eqhmcow

#!/usr/bin/env perl
use strict;
use warnings;

# apt-get install apache2-utils libyaml-syck-perl libtimedate-perl
use YAML::Syck;
use IO::File;
use Date::Parse;
use Time::Piece;

eqhmcow

#!/usr/bin/perl
use strict;
use warnings;

$|++;

use IPC::Run qw/start pump finish timeout/;
use Parallel::ForkManager;
use Getopt::Long;

eqhmcow

Big Data on Little Linux

eqhmcow

#!/bin/bash

# As the "bufferbloat" folks have recently re-discovered and/or more widely
# publicized, congestion avoidance algorithms (such as those found in TCP) do
# a great job of allowing network endpoints to negotiate transfer rates that
# maximize a link's bandwidth usage without unduly penalizing any particular
# stream. This allows bulk transfer streams to use the maximum available
# bandwidth without affecting the latency of non-bulk (e.g. interactive)
# streams.

eqhmcow

IPTABLES=iptables
IPV=${IPTABLES%tables} # ip for ipv4 | ip6 for ipv6
PROC_IPTABLES_NAMES=/proc/net/${IPV}_tables_names

/sbin/modprobe --version 2>&1 | grep -q module-init-tools \
    && NEW_MODUTILS=1 \
    || NEW_MODUTILS=0

# Do not stop if iptables module is not loaded.
[ -e "$PROC_IPTABLES_NAMES" ] || exit 1

eqhmcow

#!/usr/bin/perl
# example perl code, but this should now actually work properly, even on
# Windows

# thanks to everyone who tested this, reported bugs and suggested or
# implemented fixes!

# this code is licensed under GPL 2 and/or Artistic license;
# aka free perl software

eqhmcow

#!/bin/bash

# backup mysql tables - does an initial dump and then backs up to final
# location using rdiff-backup

# NOTE: do NOT manually edit or update the rdiff-backup managed final backup
# location. ONLY use rdiff-backup to update files in that backup directory. if
# the files are changed directly, the historical backups that rdiff-backup
# maintains will be corrupted.

eqhmcow

script to kill apache 1.3 / 2.x prefork httpd processes serving preconnect connections, in an attempt to prevent chrome from causing a DoS against the httpd

This is version 2 of https://gist.github.com/eqhmcow/4774549

The major change is the use of two processes instead of one:

One process makes server-status requests, the other kills idle prefork processes.

This allows the kill script to continue killing idle processes even when Chrome has effectively DoS'd the apache server. When this happens, the check script can't get an updated status response back immediately, but the kill script can hopefully free up a slot by killing some processes. Having an uninterrupted request in the socket queue allows us to get an updated response after killing the Chrome preconnections are killed.

eqhmcow

#!/bin/env perl

# kill_idle_cms_httpd.pl - kill idle httpd connections
# $Id$

use strict;
use warnings;

use Time::HiRes 'time';

eqhmcow

$ cat test.txt
18:36 <@Dop> also fox_1: the scottish accent isn't too bad but you need to slow it down quite a bit to get the full effect
18:37 <@fox_1> still need to tweak a little
18:37 <@fox_1> right now you're a female scot for example I think
18:37 <@Dop> och aye? rofl
18:38 <@fox_1> ;)
18:55 <@ytz> lol
18:56 <@fox_1> that was the facebook fight
18:56 <@ytz> time to assemble coffee before the main card
18:56 <@fox_1> yeah I got rockstar