Preparing Plone to start the WebDAV service and setting the permissions to allow the users to make use of it is only half the battle, actually using it, especially from automated systems like build servers is another struggle.

Using the Cadaver WebDAV client

Although WebDAV is currently well integrated in modern desktop environments, a CLI alternative is useful for automation, like Jenkins build scripts.

Automatic Login

Ideally we want password-less operation from script, both from a usability as from a security standpoint.

Cadaver supports automatically logging in to servers requiring authentication via a .netrc file, like the ftp client. The syntax is such that the file can be shared for both tools.

The file ~/.netrc may be used to automatically login to a server requiring authentication. The following tokens (separated by spaces, tabs or newlines) may be used:

machine host

Identify a remote machine host which is compared with the hostname given on the command line or as an argument to the open command. Any subsequent tokens up to the end of file or the next machine or default token are associated with this entry.

default

This is equivalent to the machine token but matches any hostname. Only one default token may be used and it must be after all machine tokens.

login username

Specifies the username to use when logging in to the remote machine.

password secret

Specifies the password to use when logging in to the remote machine. (Alternatively the keyword passwd may be used)

Example ~/.netrc file

default
login jenkins
passwd secret

Example Session

Troubleshooting

Error 409 Conflict

This can mean a lot of things, like a real conflict. However most of the time it means that the folder where the stuff is uploaded does not exist:

dav:/cmdb/Members/pti/> mput Dropbox/Apps/Byword/plone_webdav.md
Uploading Dropbox/Apps/Byword/plone_webdav.md to `/cmdb/...':
Progress: [=============================>] 100,0% of 1488 bytes failed:
409 Conflict

This actually tries to upload the file to a subfolder Dropbox/Apps/Bywordwhich does not exist, causing this confusing error.

Simply changing the local directory solves the issue:

dav:/cmdb/Members/pti/> lcd Dropbox/Apps/Byword
dav:/cmdb/Members/pti/> mput plone_webdav.md
Uploading plone_webdav.md to `/cmdb/Members/pti/plone_webdav.md':
Progress: [=============================>] 100,0% of 1488 bytes succeeded.

Cannot create folders using WebDAV

Problem: The WebDAV “make folder” method, MKCOL, requires the “Add Folders” permission. This is not normally granted to Members or Owners on the site.

dav:/cmdb/Members/jenkins/> mkcol test2
Creating `test2': Authentication required for Zope on server `cmdb-uat.elex.be':
Username: 
Password: 
Retrying: Authentication required for Zope on server `cmdb-uat.elex.be':
Username: Terminated by signal 2.

Plone asks to login again because the current user has insufficient rights.

Workaround: In the Zope Management Interface, under the “Security” tab for the Plone root, check the “Owners” and “Managers” box for the “Add Folders” permission setting.

~  ᐅ cadaver dav://cmdb-uat.elex.be:1980/cmdb/Members/jenkins
dav:/cmdb/Members/jenkins/> mkcol test2
Creating `test2': succeeded.
dav:/cmdb/Members/jenkins/> 

Source: Members Can’t Create Folders Through WebDAV

Automating Cadaver with davpush.pl

Cadaver uses an ftp like command language to interact with the WebDAV server. This is very flexible, but impractical when a large number of files and folders must be uploaded. This happens often when the documentation for a new release must replace the previous version.

Cadaver accepts its input on the stdin stream, which allows us to pipe a script of commands to it. Since it is non-trivial to create and maintain such a script by hand, a script generator is needed. The generator presented here is meant to be simple and easy to use and modify. No attempt was made to made to add advanced syncing (like removing deleted files), handle exceptions gracefully or ‘do the right thing’.

With that in mind, organize the docs in such a way that it is easy to delete the target folder and push a fresh copy to clean everything up. This is common (and good) practice anyway in order to effectively use relative links within a subsite.

The principle is to cd to the root directory of the documentation root and run the script there and point it to the target.

Usage

davpush.pl dav://_hostname_:_port_/_upload path_

Uploads all files and folders recursively to the WebDAV folder passed in the url.

