Posts Tagged ‘Open Source’

Still working on MondoRescue 3.2 to make it available ASAP

2014/03/14

Even if stuff do not progrees at the speed I’d like them to progress (lots of travels on HP side since early 2014 and 4 concerts to perform on the private side) I’ve tried to improve the 3.2 version I published unofficially as beta. Interestingly enough, even when I do not announce that packages are delivered, there are people who do use them !!! Which gave me some feedback (you can guess it wasn’t that positive), so in fact it’s already my second delivery :-) and it contains some interesting new features:

  • mindi now uses the new mr-kernel-get-modules perl script which allows now in mindi to just mention end modules names andd not dependencies anymore, which are now computed by the script !! This will help a lot to maintain the list of modules, which was always impacted by low-level dependencies changes at kernel level.
  • Support of symlinks for newest distributions based on systemd such as Fedora, Mageia, … is now finally working !! Again this was done exporting the existing wrong code into a separate perl script which now operates correclty. This is part of the global willingness to recode most of mindi and some of mondo in perl. This took quite a long time, as of course, we need to stay compatible (a word systemd team doesn’t care about of course) with other tools, and older distributions. Side note, this is probably one of the reason MondoRescue is still appreciated by its community :-)
  • The introduction of a dependency on a perl function was incorrect and people trying to install from packages gave feedback that they had errors dof course ue to that. This is now fixed, as project-builder.org indeed had an issue because a low level function was depending on a higher level function not part of the perl modules provided for MondoRescue. With 0.12.5 of project-builder.org this is completely solved.
  • Now I still have regressions with the isolinux menus, NFS on Mageia 4 and systemd not working anymore (change of network NIC name is the root cause). However the ldlinux.c32 issue for syslinux > 5.x is now solved.

Next week is the TES, so won’t have much time to work on it. Expect news the week after.

6th Technical Excellence Symposium (TES) in Grenoble 21-27 March 2014

2014/02/20

If you are an HP employee, or a Value Added Reseller (VAR), then I’d like to invite you to come to an Event I’m co-organizing:

TES logo

High Performance Computing and Open Source Linux
Technical Excellence Symposium
17 – 21 March 2014
Grenoble, France

TES fond

The goal is to gain expertise and increase your sales and earnings potential.

You are invited to the 2014 High Performance Computing and Open Source & Linux Technical Excellence Symposium. This highly regarded technological event dedicated to professionals is the unique opportunity to boost your skills and knowledge in just 4.5 days!!

Why should you attend the EMEA HPC & OSL event?

  • Receive an update on the entire HPC and Open Source & Linux solutions including Enterprise Storage, Servers and Networking (EG) portfolio.
  • Reinforce your knowledge in this focus area.
  • Connect with the HP management and peers in the HP Certified Professional Community.

Do not miss this exclusive opportunity to give your career a boost and be part of the HPC and OSL community. This is also an excellent opportunity to get the latest technology updates even if you participated in this event before.

Pre-register today to secure your seat by clicking on the following link: http://hp.inviteo.fr/HPC_OSL_TES2014/

Choose among over 100 technological and solution sessions, a mix of hands-on labs, breakout demos and round tables.

This technical event covers the HP’s broad enterprise products and solutions in an open source environment of high performance computing, web services portfolio and new product introductions (e.g. Moonshot program and big data solutions).

Along the sessions, you will meet the HPC and OSL experts to share knowledge and insights in many key business areas including competitive tools.

Back home and freshly armed with an enhanced skills set, you will be ready to provide a great business value to your customers.

Throughout the event, we will offer detailed tracks on High Performance Computing solutions portfolio using our HP Blade System, ProLiant Scalable servers, Storage Works and Network products and solutions using Open Source & Linux software products and services.

We will have a huge emphasis on topics like Hyperscale products, smart storage, purpose-built servers for big data, Moonshot program wave 2 solutions as well as open cloud solutions.

Our High-level agenda is available. The detailed one will be sent to you after your registration.

