph all in "one place".

If the management don’t want to pay for a M/Monit licence its possible to shim a dashboard with a set of well-thought NGINX proxy configuration. While it would allow to control every nodes, there would be no overall network graphs.

Screenshots

The following steps are things that aren’t been handled by Ansible and that were required to do manually so we could run FxA on a Vagrant VM using mozilla/fxa-dev.

Note that this procedure has been made after the steps described in mozilla/fxa-dev#81

Procedure

Database HTTP layer

Inside the VM, create manually the database

	[{"name":"wp-logger","time":"2012-10-07 23:48:36","message":null,"type":"join"},
	{"name":"shepazu","time":"2012-10-07 23:50:56","message":"hi, wp-logger","type":"pubmsg"},
	{"name":"shepazu","time":"2012-10-07 23:54:32","message":"paul_irish_: how do I make myself channel op? I thought we set that up a while ago...","type":"pubmsg"},
	{"name":"Ryan_Lane","time":"2012-10-08 00:00:05","message":"http:\/\/talk.webplatform.org\/chatlogs\/#home","type":"pubmsg"},
	{"name":"Ryan_Lane","time":"2012-10-08 00:00:27","message":"I wonder how often it updates","type":"pubmsg"},
	{"name":"shepazu","time":"2012-10-08 00:00:39","message":"hmm","type":"pubmsg"},
	{"name":"shepazu","time":"2012-10-08 00:00:47","message":"not very often, it seems","type":"pubmsg"},
	{"name":"shepazu","time":"2012-10-08 00:01:24","message":"I'll look for another logging solution longer term","type":"pubmsg"},
	{"name":"Ryan_Lane","time":"2012-10-08 00:01:59","message":"it's often-ish","type":"pubmsg"},
	{"name":"Ryan_Lane","time":"2012-10-08 00:02:09",

	<?php

	/**
	* Export LumberJack IRC log to text
	**/

	$username='lumberjack_user';
	$password='secretpassword';
	$database='lumberjack';
	$host='10.10.10.3';

	[client]
	ssl
	ssl-cert=/etc/mysql/client-cert.pem
	ssl-key=/etc/mysql/client-key.pem

	[mysqld]
	ssl
	ssl-cipher=DHE-RSA-AES256-SHA
	ssl-ca=/etc/mysql/ca-cert.pem
	ssl-cert=/etc/mysql/server-cert.pem

	ControlMaster auto
	ControlPath ~/.ssh/sockets/%r__%h__%p.sock
	TCPKeepAlive no
	ServerAliveInterval 60
	ServerAliveCountMax 10

	Host staging.wpdn
	Hostname salt.webplatformstaging.org
	ProxyCommand none