Recently I've discovered a secret feature in AutoYaST. Well, probably not so secret, because a SLES11 user (from our Two-Letter Customer) discovered it as well and reported a bug about it :) But all in all, if you googled for the keyword (keep_install_network) a week ago, the only hits were an AutoYaST changelog with short notice "I've added this feature" and few questions about it on opensuse-autoinstall mailing list. The documentation did not reveal any more details.

So - what does keep_install_network parameter, set to 'true' and inserted into <networking> section of AutoYaST profile, do? One can intuitively guess that it will preserve network configuration - that is, interfaces setup, resolv.conf bits, static routing, udev rules & co. - from 1st stage of installation (of course, only if the installation actually runs over the network) and it is indeed like that. Uwe blogs about it in more details

But unless you use openSUSE Factory, please don't try this at home :)

<networking>
  <keep_install_network config:type="boolean">true</keep_install_network>
</networking>

Network interfaces setup (ifcfg files) from the installation will be successfully preserved, but due to a bug, your static routing configuration (if you have any) will be moved into backup file and YaST won't create a new one and you will lose most of the information in /etc/resolv.conf. Unfortunately, nobody got the idea to test with the profile containing nothing more but keep_install_network entry in networking section so far :(

The bug is fixed now for openSUSE 11.2 and SLE11 SP1 (if nothing in particular sub-section of networking section is defined, the setup from installation is used, but AutoYaST profile is the higher authority here). However, if you use SLE11, you'd better install with profile containing at least minimal DNS and routing info, for example like this (set search domain, 1 nameserver and default gateway):

<networking>
    <keep_install_network config:type="boolean">true</keep_install_network>
    <dns>
      <dhcp_hostname config:type="boolean">false</dhcp_hostname>
      <domain>home.net</domain>
      <hostname>dolphin</hostname>
      <nameservers config:type="list"> 
        <nameserver>192.168.0.2</nameserver>
      </nameservers>
      <searchlist config:type="list">
        <search>home.net</search>
      </searchlist>
    </dns>
    <routing>
      <ip_forward config:type="boolean">false</ip_forward>
      <routes config:type="list">
        <route>
          <destination>default</destination>
          <device>-</device>
          <gateway>192.168.0.1</gateway>
          <netmask>-</netmask>
        </route>
      </routes>
    </routing>
  </networking>

Or use a workaround post-script Uwe posted to bugzilla, to backup your setup and to restore it later.

One good thing about these bugs is that not only it broadens your knowledge (when user comes up with scenario you never thought of) ;-) but it also helps us to improve things. keep_install_network feature is now documented and not secret anymore. Enjoy it!

Hi openSUSE community!

I’m glad my proposal was accepted and today I want to introduce myself and my GSoC project.

/me , thats Jan-Simon Möller and I’m just finishing my Diploma in electrical engineering at the Leibniz Universität Hannover. I’m coordinator of the openSUSE Weekly Newsletter and contribute also to the hamradio repository, the iFolder project and the openSUSE Build Service. See also my “People of openSUSE” interview.

My Project in short:  openSUSE@ARM
My aim during GSoC 2009 is to port first the base to the ARM platform. Then KIWI needs also some attention when it comes to imaging and after that the tools, Kernel and X11.

I’ll heavily use the capabilities of the openSUSE Build Service, which is now ready for ARM.

During the last few days, I’ve done many little preparations to get it all flying when GSoC coding period starts.

Stay tuned !

If you have a running program in Python and press Ctrl+C, you’ll get a traceback like this:

$ scout java foo
^CTraceback (most recent call last):
  File "/usr/bin/scout", line 11, in 
    ret = scout.ScoutCore.run()
  File "/usr/lib64/python2.6/site-packages/scout/__init__.py", line 945, in run
    result = module.ScoutModule().main(clp.module_args)
  File "/usr/lib64/python2.6/site-packages/scout/__init__.py", line 873, in main
    return self.do_query(args.query, repos, args.inversesearch)
  File "/usr/lib64/python2.6/site-packages/scout/__init__.py", line 890, in do_query
    result.add_rows(self._query(repo, query, inversesearch))
  File "/usr/lib64/python2.6/site-packages/scout/__init__.py", line 896, in _query
    r = db.query(self._sql, '%%%s%%' % term)
  File "/usr/lib64/python2.6/site-packages/scout/__init__.py", line 485, in query
    if len(row) == 1:           #(2)
KeyboardInterrupt