In addition, there will be:

  • Sponsor sessions delivered by top names in the industry
  • Presentations by HPC and Open Source experts
  • Evening events

For detailed program information, please visit the event website

Any question? We invite you to contact us at: TES-event_at_hp.com

We are looking forward to meeting you at the 2014 HPC and OSL Technical Symposium!

project-builder.org 0.12.5 is now available

2014/02/07

Finally, after having published the version 0.12.3 on time for Linux.conf.au, I found some remaining bugs in it (problem of time based realease and not enough time to test), then I made a 0.12.4 version soon after as promised to solve these, but still missed some aspects, and those are now solved with this 0.12.5 version.

So this version fixes an issue with messages printed which were done in double, and also the way we manage naming of Debian versions, following the recent change for Debian 7 (using now only the major version, and not mentioning the minor version which is now useless).

I also fixed a dependency issue, seen when trying to install the beta of MondoRescue, where a function wasn’t provided in the perl-ProjectBuilder module, while used in the code :-( This is now solved, and I’ll now start working on the next MondoRescue beta which should now be installable with yum/apt-get/urpmi.

This project-builder.org version should now be stable to use for some weeks, as I do not plan to make big changes in it, as this is perfectly fitting my needs now. Of course, more could always been done, but I can’t advance on all sides at the same time, so now it’s MondoRescue‘s turn !

Enjoy !

Mageia 4 on time for Fosdem but …

2014/02/02

Anne has announced that Mageia 4 is now available just on time for Fosdem !

But I won’t be at Fosdem this year again (I’m attending an HP event instead)

And while there are good reasons for Mageia to be my distribution of choice, I won’t update my laptop this week, as I’m presenting during this HP event, and want to stay on the safe side. And for my home computer, well, I generally do it after my laptop ;-)

But anyway, great job done by the Mageia team and lots of good apps in this new version, including OpenStack and UEFI ! Enjoy and try it. It’s really worth it !

Gouvernance informatique: Il est temps d’y intégrer l’Open Source

2014/01/24

Dans le cadre de mes activités pour le Conseil des technologistes d’HP France, j’ai écrit un article pour le Webzine IT experts sur la l’intégration de Open Source et la gouvernance informatique disponible sur http://www.it-expertise.com/gouvernance-informatique-il-faut-integrer-lopen-source/. Un grand merci à Aurélie Magniez pour m’avoir aidé à faire cette publication.

Ci-dessous, une version légèrement modifiée qui tient compte de retours et rétablit certaines formules auxquelles je tiens, quoique moins journalistiquement correctes et certains liens (jugés trop nombreux par le Webzine, mais je tiens à citer mes sources, et Tim Berners-Lee ne les a pas inventés pour que l’on ne s’en serve pas non ? :-))

Bonne lecture !

Aujourd’hui en 2013, toutes les entités, publiques comme privées, en France, comme partout dans le monde, utilisent massivement des Logiciels Free, Libres et Open Source (abrégé en FLOSS (1)). Quelques exemples de cet état de fait sont fournis par la Linux Foundation, comme les 600 000 télévisions intelligentes vendues quotidiennement fonctionnant sous Linux ou les 1,3 millions de téléphones Andoïd activés chaque jour. Le dernier rapport de top500.org, présentant les super-calculateurs mondiaux, indique une utilisation de Linux à 96,4%. Des sociétés ayant aujourd’hui un impact quotidien sur notre environnement numérique telles que FaceBook ou Twitter ont non seulement bâti leur infrastructure sur une grande variété de FLOSS, mais ont aussi publié de grandes quantités de code et des projets complets sous licence libre. Ceci concerne aussi des acteurs plus classiques du monde de l’informatique comme HP ou IBM.

Ceci peut sembler normal, car on évolue là dans le monde du numérique, mais le phénomène touche tous les secteurs comme le montre une récente étude de l’INSEE, qui reporte que 43% des entreprises françaises d’au moins 10 personnes utilisent des suites bureautique FLOSS ou encore que 15% des sociétés de construction utilisent un système d’exploitation FLOSS par exemple. Cette large adoption se trouve corroborée par le développement de la filière FLOSS en France, comme rapporté par le CNLL, représentant en 2013 2,5 milliard d’Euros et 30 000 emplois.

