grep life by Jeremy Glover » Linux

How to backup your website

Jeremy — Wed, 17 Jun 2009 04:08:50 +0000

Everyone knows (or should by now) that cheap web hosts (Bluehost, Dreamhost, MediaTemple, etc.) don’t backup your data for you. So you’d better do it yourself. If you’re on any respectable host, you should have ssh access to the box.

Connect to your box via ssh and run the following commands to create a backup of your site.

cd ~ mkdir Backup nohup zip -r Backup/YYYY-MM-DD-HHMM.zip www/ > backup_log.txt &
(Replace YYYY with the 4-digit year, MM with the 2-digit month, HH with the 24-hour format of the hour, and MM with the 2-digit minute)

cd ~ navigates to your home folder

mkdir Backup creates the backup directory in which the backups will be stored

nohup is short for no hangup and allows processes started by users at the terminal to continue running even after the user logs out

zip is a program which combines many files into one and compresses them to make the end result even more portable

-r tells zip to burrow into all subdirectories in order to grab all of the files

Backup/YYYY-MM-DD-HHMM.zip is the path to the backup file

www/ is the directory to backup (it may be html, htdocs, httpdocs, etc. on your box)

> backup_log.txt redirects all output from zip to the backup_log.txt file so you can review the file later

& tells linux to run the zip program in the background so that you can logout or perform other tasks without killing the process

Now all you need to do is download that zipped file. Use your favorite SFTP client to login to your box and snag it. I recommend FileZilla Client for all platforms. If you’re looking for an FTP server, FileZilla Server is perfect.

No related posts.

Software RAID 1 Performance on Ubuntu Server 8.04

Jeremy — Fri, 31 Oct 2008 00:29:11 +0000

A generous gift was bestowed upon me last week — Dell’s PowerEdge SC420 server. It came with two (2) 80GB hard drives which I have since replaced with two (2) 1TB hard drives. I decided to mirror these drives for the sake of redundancy using Linux’s built-in software RAID drivers.

While synchronizing the drives, I decided to monitor the progress. What I saw was truly amazing.

jeremy@fileserver1:~$ cat /proc/mdstat Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]

md0 : active raid1 sda1[0] sdb1[1] 879084736 blocks [2/2] [UU] [=>...................] resync = 8.1% (71688192/879084736) finish=116.2min speed=115709K/sec

Holy cow! 116MB/s!

That means the SATA bus is processing at least 332MB/s since it has to read the data from one drive and then write the data to the other drive. That’s 2,656Mb/s or 2,656,000,000 bits per second which equates to 2.7 bits per picosecond! I’d like to see anyone do anything 2.7 times in one picosecond.

I figured with this much throughput the CPU would be chugging away. Nope. 4%.

No related posts.

Install SSH Server on Ubuntu

Jeremy — Sat, 25 Oct 2008 16:52:03 +0000

It really couldn’t be any easier than this:

sudo apt-get install openssh-server

You’re done. Now you can use ssh to remotely connect to your Ubuntu box.

No related posts.

Where Is tracert in Ubuntu?

Jeremy — Sat, 21 Jun 2008 04:55:08 +0000

It turns out traceroute (or tracert for Windows users) is not bundled with Ubuntu; you have to manually install it. If you’re looking for an alternative, simply use tracepath although it looks like there are some complaints. Apparently tracepath doesn’t deal with NAT properly but traceroute does so you might be better off installing it:

From the shell type “apt-get install traceroute”.

Voila.

No related posts.