Recently I bought an Orange Pi 2G IoT (this little SoC is really great!). I installed the Ubuntu Server and managed to connect to it through the serial port and I could even update the system (only apt-get update) and install some packages, but here's the little problem: I would like to install packages offline (in my house I do not have an internet connection) and I tried to download the repository but it only lets me download about 9 megabytes and don't even start with the "pool" files where (I think) reside the packages, it only make a list of the files in that repo and download some of the "index.html" but nothing more.

I am using a PC with Debian Jessie amd64, I situate in a folder in a NTFS partition (check that I have enough space) and I have tried to mirror the repository from the console using the wget command, with the following syntax:

wget --mirror --convert-links --show-progress --recursive --wait=5 http://mirrors.ustc.edu.cn/ubuntu-ports/

Could someone with a better idea help me to download the repo to use it offline?

  • 1
    wgetrespects the robots.txt file when crawling recursively through web sites. You can override this behaviour with -e robots=off (see codenimbus.com/2010/08/02/override-robots-txt-with-wget)
    – ridgy
    Dec 29, 2017 at 16:58
  • Yeap, I see the robots.txt "disable" many of the folder to download (recursively?)...thanks for giving the quick questions...let my try
    – k.Cyborg
    Dec 29, 2017 at 17:08
  • That seems to work flawlessly!!! Thanks!
    – k.Cyborg
    Dec 29, 2017 at 17:42

2 Answers 2


You may be better off using debmirror instead of tryign to use wget since things won't all be linked, etc.

Here's the script I use to run debmirror note that you'll want to change the architecture, section, release, etc.


# Architecture (i386, powerpc, amd64, etc.)

# Section (main,contrib,non-free)

# Release of the system (squeeze,lenny,stable,testing,etc)

# Server name, minus the protocol and the path at the end

# Path from the main server, so http://my.web.server/$dir, Server dependant

# Protocol to use for transfer (http, ftp, hftp, rsync)

# Directory to store the mirror in

logfile=ubuntu-mirror.log-`date +%m-%d-%Y`

# Start script

debmirror -a $arch \
--no-source \
--slow-cpu \
--i18n \
--md5sums \
--progress \
--passive \
--verbose \
-s $section \
-h $server \
-d $release \
-r $inPath \
-e $proto \

# pull in all the Release.gz etc. stuff
for i in `echo $release | sed s/\,/\ /g`
     rsync -avrt rsync://us.archive.ubuntu.com/ubuntu/dists/$i $outPath/dists

# fix Translation files
cd $outPath/dists
for i in `find ./ -iname Translation-en.gz`
    do nn=`echo $i | sed s/\.gz//g`
    zcat $i > $nn
  • Wow! This script seems to be very useful! Let me modify and try it! Thanks!!!!
    – k.Cyborg
    Dec 29, 2017 at 17:13

wget respects the robots.txt file when crawling recursively through web sites. You can override this behaviour with -e robots=off More info here

Thanks to @ridgy

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .