jpmens

# graphite-web install is hardcoded in setup.cfg to /opt/graphite
sudo mkdir /opt/graphite
sudo chown brad.users /opt/graphite

# run under python2.7 virtualenv
virtualenv --python=python2.7 ~/ve/graphite
source ~/ve/graphite/bin/activate

# install the necessary python packages (simplejson is for flot graphs)
pip install graphite-web carbon whisper django django-tagging uwsgi simplejson

jpmens

logstash queries graphed with graphite.

Operation: Decouple whisper from graphite.

Method: Create a graphite function that does a date histogram facet query against elasticsearch for a given query string for the time period viewed in the current graph.

Reason: graphite has some awesome math functions. Wouldn't it be cool if we could use those on logstash results?

The screenshot below is using logstash to watch the twitter stream of keywords "iphone" "apple" and "samsung" - then I graph them each, so we get an idea of popularity. As a bonus, I also do a movingAverage() on the iphone curve to show you why this is awesome.

jpmens

diff --git a/src/lib/krb5/os/changepw.c b/src/lib/krb5/os/changepw.c
index 4ad8f32..bcea591 100644
--- a/src/lib/krb5/os/changepw.c
+++ b/src/lib/krb5/os/changepw.c
@@ -223,7 +223,7 @@ change_set_password(krb5_context context,
 {
     krb5_data                   chpw_rep;
     krb5_address                remote_kaddr;
-    krb5_boolean                use_tcp = 0;
+    krb5_boolean                use_tcp = 1;

jpmens

#!/bin/bash

# Schedule this to run once a day with cron. Doesn't matter what time since it parses yesterday's hits (by default).
# I only tested this on the Marco.org server, which runs CentOS (RHEL). No idea how it'll work on other distributions, but it's pretty basic.

# Required variables:
RSS_URI="/rss"
MAIL_TO="your@email.com"
LOG_FILE="/home/[UBERSPACE_USERNAME]/logs/access_log"

jpmens

#!/bin/zsh

# Usage: copy-rtf [file]
#
# Copies Pygmentize's RTF output of a source file to the clipboard after making it 36pt Monaco. 
#
# Copy straight into an "Auto-Shrink" text box in Keynote for efficient code highlighting in a Coding presentation.

# Depends on [Pygments](http://pygments.org/) being installed in your $PATH... 
# try `gem install pygmentize` for a standalone Ruby installation.

jpmens

" Why query me?Your IP had been logged!"
" unknown "
"( prasina aloga )"
"( surely you must be joking :) )"
"(get out - secure)"
"-"
"-=[NiN]=-"
"."
"... baubau :)"
"..:: blink-bind ::.."

jpmens

#!/usr/bin/env python

#
# This is a script we use at Tenscores, and it is specific only in the sense that we use IP addresses
# and use the variable `hostname` to find the correct nodes.  Also, it connect to user `ubuntu` by
# default.. that could be checked up in the vars too.
#
# Our hosts file looks like:
#
#[appservers]

jpmens

.ASIA @registryasia
.BIZ @Neustar
.CAT @dotcatdomain @puntcat
.COM .NET .NAME @VERISIGN 
.COOP @dotCoop
.INFO @Afilias
.MOBI @dotMobiOfficial
.ORG @ORGBuzz
.PRO @RegistryPro
.TEL @dottel @TelnicLimited

jpmens

.AR @nicargentina
.AU @auda
.BE @DNSbe
.CA @CIRANEWS
.CL @NICChile
.CO @dotco
.CR @CR_NIC
.CZ @CZ_NIC
.DE @denic_de
.EE @Eesti_Internet

jpmens

My ansible approach

Per service playbooks

Playbooks are written per service : a playbook is a collection of tasks, and eventually associated handers, templates, and files, that are required to properly install and setup a service.

Each playbook has a setup.yml entry point, which is responsible of including various necessary tasks to get the service up and running. Typically, this