Code

#!/usr/bin/perl
use File::Find;

my $script = "";

sub wanted() {
  my $f = $File::Find::name;
  if (-f $f) {
    $script .= "put $f\n";
  } else {
    $script .= "cd $target_dir\n";
    $script .= "mkdir $f\n";
    $script .= "cd $f\n"
  }
}

my $url = $ARGV[0];
print "URL: $url";

if ($url =~ m#dav://.*?(/\S*)#) {

  my $target_url = "$0";
  my $target_dir = "$1";

  find({'wanted'=>\&wanted, 'no_chdir' => 1},   ".");


  $pid = open(POUT, "| cadaver $url");
  print POUT $script;
  print POUT "bye\n";
  close POUT;

} else {
  print "Usage: davpush.pl dav://<hostname>:<port>/<upload path>\n";
  print "\n";
  print "Uploads all files and folders recursively to the WebDAV folder passed in the url.";
}

Code Notes

The standard perl File::Find module traverses the folder tree in the right order to make sure all folders are created before other files or folders are created in them. Default behavior is to chdir to the directory, but then we lose the nice paths relative from the root, which would require additional administration entering and leaving the directory. Setting the no_chdir flag in the options keeps the paths like we want them in the script. (Look at the preprocess and postprocess options to help with the directory admin, but I think the added complexity will outweigh the gains for small to moderate trees)

For every file or folder, the wanted subroutine is called. For files we just add a mput command to copy the file over, because it keeps the path intact. If there is a file already (and the permissions are not screwed up) then it is overwritten. When we enter a new folder then we create the folder. If the folder already exists we get a (harmless) 405 Method Not Allowed error. Here we make another offer to the God of Simplicity, and ignore it.

After walking the tree, we have the script in the $script variable. It is unceremoniously piped as input for cadaver. We add the bye command to close the session, and we’re done. The output of cadaver appears on the stdout for easy verification using a MkI Eyeball check or by piping it to grep.

Sample session

~/Dropbox/Apps/Byword  ᐅ perl ~/tmp/davpush.pl dav://cmdb-uat.elex.be:1980/cmdb/Members/pti
URL: dav://cmdb-uat.elex.be:1980/cmdb/Members/ptiCreating `.': failed:
405 Method Not Allowed
Uploading ./plone_webdav.html to `/cmdb/Members/pti/plone_webdav.html':
Progress: [=============================>] 100,0% of 4007 bytes succeeded.
Uploading ./plone_webdav.md to `/cmdb/Members/pti/plone_webdav.md':
Progress: [=============================>] 100,0% of 6369 bytes succeeded.
Uploading ./Untitled.txt to `/cmdb/Members/pti/Untitled.txt':
Progress: [=============================>] 100,0% of 203 bytes succeeded.
Uploading ./Uploading to `/cmdb/Members/pti/Uploading': Could not open file: No such file or directory
Uploading documents to `/cmdb/Members/pti/documents': Could not open file: No such file or directory
Uploading to to `/cmdb/Members/pti/to': Could not open file: No such file or directory
Uploading Plone to `/cmdb/Members/pti/Plone': Could not open file: No such file or directory
Uploading with to `/cmdb/Members/pti/with': Could not open file: No such file or directory
Uploading WebDAV.md to `/cmdb/Members/pti/WebDAV.md': Could not open file: No such file or directory
Creating `./foo': failed:
405 Method Not Allowed
Creating `./foo/bar': failed:
405 Method Not Allowed
Creating `./foo/bar/baz': failed:
405 Method Not Allowed
Uploading ./foo/bar/baz/plone_webdav.md to `/cmdb/Members/pti/foo/bar/baz/plone_webdav.md':
Progress: [=============================>] 100,0% of 3380 bytes succeeded.
Creating `./images': failed:
405 Method Not Allowed
Uploading ./images/SJ09_1.jpg to `/cmdb/Members/pti/images/SJ09_1.jpg':
Progress: [=============================>] 100,0% of 31637 bytes succeeded.
Uploading ./images/SJ09_2.jpg to `/cmdb/Members/pti/images/SJ09_2.jpg':
Progress: [=============================>] 100,0% of 29182 bytes succeeded.
Uploading ./images/SJ09_3.jpg to `/cmdb/Members/pti/images/SJ09_3.jpg':
Progress: [=============================>] 100,0% of 31296 bytes succeeded.
Uploading ./images/SJ09_4.jpg to `/cmdb/Members/pti/images/SJ09_4.jpg':
Progress: [=============================>] 100,0% of 31094 bytes succeeded.
Uploading ./images/SJ09_5.jpg to `/cmdb/Members/pti/images/SJ09_5.jpg':
Progress: [=============================>] 100,0% of 26886 bytes succeeded.
Uploading ./images/SJ09_6.jpg to `/cmdb/Members/pti/images/SJ09_6.jpg':
Progress: [=============================>] 100,0% of 29373 bytes succeeded.
Uploading ./images/SJ09_7.jpg to `/cmdb/Members/pti/images/SJ09_7.jpg':
Progress: [=============================>] 100,0% of 34486 bytes succeeded.
Uploading ./images/SJ09_8.jpg to `/cmdb/Members/pti/images/SJ09_8.jpg':
Progress: [=============================>] 100,0% of 28561 bytes succeeded.
Uploading ./images/SJ09_9.jpg to `/cmdb/Members/pti/images/SJ09_9.jpg':
Progress: [=============================>] 100,0% of 27381 bytes succeeded.
Connection to `cmdb-uat.elex.be' closed.
Share

