- Grml - lists.mur.at

grml virtual machine
by John G. Heim 04 Jan '14

04 Jan '14

I want to create a custom grml ISO that has a kernel patched for hardware speech synthesiers. There's a bug in the kernel code so most hardware speech synths don't work. If I can put ma custom compiled kernel on a grml disk, I can use it with my hardware speech synth. The man page for grml-live recommends running it on a virtual machine running grml. That sounds like a good idea. I have a VMWare player virtual machine booted into grml from the latest 32-bit iso file. But I am uncertain as to what to do next. Should I use grml2usb or grml-debbootstrap or something else? What's the best way to set up a virtual machine to run grml?

2 1

Mirrors for GRML
by Timo Jacob 06 Dez '13

06 Dez '13

Hello, please add our mirrors to the public mirror list: + Mirrors Japan: http://mirrors.go-parts.com/grml/ ftp://mirrors.go-parts.com/grml/ rsync://mirrors.go-parts.com/mirrors/grml/ + Mirrors-usa: http://mirrors-usa.go-parts.com/grml/ ftp://mirrors-usa.go-parts.com/grml/ rsync://mirrors-usa.go-parts.com/mirrors/grml/ + Mirrors-uk: http://mirrors-uk.go-parts.com/grml/ ftp://mirrors-uk.go-parts.com/grml/ rsync://mirrors-uk.go-parts.com/mirrors/grml/ + Mirrors-russia: http://mirrors-ru.go-parts.com/grml/ ftp://mirrors-ru.go-parts.com/grml/ rsync://mirrors-ru.go-parts.com/mirrors/grml/ + Mirrors-australia: http://mirrors-au.go-parts.com/grml/ ftp://mirrors-au.go-parts.com/grml/ rsync://mirrors-au.go-parts.com/mirrors/grml/ + Mirrors-brazil: http://mirrors-br.go-parts.com/grml/ ftp://mirrors-br.go-parts.com/grml/ rsync://mirrors-br.go-parts.com/mirrors/grml/ Contact Name: Timo Jacob address: timojacob.gp(a)gmail.com Please let me know if you have any questions and I would be happy to answer them. Kind regards Timo Jacob

1 1

Cannot find code to color by file extensions on an ls
by John 22 Nov '13

22 Nov '13

I just opened a new bug report (http://bts.grml.org/grml/issue1286) wherein I ask to have some file extensions added to the list to color red on an `ls` but find myself clueless where exactly this "color these extensions red on an ls" resides. Â The only thing I can find in /etc/zsh/zshrc begins on line #3116 but I do not understand it. Â Can someone point me to where this list is defined? Â My request should be as easy as adding the new extensions to some array I would think. Thanks!