It is useful suppress it, because user knows he breaks the program and this output should be considered as a bug. Possible solution is wrap a main function by one big try: except KeyboardInterrupt:

try:
  main() # the main function
except KeyboardInterrupt:
  pass # KeyboardInterrupt supressed

But it makes a new level of indentation which should be uncomfortable – especially in Python. Or when you have multiple entry-points, or just don’t well structured program (which is common when you write your private helper script :)), you maybe prefer another solution.

Python has a sys.excepthook, which is called for traceback printing, so we could define our own and suppress unnecessary output here. And it would be nice suppress only one exception and handle other ones using existing function. And this function make it:

def suppress_keyboard_interrupt_message():
    old_excepthook = sys.excepthook

    def new_hook(type, value, traceback):
        if type != exceptions.KeyboardInterrupt:
            old_excepthook(type, value, traceback)
        else:
            pass

    sys.excepthook = new_hook

Function suppress_keyboard_interrupt_message (it is really nice name, don’t it ;-)) stores an existing hook and register an inner function new_hook as a new one. Advantage is that old_excepthook exists only in a scope of this function, so you don’t need use global variables for it.

Update: typos fixed

Ruby has a nice, but very dangerous feature called open classes. That means you can extend any class definition if you want it.

This example shows it

#!/usr/bin/ruby
class Foo
    def do_foo
        return "foo"
    end
end

aFoo = Foo.new()

puts aFoo.do_foo()

class Foo
    def do_bar
        return "bar"
    end
end

puts aFoo.do_bar()

That means you dynamically extend the definition of Foo class, so the last line prints “bar”.

Python behavior is different, because it does not allows this extending. Python just redefines a class Foo, so new instance of Foo will have a do_bar method only. But this not affected an existing ones, like Ruby does.

class Foo:

    def do_foo(self):
        return "foo"

aFoo1 = Foo()

class Bar(Foo):

    def do_nothing(self):
        pass

print issubclass(Bar, Foo)
print isinstance(aFoo1, Foo)

class Foo:
    
    def do_bar(self):
        return "bar"

aFoo2 = Foo()
aBar  = Bar()

print issubclass(Bar, Foo)
print isinstance(aFoo1, Foo)
print isinstance(aFoo2, Foo)

print dir(aFoo1)
print dir(aFoo2)
print dir(aBar)

But what about method injection? My solution is based on ideas of Recipe 81732: Dynamically added methods to a class.

import types

def funcToMethod(func, instance, method_name=None):
    cls = instance.__class__ if type(instance) != types.TypeType else instance
    setattr(cls, \
            method_name or func.__name__, \
            types.MethodType(func, instance, cls))

And that’s all. The funcToMethod bounds func to instances’s class and allows under method_name (default name is a same as a function one). So lets do some testing.

class Foo(object):

    def list(self):

        return [meth for meth in dir(self) if meth[:3] == 'do_' and type(getattr(self, meth)) == types.MethodType]

    def do_bar(self):
        
        return "bar"

def do_foo(inst):

    return "foo"

class Bar(Foo):

    def do_nothing(self):
        pass

aBar = Bar()

aFoo1 = Foo()
print aFoo1.list()
# calling it with class instead of instance is also possible and is equivalent
#funcToMethod(do_foo, Foo, "do_foo")
funcToMethod(do_foo, aFoo1, "do_foo")

aFoo2 = Foo()
print aFoo1.list()
print aFoo2.list()

print aFoo1.do_foo()
print aFoo2.do_foo()

print aBar.list()
print aBar.do_foo()

When you run the code, you will see that funcToMethod adds a new method to Foo class and this changes both existing and new instances as Ruby does too. And subclassing is not a problem, they are be affected too.