2011-07-22 Fri 16:33
Figure out proxy settings Linux

Ubuntu has a Network Proxy chooser which allows you to select a location (a la MacOSX). This works well enough except that the UI is a bit counter-intuitive (in my humble opinion)which causes me to regularly nuke some predefined setting inadvertently. This is not a big deal though.

However for update manager (and several other tools) to pick up the new proxy settings you need to push the settings down to the system level. This takes 2 times typing your password. Now, this IS a big deal.

When I go back and forth between work and home I have to change this at least 2 times per day. Also it irks me that a detail setting like the proxy is not auto-detected and I need to login to change this ‘system’ setting. My laptop is essentially a single user system and I do not see switching the proxy as a serious security issue, even with 3 kids running around the home.

To come back to auto-detection, while this works fine at work, it fails to figure out that at home that there is a direct connection to the Internet. I can probably fix this by replacing my aging wireless router with my Time Capsule as the Internet gateway router, but I prefer to have the Time Capsule close to my desk.

In any case the Network proxy shows 2 times the authentication dialog box. A particularly nice feature (Is this new in Natty?) is that the dialog shows for which DBUS setting access is being asked.

The first dialog asks access to com.ubuntu.systemservice.setProxy. This response is configured in the file /usr/share/polkit-1/actions/com.ubuntu.systemservice.policy. This is a very readable XML file which contains a section for the setProxy action. I feel no reservation in allowing unchecked access to the setProxy. Although this might make a man-in-the-middle attack easier someone with the sophistication to pull this off, does not need to doctor my PC to do it.

<action id="com.ubuntu.systemservice.setproxy">
  <description>Set current global proxy</description>
  <message>System policy prevents setting proxy settings</message>
  <defaults>
    <!-- PTI : original settings
    <allow_inactive>no</allow_inactive>
    <allow_active>auth_admin_keep</allow_active>
    -->
    <allow_inactive>yes</allow_inactive>
    <allow_active>yes</allow_active>
  </defaults>
</action>

Retrying, and indeed one of the authentication requests dropped.

Note that the action was configured with auth_admin_keep which according to the docs would mean we should be authenticated for some time,so I would not expect the second authentication I am getting. Must be a subtlety which escapes me at the moment.

The second action is more problematic since the set-system om the system gconf settings is much less fine-grained than setProxy and can potentially cause more damage to the system.

<action id="org.gnome.gconf.defaults.set-system">
  <description gettext-domain="GConf2">Change GConf system values</description>
  <message gettext-domain="GConf2">Privileges are required to change GConf system values</message>
  <defaults>
    <allow_inactive>no</allow_inactive>
    <!-- PTI: Original setting
    <allow_active>auth_admin</allow_active>
    -->
    <allow_active>yes</allow_active>
  </defaults>
</action>

After relaxing this second method, I can finally easily switch proxies between Locations.