2 2

Re: [Grml] J'ai l'inux nom de code Canterbury
by chris21.roux＠free.fr 05 Nov '13

05 Nov '13

----- Mail original ----- De: grml-request(a)ml.grml.org Ã€: grml(a)ml.grml.org EnvoyÃ©: Vendredi 27 Septembre 2013 12:00:02 Objet: Grml Digest, Vol 95, Issue 9 Send Grml mailing list submissions to grml(a)ml.grml.org To subscribe or unsubscribe via the World Wide Web, visit http://ml.grml.org/mailman/listinfo/grml or, via email, send a message with subject or body 'help' to grml-request(a)ml.grml.org You can reach the person managing the list at grml-owner(a)ml.grml.org When replying, please edit your Subject line so it is more specific than "Re: Contents of Grml digest..." Today's Topics: 1. J'ai l'inux nom de code Canterbury (chris21.roux(a)free.fr) 2. Re: J'ai l'inux nom de code Canterbury (Thomas K?hler) ---------------------------------------------------------------------- Message: 1 Date: Thu, 26 Sep 2013 22:48:44 +0200 (CEST) From: chris21.roux(a)free.fr To: grml(a)ml.grml.org Subject: [Grml] J'ai l'inux nom de code Canterbury Message-ID: <1799077990.27381246.1380228524139.JavaMail.root(a)spooler6-g27.priv.proxad.net> Content-Type: text/plain; charset=utf-8 Le projet "J'ai l'inux" vise ? proposer une distribution Linux capable de filter les contenus notamment pornographiques, pour une utilisation ? l'?cole. Pour filtrer les contenus, il est n?cessaire d'intervenir ? l'int?rieur du syst?me pour intercepter les contenus ? filtrer (on ne peut pas intervenir au niveau des trames r?seau car ? ce niveau les contenus ne sont pas interp?tables). J'ai pens? ? utiliser des machines virtuelles pour d?l?guer ? chaque machine virtuelle une portion sp?cifique du syst?me ? assurer. Cette id?e est n?e d'un poisson d'avril (April's fool joke) qui annon?ait le projet Canterbury, visant ? r?unir le meilleur de 5 distributions Linux dans une nouvelle distribution. En tant que chr?tien, je suis tr?s sensible aux poissons d'avril qui ont pour objectif de se moquer des chr?tiens. J'ai donc choisi ces 5 distributions et j'ai fait un prototype de test 64 bits: - Archlinux pour l'interface homme-machine et la virtualisation - Gentoo pour la compilation - Grml pour les entr?es-sorties - Open-Suse pour les applications - Debian pour les applications 32 bits et la mod?lisation L'id?e sous-jacente ? ce projet, c'est un module noyau ? d?velopper qui permettra de router les appels syst?me vers chaque distribution en fonction du type d'appel syst?me et du r?le de chaque distribution dans le syst?me. Le pattern MVCA (Mod?le= donn?es, Vue=ihm, Controleur=noyau, Action=application) est utilis? entre Archlinux, Grml et OpenSuse. Le filtrage de contenu s'effectue entre les applications et les vues gr?ce au module noyau qui intercepte les donn?es ? filtrer et les dirige vers un filtreur hors noyau. Le filtreur applique des outils sur les donn?es pour d?cider si elles seront autoris?es ou non. Pour d?velopper le filtreur, un nouveau langage, Pmal, est ? cr?er. On a besoin d'un nouveau langage qui doit ?tre capable de faire ?voluer ses programmes sans avoir ? les interrompre. Ce langage sera bas? sur du typage fort, pour assurer la s?curit? du syst?me. J'ai imagin? un langage de glue pour faire ?a. Il s'agit d'un langage qui fait un usage important de la r?flexivit?: il connait ses constituants et peut agir dessus. Il est bas? sur le principe des conteneurs l?gers: les classes composantes sont assembl?es par inversion de controle. Chaque composante ne connait pas avec quel composant il sera reli? dans les conteneurs l?gers. Exemple: un objet A est reli? ? un objet de type D sur lequel il invoque une m?thode b pour imprimer. A ne sait pas de quelle classe est l'objet de typez D. De D, 2 classes h?ritent: B et C, pour 2 imprimantes diff?rentes. Chaque classe B et C poss?de un objet M?thode impl?mentant la m?thode b. Cet objet M?thode porte des attributs "tags" qui le d?crivent. Par exemple B a un tag "600dpi" et C a un tag "300dpi". Lorsque A demande a ex?cuter b, b est intercept? par le conteneur l?ger qui va filter les candidats possibles en fonction des tags. Finalement, le conteneur l?ger choisira B pour optimiser la qualit? d'impression, sans que A se soit jamais dout? de rien. On pourra aussi avoir un objet proxy qui intercepte les requ?tes b et les envoie de mani?re asynchrone. Ou encore, un compteur d'appels de m?thode incr?mental pour d?tecter les boucles infinies et les r?soudre en analysant le code invoqu? et en proposant une approximation si elles tendent vers une limite. Par exemple on pourra r?soudre le paradoxe de Z?non: Achille et la tortue font une course : Achille va 2 fois plus vite que la tortue. Chaque fois que Achille parcours la distance qui le s?pare de la tortue, celle-ci parcours la moiti? de cette distance. Ainsi une nouvelle distance plus petite les s?pare... La r?cursion est infinie. Si on compte les appels r?cursifs