chetan

Day job: Sr. Developer @ Evidon

Favorite Python project: boto

Favorite Conference: DEFCON

Python Experience Level: intermediate

chetan

#!/usr/bin/env perl

=cut
combine several hadoop (log4j) log files into a single stream, sorted by time
=cut

use File::Basename qw(basename);

my @logs;

chetan

#!/usr/bin/env perl

=cut

usage:

cat foo.txt | hadoop_snip_logs.pl <time1> <time2>

e.g.
cat foo.txt | hadoop_snip_logs.pl 12:26 12:28

chetan

#!/bin/bash

if [[ -z "$1" ]]; then
  echo "usage: $0 <host1> <host2> [host3 ...]"
  exit 1
fi

SRC_HOST=$1

# see if src is setup yet with a key, if not create

chetan

--- netverify.py.orig	2011-11-30 17:26:33.000000000 -0500
+++ netverify.py	2011-11-30 17:32:45.000000000 -0500
@@ -1278,7 +1278,7 @@
         
         # Get the hostname from the two different sources
         
-        hvar = None
+        hvar = ""
         for l in self.hostname_var:
             if l.startswith("HOSTNAME ") or l.startswith("HOSTNAME\t"):

chetan

## don't screw around with cpan! just just cpan-minus!!
##
## run as root!

cd /usr/local/bin; curl -fsSLO http://xrl.us/cpanm; chmod 755 cpanm

## then install away!
## -q = queit, -n no test (who needs those pesky things?!)

cpanm -qn Moose

chetan

#!/bin/bash

# a simple dependency installer using cpanm and the DEPS file


if [ ! -f `which cpanm` ]; then
  echo "need to install cpanm!"
  cd /usr/local/bin
  sudo curl -fsSLO http://xrl.us/cpanm
  sudo chmod 755 cpanm

chetan

#!/usr/bin/env ruby

LONG_UA="-H \"Mozilla/5.0 (Windows NT 6.1; WOW64; rv:8.0) Gecko/20100101 Firefox/8.0fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff0fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff\""

DIR=ARGV.shift

URL="http://sewer3.evidon.com:8080/"

`mkdir -p #{DIR}`
(1..2).each do |i|

chetan

#!/bin/bash

if [ ! -f /etc/apt/sources.list.d/cloudera.list ]; then
  echo deb http://archive.canonical.com/ lucid partner > /etc/apt/sources.list.d/cloudera.list
  echo deb http://archive.cloudera.com/debian lucid-cdh3u0 contrib >> /etc/apt/sources.list.d/cloudera.list
  apt-get update
  apt-get install --yes --force-yes hadoop
fi

if [ `uname -m` == "i686" ]; then

chetan

require 'logger'
require 'stringio'

# simple little wrapper on logger that keeps a buffer of all logged messages
class LogWrapper

  def initialize(file, level, rotation)
    @file_logger = Logger.new(file, rotation)
    @file_logger.sev_threshold = level