There are several things bugging me:

  1. The set-system method really is too wide in scope.
  2. There should be a more elegant way that modifying files under /usr/share
  3. My system should actually switch location unaided.

For the time being, I fixed a frustration and learned something in the process. The result is not yet fully satisfactory, but that will improve over time. **

Of course there is a more elegant solution than editing files under /usr/share folder.

Everything is explained in the man pages pklocalauthority and PolicyKit.conf (among others).

But that’s for another day…

Share

Overview and Goals

We build our solutions mostly on Ubuntu Natty and we deploy to Debian (currently lenny). One problem we face is that Debian has a slow release cycle and the packages are dated. Before a new release is approved and deployed to our target servers it can still take many months causing us to have to use up to 3 year old technology.

So we are often faced to ‘backport’ packages or debianize existing packages if we want to use the current releases.

In the past we had different build servers for the target architectures. However this is a heavy solution and scales poorly. It also makes upgrading to the next release that much heavier.

So we need a system for building debian packages that is :

  1. Fully automated
  2. Target multiple distributions Debian (lenny,squeeze) and Ubuntu(natty, maverick)
  3. Build on development machines(a) and Jenkins/Hudson CI servers(b)
  4. easily configurable
  5. memorizable process

The goal is to make packages for internal consumption, and the process outlined here falls short of the community standards.

Enter pbuilder

Of course we are not the first or only one with this issue. In fact we are laggards and there are excellent articles on the ‘net to help us with these goals.

e.g.

The pbuilder program create a clean room environment of a freshly installed empty debian or ubuntu distro, chroot into it and starts building based on the project metadata, mostly from the debian/control file.

It does this by unpacking a preconfigured base image of the selectable target , installing the build dependencies, building the package in the cleanroom, moving the artifacts to the hosting machine and cleaning everything up again. And it does this actually surprisingly fast. This clearly satisfies goals 1 and 2 (and half of 3 if we assume a developer has full control over his laptop).

The pbuilder is configured through commandline options, which are clear and friendly enough but you end up with commandlines of several lines long which are impossible to type in a shell and are a maintenance nightmare in build scripts (clearly conflicts with point 5). Also in the ideal world we would be able to retarget a build without touching the checked out files, e.g. with environment variable (see goals 3 and 4).

Configuring pbuilder

On the Pbuilders Tricks page I found a big smart shell script to use as the pbuilder configuration file ~/.pbuilderrc.

# Codenames for Debian suites according to their alias. Update these when
# needed.
UNSTABLE_CODENAME="sid"
TESTING_CODENAME="wheezy"
STABLE_CODENAME="squeeze"
OLDSTABLE_CODENAME="lenny"
STABLE_BACKPORTS_SUITE="$STABLE_CODENAME-backports"

# List of Debian suites.
DEBIAN_SUITES=($UNSTABLE_CODENAME $TESTING_CODENAME $STABLE_CODENAME $OLDSTABLE_CODENAME
    "unstable" "testing" "stable" "oldstable")

# List of Ubuntu suites. Update these when needed.
UBUNTU_SUITES=("natty" "maverick" "jaunty" "intrepid" "hardy" "gutsy")

# Mirrors to use. Update these to your preferred mirror.
DEBIAN_MIRROR="ftp.be.debian.org"
UBUNTU_MIRROR="mirrors.kernel.org"

# Optionally use the changelog of a package to determine the suite to use if
# none set.
if [ -z "${DIST}" ] && [ -r "debian/changelog" ]; then
    DIST=$(dpkg-parsechangelog | awk '/^Distribution: / {print $2}')
    # Use the unstable suite for Debian experimental packages.
    if [ "${DIST}" == "experimental" ]; then
        DIST="unstable"
    fi
fi

# Optionally set a default distribution if none is used. Note that you can set
# your own default (i.e. ${DIST:="unstable"}).
: ${DIST:="$(lsb_release --short --codename)"}

# Optionally set the architecture to the host architecture if none set. Note
# that you can set your own default (i.e. ${ARCH:="i386"}).
: ${ARCH:="$(dpkg --print-architecture)"}