avec une interruption par exemple au bout de 10 it?rations, en analysant le code on peut d?duire que la limite tend vers 0 et retourner cette r?ponse comme approximation et mettre fin ? la boucle. Si il n'y a pas de limite on interrompra aussi la boucle pour ?viter un engorgement syst?me. Avec le niveau d'abstraction o? une m?thode est un objet on pour attacher cette m?thode ? un seul objet sans l'attacher ? tous les objets de sa classe, ? des fins de tests. Une fois test?e, cette m?thode pourra ?tre g?n?ralis?e ? toute la classe. A ce moment, la m?thode sera intercept?e par un proxy, puis mise en place, puis activ?e en d?bloquant le proxy, permettant ainsi une ?volution du code et des classes du programme sans interruption. De plus, bien que fortement typ?, ce langage pourra ?tre ex?cut? par du javascript, du fait de son niveau d'abstraction : les objets, les classes, les attributs et les m?thodes sont tous des objets manipulables. Actuellement, je dispose d'un prototype faisant fonctionner les 5 distributions ensemble. J'ai presque termin? le compilateur pour les m?thodes du langage en Java 1.5. Il faut d?velopper l'ensemble du langage Pmal, le module du noyau pour router les appels syst?me et aussi un gestionnaire de paquets pour l'ensemble du syst?me avec les 5 distributions. Il s'agit d'un travail ?norme, mais le proverbe dit: "La foi d?place des montagnes". Contactez-moi si vous d?sirez participer: chris21.roux(a)free.fr subject "J'ai l'inux". ------------------------------ Message: 2 Date: Fri, 27 Sep 2013 08:28:13 +0200 From: Thomas K?hler <jean-luc(a)picard.franken.de> To: grml(a)ml.grml.org Subject: Re: [Grml] J'ai l'inux nom de code Canterbury Message-ID: <20130927062813.GA15925(a)picard.franken.de> Content-Type: text/plain; charset=iso-8859-1 Hello, First thing: I did learn some French at school, but that was two decades ago. I might completely misunderstand everything here. It might even be that this whole thing is an April's fool that went wrong (in regards to the calendar, that is). Just a few random thoughts from my side follow anyway. chris21.roux(a)free.fr wrote: > Le projet "J'ai l'inux" vise ? proposer une distribution Linux > capable de filter les contenus notamment pornographiques, pour > une utilisation ? l'?cole. If I understand this correctly, you want to filter out pornographic material at school, and the target is to create a linux distribution for this application. Answer from c.roux: ******************* I want to filter pornography and other perversions, based on content analysis, so this explains why I need a new distribution: it is not possible to analyse an image content at network level. The MVC pattern allows to filter content at the moment when the application sends the image to the view, inside the whole system. I developed a multi-virtual machines system in order to have one for the application, another for the view and a third for input and output, allowing to filter at the communication level between the virtual machines. ******************* > Pour filtrer les contenus, il est n?cessaire d'intervenir ? > l'int?rieur du syst?me pour intercepter les contenus ? filtrer > (on ne peut pas intervenir au niveau des trames r?seau car ? ce > niveau les contenus ne sont pas interp?tables). If I understand this correctly, you thing you can only do this on the system itself, because you think it's not possible to intervene at the network layer. Well, actually, it is. If you connect your systems to the internet via a firewall, you can block everything you want. You can also install a transparent proxy, and your proxy may well be able to filter content. You could use junkbuster for this. > J'ai pens? ? utiliser des machines virtuelles pour d?l?guer ? > chaque machine virtuelle une portion sp?cifique du syst?me ? > assurer. Of course you can use virtual machines for what you want to achieve, but it seems unnecessary here. > Cette id?e est n?e d'un poisson d'avril (April's fool joke) qui > annon?ait le projet Canterbury, visant ? r?unir le meilleur de > 5 distributions Linux dans une nouvelle distribution. So your idea is born from an April's fool joke, but in the end, it looks like one itself. > En tant que chr?tien, je suis tr?s sensible aux poissons > d'avril qui ont pour objectif de se moquer des chr?tiens. For this sentence, I needed google's translate in order to possibly make sense of it, but then I quite don't. "As a Christian, I am very sensitive to April's fool jokes which aim to make fun of Christians." Well, as a Christian I don't see how this "5 Linux distributions now become one in the Canterbury project" aims to make fun of Christians. Furthermore, there are quite a few possible explainations on how the "April's fool jokes" emerged, but none of the ones I know particularly targets at Christians. You might to choose to explain this sentence again (in English) so