HOWTO sync your files/pt

In other languages: English | Afrikaans | Albanian | Arabic | Brazilian | Bulgarian | Catalan | Chinese | Croatian | Czech | Danish | Dutch | Esperanto | Estonian | Filipino | Finnish | Flemish | French | German | Greek | Hebrew | Hindi | Hungarian | Indonesian | Italian | Japanese | Latvian | Lithuanian | Macedonian | Malay | Malayalam | Norwegian (Bokmål) | Norwegian (Nynorsk) | Persian | Polish | Portuguese | Romanian | Russian | Serbian | Slovak | Slovenian | Spanish | Swedish | Turkish | Ukrainian | Urdu

Como sincronizar os teus ficheiros

Neste artigo nós iremos te explicar como sincronizar vários ficheiros através do teu computador e como recuperar informação a múltiplas contas shell. A solução abrangida neste artigo não irã ser necessária alguma instalação ou configuração especial no servidor remoto.

Como começar

Primeiro, tu precisas de criar "repositório" e colocar toda a informação que tu queres sincronizar.

mkdir ~/.home
cp -fr .tcshrc ~/.home
cp -fr .irssi ~/.home

Also, you should create Makefile in your repository. The idea is to run it after your data will be transfered to remote machine.

all:
	chmod 711	$(HOME)/.ssh \
			$(HOME)/.irssi
	chmod 600	$(HOME)/.tcshrc \
			$(HOME)/.bash_profile \
			$(HOME)/.logout \
			$(HOME)/private.tar.gz.bf64 \
			$(HOME)/CHECKSUM

We also recommend you to read this: HOWTO SSH without a password.

ClusterIt!

We will use clusterit to transfer data from your repository to remote machine. Clusterit is a collection of clustering tools to turn your ordinary everyday pile of UNIX workstations into a speedy parallel beast. This facilitates managing a set of program executing in parallel with a very flexible authentication scheme. If you don't want to use it, you will have to edit synchronization script. First, you need to install it. Under FreeBSD you may do:

make -C /usr/ports/net/clusterit/ WITHOUT_X11=yes install clean

Config file for clusterit is a file pointed to by CLUSTER enviroment variable. It's just a list of cluster nodes.

GROUP:shells
shellium.org
another.free.shell.isp
GROUP:work
work.machine1
work.machine2
LUMP:all
shells
work

Scripting

The following script will compress ~/private directory, encrypt it and copy all files to your cluster. After this "make" command will be executed on all cluster nodes.

#!/bin/sh
#
# $Header: $
#

passwd=""
prx=""
prx_type=""
repo="${HOME}/.home"
group=""
node=""
objs="private"
dir="${HOME}"

while getopts p:s:t:r:g:w:o:d:c:h opt; do
	case ${opt} in
		p) passwd=${OPTARG}	;;
		s) prx=${OPTARG}	;;
 		t) prx_type=${OPTARG}	;;
 		r) repo=${OPTARG}	;;
 		g) group=${OPTARG}	;;
		w) node=${OPTARG}	;;
		o) objs=${OPTARG}	;;
		d) dir=${OPTARG}	;;
		c) CLUSTER=${OPTARG}	;;
		h) echo backup data and sync configs with clusterit tool:
		   echo http://www.garbled.net/clusterit.html
		   exit 0		;;
		*)
	esac
done

CLUSTER=${CLUSTER:-"${HOME}/.cluster"}

if [ "${passwd}" = "" ]; then
	echo -n 'Password: '
	read passwd
fi

if [ "${prx}" = "" ]; then
	RCMD_CMD=${RCMD_CMD:-"ssh"}
	RCP_CMD=${RCP_CMD:-"scp"}
else
	RCMD_CMD="ssh -o ProxyCommand='nc -X ${prx_type} -x ${prx} %h %p'"
	RCP_CMD="scp -o ProxyCommand='nc -X ${prx_type} -x ${prx} %h %p'"
fi

cd ${dir}

echo compressing...
for obj in ${objs}; do
	tar czf ${repo}/${obj}.tar.gz ${obj}
done

cd ${repo}

echo encrypting...
: > CHECKSUM
for obj in ${objs}; do
	openssl bf -a -k ${passwd} -in ${obj}.tar.gz -out ${obj}.tar.gz.bf64
	rm -rf ${obj}.tar.gz
	openssl md5  -c ${obj}.tar.gz.bf64 >> CHECKSUM
	openssl sha1 -c ${obj}.tar.gz.bf64 >> CHECKSUM
	chmod 600 ${obj}.tar.gz.bf64
done
chmod 600 CHECKSUM

echo Total size: `du -sh ${repo} | cut -d '	' -f 1`

if [ "${group}" != "" ]; then
	group="-g ${group}"
fi

if [ "${node}" != "" ]; then
	node="-w ${node}"
fi

echo ClusterIt!
pcp -c -r ${group} ${node} `ls -a ${repo} | awk '! /^(.|..)$/ { print "'${repo}'/" $1 }'` .
dsh ${group} ${node} make

# # if you don't have ClusterIt...
#
# for host in `...`; do
#  scp -r ${repo}/ ${host}:. &
# done
#
# wait
#
# for host in `...`; do
#  ssh ${host} make &
# done