Enfin, le secteur public n’est pas en reste avec la publication en septembre 2012 de la circulaire du premier ministre qui reconnait la longue pratique de l’administration des FLOSS, et incite celle-ci, à tous les niveaux, à un “bon usage du logiciel libre”, ce qui se vérifie dans certains ministères comme celui de l’intérieur ou de l’économie. Le ministère de l’éducation nationale a ainsi déployé 23 000 serveurs EOLE sous Linux et utilise de nombreux projets FLOSS pour la gestion multi-fonctions (réseau, sécurité, partage) des établissements scolaires.

Services impliqués dans la gouvernance FLOSS

Dans ce contexte d’utilisation généralisée, se posent certaines questions quant à la gouvernance particulière à mettre en place ou l’adaptation de celle existante pour accroître l’usage, la distribution, la contribution au FLOSS, tant pour les fournisseurs que pour les utilisateurs de ces technologies. En effet, les FLOSS ont des spécificités tant techniques qu’organisationnelles (rapport à la communauté, méthodologie de développement, licence utilisée) qui ont un impact sur la façon de les gérer dans une entité. La Gouvernance Open Source, aujourd’hui, doit donc être partie intégrante d’une Gouvernance Informatique.

Contrairement à ce qu’une rapide analyse pourrait laisser penser, ce n’est pas uniquement le service informatique qui est concerné par l’utilisation des FLOSS. Celle-ci touche la totalité de l’entité et le modèle de gouvernance doit donc être adapté en conséquence. En effet, le service des achats se voit souvent court-circuité par l’utilisation de composants logiciels téléchargés et non achetés en suivant les procédures qu’il met en place, le service du personnel ne dispose pas de contrats de travail statuant sur les contributions des employés à des projets FLOSS (ne parlons pas des stagiaires ou co-traitants), le service juridique doit apprendre à distinguer la licence Apache de la GPLv2, ou v3, le service de propriété intellectuelle considérer si telle modification faite à un projet FLOSS peut ou doit être reversée au projet, et dans quel contexte, voire le PDG évaluer, lors d’une scission de sa société en différentes entitées juridiques, l’impact représenté sur la redistribution de logiciels faite à cette occasion et le respect des licences utilisées. Ce ne sont que quelques exemples des questions auxquelles les entités doivent répondre dans le cadre d’une Gouvernance Informatique intégrant les FLOSS.

Ceci n’est pas un débat oiseux: il y a eu maintenant trop d’exemples allant jusqu’au procès et sur des problématiques de non-respect des licences FLOSS pour que les entreprises et services publics ignorent le problème. Les conséquences tant financières que sur leur image de marque peuvent être très importantes et causer des dommages beaucoup plus graves que ne le représente la mise en conformité (qui consiste le plus souvent en la seule publications des codes sources modifiés).

Il ne s’agit pas ici d’énoncer des éléments qui tendraient à restreindre l’utilisation des FLOSS dans une entité. Au contraire, les bénéfices de leur utilisation sont aujourd’hui trop évidents, la baisse des coûts induite par la mutualisation, les gains technologiques d’avoir des souches logicielles si versatiles et éprouvées doivent juste s’accompagner des mesures de gestion nécessaires pour en retirer tous les bénéfices annoncés. L’analyse des risques fait partie des choix quotidiens exercés au sein d’une entité et de même que pour une démarche qualité, l’impulsion doit venir du sommet de la hiérarchie de l’entité. Celle-ci doit soutenir la création des instances nécessaires à l’établissement d’une gouvernance FLOSS en leur donnant le pouvoir requis et l’interaction avec les différents services de l’entité.

Composants d’une gouvernance FLOSS

Tout d’abord, il s’agira de développer la compréhension de l’écosystème libre au sein de l’entité pour en appréhender les spécificités.