that I might have a chance to really understand what you wanted to say. answer from c.roux: ******************* The origin of April's fool jokes is based on 1st of April day, close to Easter. On that day, people stick a fish (symbol of christans) on the back of christians. The joke is the following: try to make beleive false things to those people, in order to discredit christians as people who beleive everything even the resurrection of Christ. ****************** > J'ai donc choisi ces 5 distributions et j'ai fait un prototype > de test 64 bits: > - Archlinux pour l'interface homme-machine et la virtualisation > - Gentoo pour la compilation > - Grml pour les entr?es-sorties > - Open-Suse pour les applications > - Debian pour les applications 32 bits et la mod?lisation So you use 5 distributions in order to do 5 different jobs in one single project. I believe this is an April's fool joke just because it seems so stupid to select 5 different methods of administrating your systems (all those distributions use their own toolset for configuration, apart from Grml and Debian, as Grml is built on top of Debian) just in order to use all of them in one single project. answer from c.roux: ******************* This idea is defined in the April fool's joke: try to use the best of each distribution as specialized in a specific task. ******************* > L'id?e sous-jacente ? ce projet, c'est un module noyau ? > d?velopper qui permettra de router les appels syst?me vers > chaque distribution en fonction du type d'appel syst?me et du > r?le de chaque distribution dans le syst?me. If I understand this correctly, you want to route systems calls to each distribution dependend on the role of the distribution in your project. Somehow, that doesn't make any sense to me. answer from c.roux: ******************* Can you explain why? ******************* > Le pattern MVCA (Mod?le= donn?es, Vue=ihm, Controleur=noyau, > Action=application) est utilis? entre Archlinux, Grml et > OpenSuse. > > Le filtrage de contenu s'effectue entre les applications et les > vues gr?ce au module noyau qui intercepte les donn?es ? filtrer > et les dirige vers un filtreur hors noyau. Le filtreur applique > des outils sur les donn?es pour d?cider si elles seront > autoris?es ou non. Seems you also want to create a kernel module to decide what to intercept when. I don't see how this should be necessary. answer from c.roux: ******************* Can you develop. Any save of time is interesting me. ******************* > Pour d?velopper le filtreur, un nouveau langage, Pmal, est ? > cr?er. On a besoin d'un nouveau langage qui doit ?tre capable > de faire ?voluer ses programmes sans avoir ? les interrompre. > Ce langage sera bas? sur du typage fort, pour assurer la > s?curit? du syst?me. J'ai imagin? un langage de glue pour faire > ?a. For the filter, you seem to want to create a new language (which you call Pmal) that should be able to intercept what a program does without actually disturbing the program. Sounds like "let's reinvent something like strace, but add the possibility to change the system calls' return values on the fly". That's actually something every virtualisation does in some way or another, but not for filtering content, but for intercepting and filtering hardware access. I still think it's better to use junkbuster and friends. answer from c.roux: ******************* I am not a wizard. If I post this message, it is because I need help. ******************* [language specification ideas deleted] > De plus, bien que fortement typ?, ce langage pourra ?tre > ex?cut? par du javascript, If you want to run a language in javascript, you usually aim at running the stuff on the browser end. You should be aware, however, that at the browser end, the USER is in control and can modify your stuff however HE sees fit, especially circumventing any blocking or filtering that happens on that level. answer from c.roux: ******************* Can javascript be used outside a browser? ******************* In the end of the day, I still think I didn't understand what you really want to achieve (no wonder, as I didn't use French in 20 years), but if I understood any of your intentions correctly, I just would think twice if there isn't an easier way to achieve your goals. answer from c.roux: ******************* Thanks for your advices, and I hope to convince you to join the project. ******************* Bye, Thomas -- Thomas K?hler Email: jean-luc(a)picard.franken.de <>< WWW: http://gott-gehabt.de IRC: tkoehler Freenode: thkoehler PGP public key available from Homepage! ------------------------------ _______________________________________________ Grml mailing list - Grml(a)ml.grml.org http://ml.grml.org/mailman/listinfo/grml join #grml on irc.freenode.org grml-devel-blog: http://grml.supersized.org/ End of Grml Digest, Vol 95, Issue 9 ***********************************