NAME="$DIST"
if [ -n "${ARCH}" ]; then
    NAME="$NAME-$ARCH"
    DEBOOTSTRAPOPTS=("--arch" "$ARCH" "${DEBOOTSTRAPOPTS[@]}")
fi

BASETGZ="/var/cache/pbuilder/$NAME-base.tgz"
DISTRIBUTION="$DIST"
BUILDRESULT="/var/cache/pbuilder/$NAME/result/"
APTCACHE="/var/cache/pbuilder/$NAME/aptcache/"
BUILDPLACE="/var/cache/pbuilder/build/"

# make sure folders exist
mkdir -p $BUILDRESULT
mkdir -p $APTCACHE

echo "Target : $BUILDRESULT" >>/tmp/dist

if $(echo ${DEBIAN_SUITES[@]} | grep -q $DIST); then

    OTHERMIRROR="deb file:///var/cache/pbuilder/$NAME/result ./"
    BINDMOUNTS="/var/cache/pbuilder/$NAME/result"
    HOOKDIR="/var/cache/pbuilder/$NAME/hooks"
    EXTRAPACKAGES="apt-utils"
    # Debian configuration
    MIRRORSITE="http://$DEBIAN_MIRROR/debian/"
    COMPONENTS="main contrib non-free"
    DEBOOTSTRAPOPTS=("${DEBOOTSTRAPOPTS[@]}" "--keyring=/usr/share/keyrings/debian-archive-keyring.gpg")
    if $(echo "$STABLE_CODENAME stable" | grep -q $DIST); then
        EXTRAPACKAGES="$EXTRAPACKAGES debian-backports-keyring"
        OTHERMIRROR="$OTHERMIRROR | deb http://www.backports.org/debian $STABLE_BACKPORTS_SUITE $COMPONENTS"
    fi
elif $(echo ${UBUNTU_SUITES[@]} | grep -q $DIST); then
    # Ubuntu configuration
    MIRRORSITE="http://$UBUNTU_MIRROR/ubuntu/"
    COMPONENTS="main restricted universe multiverse"
    DEBOOTSTRAPOPTS=("${DEBOOTSTRAPOPTS[@]}" "--keyring=/usr/share/keyrings/ubuntu-archive-keyring.gpg")
else
    echo "Unknown distribution: $DIST"
    exit 1
fi

I just updated the distribution names to the current situation and added the directory where the packages are collected as a repository so subsequent builds can use these packages as dependencies. I also specified the keyrings to use for Debian and Ubuntu and made sure the expected folders are created to mount them in the clean room.

I created this in my account on my development laptop and added a symbolic link in ~root/.pbuilderrc to this file so I can update it from my desktop environment and do not have to get my brain all twisted up to try to remember with which configuration I am busy in my shell, sudo, su -, …

THe way the script works is that the configuration adapts itself to the content of the DIST and ARCH environment variables. So to configure lenny-amd64 as target is sufficient to do

~ > export DIST=lenny
~ > export ARCH=amd64

This approach is also perfect Jenkins or Hudson to determine the target build from checked out sources, since it can be specified in the build recipe. (satisfies goals 3b, 4 and 5)

Since we have to run these programs using sudo we must make sure the environment variables are passed by sudo. We can do this in the Defaults line of the /etc/sudoers file with the envkeep instruction.

...
Defaults  env_reset,env_keep="DIST ARCH http_proxy ftp_proxy \
 https_proxy no_proxy"
... snip ...
# Cmnd alias specification
Cmnd_Alias PBUILDER=/usr/sbin/pbuilder, /usr/bin/pdebuild
... snip to end of file ...
# Allow members of group sudo to execute any command
%sudo   ALL=(ALL:ALL) ALL

pti     ALL=(ALL) NOPASSWD: PBUILDER
jenkins ALL=(ALL) NOPASSWD: PBUILDER
#includedir /etc/sudoers.d

You add the DIST and ARCH variables there. I also included the environment variables for proxying so I can easily switch between environment on my laptop and these changes propagate to sudo (which is also useful for plain apt-get, by the way).