La première d’entre elles est la licence gouvernant les FLOSS. Comme pour toute utilisation d’un logiciel, ou d’un service, un utilisateur se voit décrit ses droits et ses devoirs au sein de ce document. Ceux-ci diffèrent selon que la licence est permissive (type Apache v2 par exemple), qui permet une utilisation (y compris pour des développement non-FLOSS) et une redistribution avec peu de contraintes (mentions légales et paternité par exemple). Elle permet ainsi à des sociétés de vendre des versions propriétaires d’Andoïd distribué sous Licence Apache v2 embarquées dans leurs téléphones portables. C’est ce qui permet de considérer cette licence comme “libre”. En regard on donnera également l’exemple des licences de gauche d’auteur (copyleft en anglais, type GPL v2 par exemple), qui permettent une utilisation tant que le logiciel distribué s’accompagne des sources (éventuellement modifiées) servant à le fabriquer. Elle permet à des projets comme le noyau Linux d’être développé par des milliers de développeurs tout en restant toujours accessible dans toutes ses variantes par la mise à disposition de son code source, dû à cette contrainte. C’est ce qui permet de considérer cette licence comme “libre”. Simplement les libertés sont vues ici sous l’angle du projet (qui le reste ad vitam aeternam) plutôt que sous celui de l’utilisateur comme dans l’autre cas. C’est la raison pour laquelle toutes ces licences sont considérées comme Open Source par l’OSI.

Une entité doit donc choisir les briques FLOSS qu’elle souhaite utiliser en fonction de l’usage prévu pour respecter les droits et devoirs d’usage codifiés dans les licences (ni plus ni moins qu’avec une offre non-FLOSS), sachant que, dans la plupart des cas, l’élément déclenchant l’application de la licence est la distribution du logiciel. Ainsi une société peut parfaitement utiliser un logiciel sous licence GPL v2, y faire des modifications et ne pas les publier, tant que l’usage reste interne à sa structure juridique (cas fréquent en mode utilisation de logiciel dans un département informatique). En revanche, si elle l’incorpore à un produit qu’elle commercialise, elle devra juste se mettre en conformité avec la licence et fournir en parallèle du produit un acccès aux dites sources.

Ceci n’est finalement pas si compliqué, eu égard aux gains énormes qu’elle peut en retirer en bénéficiant d’une brique logicielle éprouvée qu’elle n’a ni à développer, ni à maintenir. Dans tous les cas, il est important que son service juridique ait une compréhension des droits et devoirs des licences utilisées pour apporter le conseil requis, comme lors de la signature de contrats avec tout fournisseur.

On le voit, la formation du service juridique est à la base de la mise en place de toute gouvernance. D’autre part, il faut organiser au sein de l’entité la mise en relation entre ce service juridique et les équipes de développement. Non seulement pour qu’elles apprennent à se connaître, mais aussi pour qu’elles échangent sur leurs besoins réciproques et qu’elles comprennent comment chacune cherche à protéger l’entité pour laquelle elle oeuvre. Les uns le faisant eu égard au respect des règles de droit, ce qui comprend l’explication envers les développeurs des licences libres, les autres eu égard au mode d’utilisation des composants techniques spécifiques des équipes de développement.

Personnellement, en tant qu’ingénieur de formation, il m’a été très bénéfique de discuter avec divers avocats spécialistes des licences libres, pour mieux comprendre leur volonté de protéger l’entreprise pour laquelle ils travaillent et comment ils devaient le faire dans ce contexte. Et réciproquement, je sais que les informations techniques et exemples parfois complexes d’agrégats de composants logiciels les aident en retour à mieux tenir compte des cas particuliers qui peuvent se faire jour. La communication sur ce sujet doit dépasser dans l’entité les structures classiques et fonctionner comme une communauté.

Du reste, la seconde spécificité du logiciel libre est le fait qu’il est développé par une communauté de personnes partageant un intérêt pour ce logiciel. Il en existe de toute taille (d’un développeur assurant tout, jusqu’à plusieurs centaines de personnes comme les larges fondations comme Apache ou OpenStack). Etudier une communauté avant d’utiliser le composant libre qu’elle produit est une bonne pratique pour avoir des informations sur sa vitalité, son organisation, sa feuille de route, en plus des caractéristiques purement techniques du composant. Certains sites comme Ohloh peuvent aider à se forger une opinion dans ce domaine, pour les projets suivis. De même qu’il peut être alors pertinent de se poser la question des modes de contributions en retour. Cela peut consister en des correctifs, du code apportant de nouvelles fonctions, de la documentation, des traductions, une animation de communauté, de l’achat de prestation intellectuelle auprès de professionnels oeuvrant sur le composant ou un soutien financier à l’organisation d’un événement permettant le rassemblement physique de la communauté. Certaines entreprises, comme la Compagnie Nationale des Commissaires aux Comptes témoignent de leurs contributions en retour envers un projet tel que LibreOffice.