On Saturday April 25th, the Nicaraguan OpenSUSE Community joined forces with the Free Software Community (GUL-Nic) to celebrate the Latin American Free Software Installation Festival, (FLISOL in Spanish).

The Festival took place at the Universidad Nacional de Ingenieria Campus (UNI-IES), reaching more than 300 visitors.

Although there were several LUGs performing, OpenSUSE was one of the most solicited distros for installing. We installed the 11.1 version on 7 laptops.  And as usual, all others distros were after our stuffed penguins and geekos!

We continue working spreading the word of the Green Lizards!

*Update:  I corrected the name… shame of me! hehe*

Disclaimer: The information value of this posts tends to zero. Take it as an example of girlie chit-chat.

And now repeat after me:

To begin with - for those who don't know it yet, I got myself a new toy and installing openSUSE 11.1 on it and playing with it looked like a cool way to spend Friday night, now didn't it? Foolishly enough, I decided that I want to have dual boot with pre-installed Windows and I don't want to lose ThinkVantage stuff. Thus, before leaving the office for the weekend, I was briefed by our YaST bootloader wizard which options in bootloader configuration to tick and which leave blank - basically, I must not let YaST overwrite master boot record if I ever want to be able to use ThinkVantage. He did not forget to add that if anything went wrong, I should bring the box over to him on Monday and he'll fix it up for me ...

A small intermezzo: though I've been pretending that I'm doing some YaST development for almost 3 years already ;-) I'm not really a technical person. I know a lot about widgets, user interface, usability and all that fancy stuff, but my knowledge of the system configuration ends in /etc/sysconfig or /etc/fstab at most. When it comes to something like system boot, I'm completely lost. At least I admit to it :-) :-D

The installation went on fine - I picked KDE4 as my default desktop environment, had the partitioner shrink Windows partition, created root and /home partition, double-checked whether I got the bootloader installation options right ("Boot from root partition" - check, "Boot from MBR" - uncheck, "Set active flag for boot partition" - check, "Write generic code to MBR" - uncheck) and watched the slideshow. The standard set of 1st stage finish scripts followed and then bang! - "Failed to install the bootloader. Try the configuration again?" (or something along those lines). Hm, if I answer "Yes" now, I'm smart enough to figure out it'll fail again, I thought, unless I change something - but what? Looking at y2log did not make me any less helpless - the last thing I saw was how bootloader uses some cryptic Perl script (/usr/lib/YaST2/bin/tp_mbr, to be exact) to detect ThinkVantage sequence in MBR and preserve it.

Nevermind, I rebooted, picked "Repair installed system" from the installation options and hoped that yast2-repair will help me find out what went wrong. Not surprisingly, it did not. It only showed me already well-known message about failed bootloader installation - quite a cruel joke :) At this point, I gave up. Trying to solve a problem where I don't understand a single bit of its principles is not that much like me. And after all, there are more pleasant ways how to spend Friday night - dancing salsa, for example ... :)

But don't get desperate, my friends - this story has a good ending. I brought my laptop to the wizard :-) I was not looking over his shoulder as he was working, so I don't know what magic spells and potions he used, but I know he did the following:

1. Created partitioning setup with separate /boot partition and told the bootloader to boot from it (that was maybe the crucial point)
2. Added a repository with openSUSE 11.1 updates as an add-on during installation
3. Made extended partition (with swap, /, /home and /boot) 1 cylinder smaller, as ThinkVantage partition at the very end of the disk did not begin exactly at the cylinder boundary

The last two steps were probably superfluous and it could've well worked out without them, but ... now I can't tell. Anyway the installation finished, the system booted and now I have shiny new ThinkPad with openSUSE 11.1. In this post, I'm writing about interesting host naming schemes people use - as for me, all my machines have animal names. So, say hello to Dolphin - he is named after one of my favourite books by Eric Lambert (it is so little known that even Google does not find it ;-) )