1 0

using persistence
by John G. Heim 22 Okt '13

22 Okt '13

I can't find any documentation on what persistence can do for me. What can I do with it? What I need to do is create a flash drive that runs a script upon boot. Can I do that? I created a flash drive with persistence but now I'm not sure how to use it. - --- John G. Heim, 608-263-4189, jheim(a)math.wisc.edu

2 2

LANG variable, locales
by adrelanos 15 Okt '13

15 Okt '13

Hi! Would you agree, that a system created by grml-debootstrap should be as close to stock Debian as possible? - In a default Wheezy installation, the LANG variable is set to en_US.UTF-8. - In system created by grml-debootstrap. the LANG variable is empty. Also the output of `locale` differs. [1] Am I missing an option in grml-debootstrap? Should this be added to grml-debootstrap? What is the recommended way (from a Debian derivative perspective) changing the default of `locale`? Cheers, adrelanos [1] Default Wheezy: LANG="en_US.UTF-8" LANGUAGE= LC_CTYPE="en_US.UTF-8" LC_NUMERIC="en_US.UTF-8" LC_TIME="en_US.UTF-8" LC_COLLATE="en_US.UTF-8" LC_MONETARY="en_US.UTF-8" LC_MESSAGES="en_US.UTF-8" LC_PAPER="en_US.UTF-8" LC_NAME="en_US.UTF-8" LC_ADDRESS="en_US.UTF-8" LC_TELEPHONE="en_US.UTF-8" LC_MEASUREMENT="en_US.UTF-8" LC_IDENTIFICATION="en_US.UTF-8" LC_ALL= grml-debootstrap: LANG= LANGUAGE= LC_CTYPE="POSIX" LC_NUMERIC="POSIX" LC_TIME="POSIX" LC_COLLATE="POSIX" LC_MONETARY="POSIX" LC_MESSAGES="POSIX" LC_PAPER="POSIX" LC_NAME="POSIX" LC_ADDRESS="POSIX" LC_TELEPHONE="POSIX" LC_MEASUREMENT="POSIX" LC_IDENTIFICATION="POSIX" LC_ALL=

2 1

(grml-)debootstrap network timeout / retry settings?
by adrelanos 09 Okt '13

09 Okt '13

Hi, is it possible to add apt-like options to (grml-)debootstrap to get higher timeouts / more retries when network is slow / flaky? Options comparable to "--option Acquire::http::Timeout=180" or "--option Acquire::Retries=3"? Cheers, adrelanos

2 3

Grml - new stable release 2013.09 available
by Michael Prokop 08 Okt '13

08 Okt '13

Hi, we just released Grml 2013.09 - Hefeknuddler. This Grml release provides fresh software packages after the Debian stable release (AKA wheezy) has been released. As usual it also incorporates up2date hardware support and fixes known bugs from the previous Grml release. More information is available in the release notes of Grml 2013.09: http://grml.org/changelogs/README-grml-2013.09/ Grab the latest Grml ISO(s) and spread the word! http://grml.org/download/ Thanks everyone and happy grml-ing, -mika-

4 6

[HOWTO] Packaging up out of tree Intel network drivers and adding them to grml
by Andras Korn 06 Okt '13