Comme précédemment, chacun de ces aspects pourra faire l’objet d’une étude dans le volet Open Source de la Gouvernance Informatique. On notera que la gestion de la proprété intellectuelle sera à considérer tout particulièrement pour les contributions sous forme de code, et en liaison avec la licence utilisée. Mais cet aspect peut aussi avoir un impact sur les contrats de travail des employés, des co-traitants, des stagiaires, afin de déterminer sous quelles conditions leurs contributions sont autorisées.

Encore une fois, il s’agit d’inciter les entités utilisatrices de logiciels libres à ne pas se contenter d’être de simples utilisatrices de FLOSS, mais à être actrices de l’écosystème et à contribuer à leur tour à l’améliorer en s’intégrant dans les communautés. Le dynamisme actuel autour des FLOSS est le fait du soutien très actif de nombreux utilisateurs. Pour ne citer qu’un exemple, on regardera la synergie créée autour du projet GENIVI par ses 120+ membres, dont de nombreuses sociétés hors secteur informatique.

Enfin la dernière spécifcité du logiciel libre est la méthodologie de développement utilisée par la communauté. Quoiqu’elles soient toutes attachées à l’accès au code, elles varient énormément d’un projet à l’autre, en fonction de sa taille, de son style de gouvernance, des outils utilisés et de son historique. Mais il est important pour une entité qui souhaite interagir avec une communauté d’en comprendre la culture. Si le noyau Linux a une méthodologie organisée autour d’un “dictateur bénévole” (Linus Torvalds) qui prend les ultimes décisions et de ses lieutenants, nommés, en qui il a toute confiance pour prendre les décisions concernant une branche de développement, d’autres projets comme OpenStack cherchent à adopter le mode le plus “méridémocratique” en procédant à l’élection des représentants techniques des branches du projet par les développeurs, et à celle des représentants au conseil d’administration par la totalité des membres de la fondation, quels que soient leurs rôles. Le processus d’intégration continue d’OpenStack implique des étapes précises pour y ajouter un patch par exemple. Cela nécessite d’abord une application sur l’arbre courant sans erreur, avant de devoir recevoir deux votes positifs puis de satisfaire le passage de l’ensemble des tests automatiques prévus. Et ceci s’applique aussi bien aux représentants techniques des branches du projet qui proposent des centaines de patches par an, ou au contributeur occasionnel faisant une modification mineure de documentation. En revanche, celui qui souhaite soumettre une modification sur le noyau Linux devra passer par des listes de diffusion où les échanges peuvent parfois se révéler vifs, et s’adapter aux desiderata potentiellement différents des mainteneurs de branches.

Bonnes pratiques de gouvernance FLOSS

Face à tous ces aspects de ce monde foisonnant, certaines bonnes pratiques simples peuvent permettre aux entreprises de faire les bons choix et de s’assurer une utilisation optimale des FLOSS en en tirant le meilleur profit sans mettre à risque leur bonne réputation par des actions mal vues des communautés.