In conclusion, I consider myself to be quite lucky that I'm surrounded by many wizards (bootloader wizard, installation wizard, KDE wizard, some wireless stuff wizards, suspend wizard, ... ) ;-) What do normal openSUSE users who can't grab their machine and bring it to a wizard do, if the same (or similar) thing happens to them? "Oh, they go to #suse and cry there. Or they simply give up and go using Ubuntu" says Darix (whom I sometimes chat with when I suffer from insomnia and he stays at work at night). Sad to say that, but it's probably true ... :-(

With the help of a unique identifier that is stored in every binary executable matching the executable, a coredump and the corresponding debuginfo together becomes really easy. You don’t need to know the package name and the version-release string to download and install the correct debuginfo package. This is achieved by extending the linker, some additional tools and the package management itself.

The build id is a unique identifier stored in the .note.gnu.build-id note section of the executable file and loaded into the process image during run-time. Different means to compute the unique identifier are supported although the default setting is to use a 20 byte SHA1 hash of the unstripped executable (see ld linker documentation for further information about the –build-id option).

To be able to read the build id from a core dump the kernel must include the ELF header pages in file-backed private memory areas (see documentation on /proc/<pid>/coredump_filter). This is the default setting on recent openSUSE kernel versions.

Different tools can be used to print out the build-id. eu-readelf (from the elfutils package) prints the contents of the note sections given the -n option in a human readable fashion. pbuildid (from the ptools package) is prints the build-id from executables and from core files including the build-id of all loaded shared objects during the crash.

Now, zypper can be asked to install the debuginfo package that provides the debuginfo for the given build-id.

# zypper install -C "debuginfo(build-id)=b75bab63c9a25eb13264bb95f1fef190e157f865"
Loading repository data...
Reading installed packages...
Resolving package dependencies...

The following NEW package is going to be installed:
  bash-debuginfo


Overall download size: 605.0 K. After the operation, additional 2.1 M will be used.
Continue? [YES/no]: yes
Retrieving package bash-debuginfo-3.2-148.2.x86_64 (1/1), 605.0 K (2.1 M unpacked)
Retrieving: bash-debuginfo-3.2-148.2.x86_64.rpm [done]
Installing: bash-debuginfo-3.2-148.2 [done]
# 

If you are really lazy you can call the following script to do the job for you:

#!/bin/bash                                                                     
#                                                                               
# Sample script how to install debuginfo packages by build-id                   
#                                                                               

IDS=''
for f in "$*"; do
    case "$(file -L $f)" in
        *ELF*)
            IDS+="$(pbuildid $f 2>/dev/null | awk '{print $NF}') "
            ;;
        *)  
            ;;
    esac
done

echo "Install Debuginfo for following build-ids: $IDS"

CMDLINE=""
for i in $IDS; do
    CMDLINE+="debuginfo(build-id)=$i "
done

echo $CMDLINE | xargs zypper install -C

The issue caused by the OBS worker update on arm builds is fixed by a new qemu.

This new qemu version also has fixed the Fedora 10 @ ARM build problem.

So we have the following working ARM target distros available for ARM: Fedora 10, Debian 5.0 and Ubuntu 9.04.

Have fun.

A lot of my day job now involves working with Ruby on Rails. At first I wasn’t sure how much I would like Rails or ruby, given that I had been doing a lot of C#/C/whatever desktop work before. Not surprisingly, though, I’ve become quite addicted. The test-driven nature of development is a welcome change — most desktop apps I worked on didn’t even have tests. The Rails community has done a great job of banging automated testing into people’s heads. Almost every tutorial, book, or random blog post I’ve seen emphasizes the importance of good automated tests. Hopefully it has helped decrease the instances of ‘snorpage’, but perhaps my co-workers would disagree :)

Anyway, I love Rails so much that I’ve converted my blog from Wordpress to Enki. Enki is more of a create-your-own-blog construction kit than a turn-key solution like Wordpress. That was one of the main reasons I chose it over Typo or Mephisto — I wanted to be able to easily hack on it.

I wrote a quick and dirty script to help me import the Wordpress posts into Enki. Any fellow Enki hackers can grab it here.