Victor Yang pydevops

localhost:9200/_cat/indices?pretty

basic concept

brew bundle --file=.Brewfile
brew bundle dump

apt-get install libmemcached-tools
memcstat --servers=203.0.113.12 11211 | grep get_hits

get version

memcstat --servers=10.2.0.4 | grep version

dump keys

memcdump --servers=localhost

dump value given a key

yum

sudo yum --enablerepo=rpmforge install tmux

yum clean all
yum info yum-plugin-versionlock

rpm

check health

(leader only) echo dump | nc localhost 2181
echo stat | nc localhost 2181 
echo conf | nc localhost 2181
echo cons | nc localhost 2181
echo envi | nc localhost 2181

https://github.com/knife-block/knife-block

# knife and berks switch server
# .chef/knife.rb
# .berkshelf/config.json
knife block list
knife block show
knife block use <server>

# metadata.rb version bump

	#!/bin/bash
	set -euo pipefail

	start_stop() {
	while read name zone ; do
	echo "$name with $zone"
	gcloud compute instances $CMD $name --zone $zone &
	done < <(gcloud compute instances list --filter='tags.items~^stoppable$' --format='value(NAME,ZONE)')
	wait
	gcloud compute instances list --filter=running

	>mdadm --create /dev/md10 --level=0 --raid-devices=3 /dev/xvdba /dev/xvdbb /dev/xvdbc
	mdadm: Defaulting to version 1.2 metadata
	mdadm: array /dev/md10 started.
	>mdadm --detail /dev/md10


	Query: ntpq -pn

	Update: sudo ntpdate -bvu 0.amazon.pool.ntp.org

	(needs to stop first, run ntpdate, then start)

	#A standard configured Salt master refreshes the Pillar data ones a minute. If you don’t have so much time, you can manually refresh it with:
	salt '*' saltutil.refresh_pillar