Une première bonne pratique peut consister à créer un comité Open Source. Par exemple, pour un grand groupe, il peut être utile pour la direction générale de nommer des représentants des différents services (achats, ressources humaines, informatique, technique, juridique, propriété intellectuelle) pour définir la politique à mettre en place. Ce comité devra se réunir régulièrement, tant dans la phase de définition de la partie Open Source de la Gouvernance Informatique, qu’ultérieurement pour la réviser sur la base des retours des utilisateurs et l’évolution de projets. Il devra également avoir les moyens associés à ses missions. Un groupe de travail du Syntec Numérique a développé, pour les aider dans cette activité, des contrats types pour leurs fournisseurs, leur demandant de préciser avec leur livraison logicielle, l’inventaire exhaustif des licences utilisées. Une présentation sur les contrats faite au sein de ce groupe pourra être aussi consultée avec profit. La FSF France propose aussi des avenants de contrats de travail type pour les employés contribuant à des projets libres, et l’AFUL des modèles économiques et financement de projets FLOSS ou de communautés. Il sera ensuite facile de donner des missions et des pouvoirs plus étendus à ce groupe de personnes quand l’utilisation des FLOSS augmente. Dans le cadre d’une PME, un correspondant FLOSS sera sans doute suffisant (comme il peut y avoir un correspondant sécurité ou CNIL), tâche qui pourra même être sous-traitée à des sociétés specialisées dans le domaine.

Une fois le comité/correspondant nommé et la politique FLOSS établie, il faudra prévoir des cycles de formations. D’une part pour le service juridique pour le cas où il manquerait de compétences sur le domaine spécifique des licences libres. La société Alterway propose par exemple une formation par un juriste pour des juristes. D’autre part, en interne, auprès de l’ensemble du personnel pour expliquer cette nouvelle politique FLOSS.

En parallèle, il est important d’avoir une vision précise de l’utilisation actuelle des FLOSS dans son entité. Notamment pour vérifier que leur utilisation est conforme aux licences sous lesquelles ils sont utilisés. Les non-conformités sont plus souvent dûes à la méconnaissance qu’à une réelle volonté d’enfreindre les licences. Cette tâche peut paraître fastidieuse de prime abord, mais elle est à mon sens fondamentale pour se prémunir, en particulier si votre activité vous amène à redistribuer du logiciel à vos clients. Heureusement des outils existent pour automatiser ce travail d’inventaire et faciliter l’analyse des licences utilisées. Le premier à recommander est libre: FOSSology a été développé par HP pour son utilisation interne, puis rendu libre en 2007 sous licence GPLv2. Il collecte dans une base de données toutes les meta-données associées aux logiciels analyés (il peut traiter des distributions Linux entières sans problème) et permet l’analyse des licences réellement trouvées dans le code depuis une interface Web. De nombreuses entités outre HP comme Alcatel-Lucent, l’INRIA ou OW2 l’utilisent, y compris pour certains, en couplage avec leurs forges de développement. Mais son accès libre et sa facilité de mise en oeuvre ne le réserve pas qu’aux grands groupes et il devrait être systématiquement utilisé comme complément naturel d’un gestionnaire de source, ou d’outillage d’intégration continue. En complément, des outils non-FLOSS peuvent également aider à ce travail d’inventaire en donnant accès à des bases préétablies de composants connus et déjà inventoriés et fournissent de nombreuses autres fonctions. La société française Antelink, émanation de l’INRIA, a développé une grande expertise dans ce domaine et a couplé son outillage avec FOSSology. D’autres acteurs tels que Blackduck et Palamida ont également un outillage complémentaire à considérer.

On pourra de plus prévoir ultérieurement un mode de déclaration des usages de FLOSS, voire, si les requêtes sont nombreuses et régulières, créer un comité de revue spécifique en charge de les évaluer et de les approuver.

Enfin certains documents de référence tel que le Guide Open Source du Syntec Numérique, les fondamentaux de la Gouvernance des logiciels libres, la vision des grandes entreprises sur la gouvernance et maturité de l’Open Source et le site de référence FOSSBazaar pourront permettre un approfondissement des sujets évoqués et donner des bonnes pratiques additionnelles quant à la mise en oeuvre d’une gouvernance Open Source.

Et pour ceux qui souhaiteraient être accompagnés dans la démarche, des sociétés telles que Smile, Alterway, Linagora, Atos, Inno3 ou HP disposent de prestations d’aide à la mise en oeuvre d’une gouvernance Open Source. Mais que vous le fassiez seuls ou accompagnés, il est temps et j’espère que cet article vous aura donné quelques clefs pour intégrer l’Open Source dans votre politique de Gouvernance Informatique.