I also added a line to show how to make the tools available for a user without having to give their password. This is not needed for interactive work, but very much so for the user as which the CI server is running (in our case jenkins). Note that the definition should be after the group definitions, otherwise these take precedence and jenkins has to provide his password (read: is hanging during the build).

Creating the target base images

The heavy lifting is now done. Let’s create an base.tgz for lenny-amd64.

~ > export DIST=lenny
~ > export ARCH=amd64
~ > sudo pbuilder create

Now go and have a cup of coffee (or read some emails).

Rinse and repeat for the other target platforms.

Backporting existing packages

In theory backporting would be as simple as

~  ᐅ cd tmp
~/tmp  ᐅ apt-get source mongodb
Reading package lists... Done
Building dependency tree
Reading state information... Done
Need to get 1,316 kB of source archives.
Get:1 http://be.archive.ubuntu.com/ubuntu/ natty/universe mongodb 1:1.6.3-1ubuntu2 (dsc) [2,276 B]
Get:2 http://be.archive.ubuntu.com/ubuntu/ natty/universe mongodb 1:1.6.3-1ubuntu2 (tar) [1,285 kB]
Get:3 http://be.archive.ubuntu.com/ubuntu/ natty/universe mongodb 1:1.6.3-1ubuntu2 (diff) [29.0 kB]
Fetched 1,316 kB in 1s (679 kB/s)
gpgv: Signature made Thu 17 Mar 2011 11:49:37 PM CET using RSA key ID D5946E0F
gpgv: Can't check signature: public key not found
dpkg-source: warning: failed to verify signature on ./mongodb_1.6.3-1ubuntu2.dsc
dpkg-source: info: extracting mongodb in mongodb-1.6.3
dpkg-source: info: unpacking mongodb_1.6.3.orig.tar.gz
dpkg-source: info: unpacking mongodb_1.6.3-1ubuntu2.debian.tar.gz
dpkg-source: info: applying debian-changes-1:1.6.3-1
dpkg-source: info: applying build-process-remove-rpath
dpkg-source: info: applying mozjs185
~/tmp  ᐅ DIST=lenny ARCH=amd64 sudo pbuilder build mongodb_1.6.3-1ubuntu2.dsc
I: using fakeroot in build.
I: Current time: Thu Jul 14 14:28:17 CEST 2011
I: pbuilder-time-stamp: 1310646497
I: Building the build Environment
I: extracting base tarball [/var/cache/pbuilder/lenny-amd64-base.tgz]
...

and you should get a nice set of debian packages in */var/cache/pbuilder/lenny-amd64.

In practice you will often end up with errors like :

... snip ...
The following packages have unmet dependencies:
  pbuilder-satisfydepends-dummy: Depends: xulrunner-dev (>= 2.0~) but it is not installable
The following actions will resolve these dependencies:

Remove the following packages:
pbuilder-satisfydepends-dummy

Score is -9850

Writing extended state information... Done
... snip ...
I: cleaning the build env
I: removing directory /var/cache/pbuilder/build//6279 and its subdirectories

In these case you have to walk the dependency tree till you find the leafs, and walk back up the branches to the trunk. Note also that chances are that unless you target machines which only serve a very specific purpose, you might end up with packages which are uninstallable since you pull out the rug from other installed packages. However we have the principle to use 1 virtual host to deliver 1 service, hence there are very little packages deployed to them and nothing complicated like desktop environments.

Simple leaf packages often build without a hitch:

~/tmp  ᐅ DIST=lenny sudo pbuilder build libevent_1.4.13-stable-1.dsc
I: using fakeroot in build.
I: Current time: Thu Jul 14 14:44:00 CEST 2011
I: pbuilder-time-stamp: 1310647440
I: Building the build Environment
I: extracting base tarball [/var/cache/pbuilder/lenny-amd64-base.tgz]
I: creating local configuration
I: copying local configuration
I: mounting /proc filesystem
I: mounting /dev/pts filesystem
I: Mounting /var/cache/pbuilder/ccache
... snip ...
dpkg-genchanges: including full source code in upload
dpkg-buildpackage: full upload (original source is included)
W: no hooks of type B found -- ignoring
I: Copying back the cached apt archive contents
I: unmounting /var/cache/pbuilder/lenny-amd64/result filesystem
I: unmounting /var/cache/pbuilder/ccache filesystem
I: unmounting dev/pts filesystem
I: unmounting proc filesystem
I: cleaning the build env
I: removing directory /var/cache/pbuilder/build//15214 and its subdirectories
I: Current time: Thu Jul 14 14:49:57 CEST 2011
I: pbuilder-time-stamp: 1310647797
~/tmp  ᐅ ls -al /var/cache/pbuilder/lenny-amd64/result
total 5260
drwxr-xr-x 2 root root    4096 2011-07-13 19:47 .
drwxr-xr-x 5 root root    4096 2011-07-13 19:13 ..
-rw-r--r-- 1 pti  pti     2853 2011-07-14 14:49 libevent_1.4.13-stable-1_amd64.changes
-rw-r--r-- 1 pti  pti     9129 2011-07-14 14:49 libevent_1.4.13-stable-1.diff.gz
-rw-r--r-- 1 pti  pti      907 2011-07-14 14:49 libevent_1.4.13-stable-1.dsc
-rw-r--r-- 1 pti  pti   499603 2009-12-05 23:04 libevent_1.4.13-stable.orig.tar.gz
-rw-r--r-- 1 pti  pti    61956 2011-07-14 14:49 libevent-1.4-2_1.4.13-stable-1_amd64.deb
-rw-r--r-- 1 pti  pti    31262 2011-07-14 14:49 libevent-core-1.4-2_1.4.13-stable-1_amd64.deb
-rw-r--r-- 1 pti  pti   172950 2011-07-14 14:49 libevent-dev_1.4.13-stable-1_amd64.deb
-rw-r--r-- 1 pti  pti    51588 2011-07-14 14:49 libevent-extra-1.4-2_1.4.13-stable-1_amd64.deb
-rw-r--r-- 1 root root    9051 2011-07-14 14:48 Packages
~/tmp  ᐅ

Using pdebuild for building packages

Many of our packages are debianized and can be build using debuild.

I use here the Ubuntu sources of tokyocabinet as an example (which uses the libevent package we just built, btw):

~/tmp/tokyocabinet-1.4.37  ᐅ DIST=lenny ARCH=amd64 pdebuild
...snip...
 dpkg-genchanges  >../tokyocabinet_1.4.37-6ubuntu1_amd64.changes
dpkg-genchanges: not including original source code in upload
dpkg-buildpackage: binary and diff upload (original source NOT included)
W: no hooks of type B found -- ignoring
I: Copying back the cached apt archive contents
I: unmounting /var/cache/pbuilder/lenny-amd64/result filesystem
I: unmounting /var/cache/pbuilder/ccache filesystem
I: unmounting dev/pts filesystem
I: unmounting proc filesystem
I: cleaning the build env
I: removing directory /var/cache/pbuilder/build//4199 and its subdirectories
I: Current time: Thu Jul 14 15:05:27 CEST 2011
I: pbuilder-time-stamp: 1310648727
~/tmp/tokyocabinet-1.4.37  ᐅ ls /var/cache/pbuilder/lenny-amd64/result
...snip...
tokyocabinet_1.4.37-6ubuntu1_amd64.changes
tokyocabinet_1.4.37-6ubuntu1.debian.tar.gz
tokyocabinet_1.4.37-6ubuntu1.dsc
tokyocabinet_1.4.37.orig.tar.gz
tokyocabinet-bin_1.4.37-6ubuntu1_amd64.deb
tokyocabinet-doc_1.4.37-6ubuntu1_all.deb

Sometimes the dependencies break on the version of debhelpers. This version is added conservatively by the dh* scripts and often is overly conservative. Many packages build just fine with older versions of the debhelpers. ** *

Setting up automated build

To set this up on the build server we have to replicate the steps above

  1. Create the ~/.pbuilderrc file
  2. Symbolic link to this file in ~root/.pbuilderrc
  3. Allow jenkins to use sudo for building packages
  4. Create a jenkins job to (re)build the packages
  5. Create jobs to build the packages

* *

Share