06 Okt '13

Hi, I packaged a few of Intel's out of tree NIC drivers from http://sourceforge.net/projects/e1000/ as dkms based .debs and included them in a custom grml image. I'll explain how; if there is interest, I can also share my packages (although, as you'll see, they're likely sub-par, because I did not take the time to really understand what I was doing; in particular, I fear the e1000e module may not always build correctly at grml-live time, while it builds correctly on the live system). 1. I grabbed some tarballs from http://sourceforge.net/projects/e1000/: -rw-r--r-- 1 root root 269997 Sep 5 16:25 e1000e-2.5.4.tar.gz -rw-r--r-- 1 root root 303648 Sep 17 02:23 igb-5.0.6.tar.gz -rw-r--r-- 1 root root 362747 Sep 30 23:21 ixgbe-3.18.7.tar.gz -rw-r--r-- 1 root root 140096 Sep 30 23:31 ixgbevf-2.11.3.tar.gz 2. I extracted each. 3. I moved the contents of foo-1.2.3/src to foo-1.2.3 itself in each. 4. I created a dkms.conf like the following in the root of every driver root directory (i.e. foo-1.2.3/): AUTOINSTALL="yes" BUILT_MODULE_NAME[0]="igb" CLEAN="make clean" DEST_MODULE_LOCATION="/updates" DEST_MODULE_NAME[0]="igb" PACKAGE_NAME="igb" PACKAGE_VERSION="5.0.6" 5. I made sure the Makefiles get the kernel version to build for from the KERNELRELEASE ennvar (not BUILD_KERNEL or similar), because this is what dkms sets: sed -i 's/BUILD_KERNEL/KERNELRELEASE/g' Makefile (In at least one case, I also replaced an #ifdef CONFIG_PM_SLEEP with #ifdef CONFIG_PM in netdev.c because the module didn't build for my kernel otherwise, but this problem doesn't manifest with the grml kernel because it defines either both or neither.) 6. I issued the following dkms commands (from the source directories): dkms add -m foo -v 1.2.3 dkms mkdeb -m foo -v 1.2.3 --source-only (Naturally, you have to replace 'foo' with the name of the module and 1.2.3 with the actual version number.) The second command will create a .deb and dump it somewhere under /var/lib/dkms/. 7. I imported the .debs into my private repository (managed by reprepro). 8. I installed grml-live. I created a new fai 'class', called EXTRANIC. This involved creating the following files: /etc/grml/fai/config/hooks/instsoft.EXTRANIC (mode 755): #!/bin/bash set -u set -e . "$GRML_LIVE_CONFIG" $ROOTCMD apt-get --yes --force-yes install \ dkms build-essential linux-headers-3.10-1-grml-amd64 (You'll note this is inelegant because it hardcodes the kernel version; the point here is to make sure the kernel-headers are already installed by the time the driver packages are installed.) /etc/grml/fai/config/scripts/EXTRANIC/50-dkms (mode 755): #!/bin/zsh set -u set -e . "$GRML_LIVE_CONFIG" for kernelversion in $(${=ROOTCMD} sh -c 'ls -1 /boot/vmlinuz-*' \ | sed 's@.*boot/vmlinuz-@@'); do echo "Building dkms modules for kernel $kernelversion." ${=ROOTCMD} /etc/kernel/postinst.d/dkms "$kernelversion" done (This may not be necessary. It attempts to build all dkms modules once more, after all requested packages have been installed.) /etc/grml/fai/config/package_config/EXTRANIC: e1000e-dkms igb-dkms ixgbe-dkms ixgbevf-dkms broadcom-sta-dkms r8168-dkms (Yeah, so not just Intel. The others come from regular Debian.) /etc/grml/fai/config/files/etc/apt/sources.list.d/myrepo.list/EXTRANIC: (Contains the deb http://... line for my repository.) /etc/grml/fai/config/package_config/ZFSONLINUX.asc: (Contains the public key of the repository signing key; I'm not sure this is really the correct way of injecting it in the build process.) 9. In /etc/grml/grml-live.local, I have: OUTPUT=/tmp/grml/grml-live SUITE="sid" FAI_DEBOOTSTRAP="sid http://cdn.debian.net/debian" VERSION=$(date +%F) RELEASENAME="grml64-extranic" GRML_NAME="grml64-extranic" HOSTNAME="grml-extranic" ARCH="amd64" DISTRI_NAME="grml-extranic" NO_ADDONS_BSD4GRML='1' EXIT_ON_MISSING_PACKAGES='1' DEFAULT_BOOTOPTIONS="ssh=sekrit" # Be sure you understand what this does before using it 10. I built a new iso using the following command: echo 'y\nn\ny\ny'\ | env -u TMPDIR -u TMP \ nice -n 20 \ grml-live -A -u -e /path/to/grml64-full_sid_20131001.iso \ -c DEBORPHAN,GRMLBASE,RELEASE,AMD64,IGNORE,EXTRANIC I'm posting this in the hope someone finds it useful. I'm not subscribed to the list, so please Cc: me on replies. The resulting image is larger than regular grml as it includes build-essential and its dependencies; if that's a problem, I suppose you can add a script after 50-dkms that purges now-unneeded packages. Incidentally, zfsonlinux can be added to grml in much the same way (that was the original point of the exercise for me). AndrÃ¡s -- At some point in the project you're going to have to break down and finally define the problem.

1 2

First Release Candidate of Grml version 2013.09 released
by Michael Prokop 27 Sep '13

27 Sep '13

Hi, so we did it again, the first release candidate of the upcoming Grml version 2013.09, code-named 'Hefeknuddler' was released. This Grml release provides fresh software packages after the Debian stable release (AKA wheezy) has been released. As usual it also incorporates up2date hardware support and fixes known bugs from the previous Grml release. For detailed information about the changes between 2013.02 and 2013.09 have a look at the official release announcement: http://grml.org/changelogs/README-grml-2013.09-rc1/ Please test the ISOs and everything you usually use and report back so we can complete the stable release soon. If no major problems come up, the next iteration will be the stable release, which is scheduled for the end of September. regards, -mika- - for the Grml.org project

3 8