(1): Dans tout ce document, on utilise le terme de FLOSS comme terme générique recouvrant aussi bien la notion de « logiciel libre », « Free Software » qu’« Open Source », tout en sachant que des nuances existent.

Meet at HP ExpertOne Technology & Solutions Summit

2014/01/22

For those of you working for an HP value added reseller (VAR) and attending the HP ExpertOne Technology & Solutions Summit on Barcelona from the 3rd to the 7th of February, feel free to meet during one of the session I’ll (co-)deliver:

For those of you wondering why project-builder.org and MondoRescue versions are not published more regularly, that gives you a hint ;-)

See you there to talk of these subjects or something else as you see fit. I’m around the whole week.

Last Day at Linux.conf.au

2014/01/21

Those interested by the previous part of the story can read the previous article on LCA14.

Time goes quick ! It was already time to pack and prepare to fly back. But before, a full day of interesting sessions was ahead of me.

Keynote at LCA

And the first one, the keynote, was particularly impressive. EVen if I must confess that space never really made me dream that much (sorry dale !), I was pretty impressed by the work done to send 15cm3 satellites, that Jonathan Oxer called ArduSat (in reference to the Arduino project), and the willingness to open source everything.

Jonathan Oxer with an ArduSat

Jonathan explained how he worked with the NASA in order to be able to launch his stellites alongside a more “normal” one, in order to benefit from an existing launch. I think that his idea to make them accessible to schools so they can create their own usage of such “tool”. He also underlined that satellite are still considered as a weapon in the USA and thus it wasn’t possible for him to disclose all what he would have liked to do. He mentioned that he is working with Bdale Garbee on trying to relax this.

ArduSat goal

I then attended a session on “Building APIs Developers Love”. While the presentation gave good overall recommendations, I’d have expect a bit more concrete examples, mention of tools to help you doing so, and a bit more code examples.

Darcy Laycock

After this, I was again more interested to follow closely what happens on the OpenStack side, so I followed the 3 sessions around that topic which were following.

The first was from Robert Collins, HP around Diskimage-builder. I was curious to see the commonalities with what I do around MondoRescue. But in fact the tool is really focussing on an OpenStack environment (as it should !) and they don’t need to manage physical setup, as they target virtual machines (IIUC). Also they really build the target image, not analyze an existing content to create an image from it, so the architecture i pretty different. For example when looking at their way to get the kernel needed, they use a qcow2 container in which they look for the kernel they want. On my side, I look in the local file system to do the same. So this shows that we could probably merge some techniques used, but it would take time and need some work. Maybe I can take that in account when refactoring mindi’s code, as I started to do with the new 3.0 version.
Anyway, again another tool to put on the TODO list and to look at.
Robert Collins

Next one was about Log content analysis in the continuous integration process of OpenStack by Clark Boylan, HP. Clark explained the tools they used and integrated to support this activity (Gearman, Logstash, and ElasticSearch), and as usual in the open. In particular, he showed how to extract useful information from billions of lines of log in their CI process.

Clark Boylan

Finally, the last session was on Bare metal provisioning with OpenStack by Devananda van der Veen, HP, who is the PTL for this Ironic project. Again my interest is also in relationship with MondoRescue which does similar stuff. But yet again from a different angle: deploying OpenStack with OpenStack. The project is still new, but progressing rapidly. It aimed at replacing crowbar typically. I really like the way Devananda present as I feel much clever at the end of his presentation, as this is always very clear and easy to get ! He even finished earlier his talk (which was very convenient for me as I had to leave quickly to catch my plane !). He could hav used that time for a short demo (I like demos it help me understand better !)

Devananda van der Veen

As you can see, HP’s involvement in OpenStack is really strong and not only throwing money in the project (thing that we also do !) but bringing expertise, project lead, build resources, code to it.

UWA
LCA is really a great event, focussing on various Open Source and Linux communities. It’s also one of the oldest worldwide, and the locations are always very nice, and people are just awsome there ! That’s why I’ll continue to submit proposals and hopefully be able to attend again in the future if they are retained. However, while I think that LCA’s reputation is well deserved, I think the largest event today for our communities is FOSDEM in Europe, attracting more than two thousands people each year only over a week-end however, but maybe without the same involvement of the Linux Kernel community. I won’t be there this year due to a conflict with another HP event, but I also recommend you to attend as a lot happens there as well.

LCA

Reminder, all the photos I took during the event are available at https://plus.google.com/photos/+BrunoCornec/albums/5966194867275136465

Day 4 at Linux.conf.au

2014/01/19

Those interested by the previous part of the story can read the previous article on LCA14.

This time, even if still tired by night work, I had to be on time in order to attend the keynote.

Matthew Garrett's Keynote

Matthew Garrett was talking about Security in general Secure Boot in particular (a subject he has been covering for a long time now). This was complementary to the Keynote given on Day 1. Matthew insisted on the need, in a post Snowden area, for technologies such as Secure Boot to improve Security which are more and more turned on by default. Having a trusted and verified boot chain is a critical aspect of system security, which Secure Boot provides. And this letting users replacing their keys, so leaving freedom to them (on x86). And today Linux distributions are indeed supporting pretty well Secure Boot on UEFI, as I was able to test myelf during the UEFI Plugfest we had last year. Matthew posed more questions in fact than he gave answers around security: What level of trust to have for hosters, Software or HyperVisors or or Firmware or Hardware manufacturers. Of course as he said rigtly “Imperfect security is better than no security” and he mentionned that most of the attacks are model specific, so we may still have a certain level of trust in our platforms. He also reminded that Security agencies may not be the biggest concern with regards to security.
And BTW as Matthew said “if we don’t trust our Hardware, then why are we running a computer”.

I found that keynote very relevant, and making you think once more to how you treat security and how you can do that better to protect your contents.

Matthew Garrett

After the traditional morning coffee, during which people exchange on various topics of their choices such as shoes, it was time to attend the 2 sessions of the morning.

Coffee pause

The first was on Python Packaging 2.0 by Nick Coghlan which I was curious to see in order to measure the impact on project-builder.org. Unfortunately, it went rapidly into a lot of concepts intimately linked to python, which I’m not savy enough in to follow closely and PEP proposals of evolutions I’d need to read before being able to understand stuff better. So with a doubt a good presentation, but I wasn’t the right audience ;-)

Nick Coghlan

And that was the same with the next one. Very good presentation from Katie Miller on functional languages (and in particular Elixir), but that’ very far from what I do and know. I hadn’t a better choice for that time slot, so at least enjoyed the pictures she used.

That was much better after lunch for me. The first session was full, and after having seen it I undertsood why !

Full session on mosh

mosh belongs to this category of small software that rapidly becomes mandatory in your environment. If you use ssh, and screen to manage disconnecitons, then mosh is for you. What does it do for you: well it uses ssh to login to your remote system and after that it does communicate with it using UDP on another port, and keeps the connection available and ready to use, even if you disconnect, and reconnect from another IP e.g. Typical usage when you hibernate to go back home. It also echo locally characters, and underline them when they are also received on the other side. The demo was pretty impressive from that perspective. I alwyas found scren annoying as unable to keep my keyboard shortcuts and here, mosh does the job for you. The only point is that you need an open port on your firewall to go through with the UDP dialog, but that’ a light drawback I’m ready to accept to benefit from the services it provides. Ah; and ^C is just immediate ! That was really a useful and great presentation. Thanks Jim Cheetham to make it obvious it was the tool I needed !

Jim Cheetham

The next session I chose on Petitboot, was also a good one. Again the principle is easy to understand: what if you were using Linux (the kernel) as your boot loader ? Instead of redeveloping each time driver support for each boot loader, and dealing with bugs already solved in the Linux kernel. Thats’ the simple yet powerful idea behind Petitboot. Jeremy explained clearly the advantages of his approach such as native support of a network stack and easy addition of client protocol for http, ftp, nfs… From that initial kernel acting as boot loader, the launch of the “real” kernel is done with kexec. Definitely worth exploring. So again added to my TODO list which never gets empty !

Jeremy Kerr

After the afternoon pause, it was time to go to Bdale’s session. I arrived early in the room, and that was a good choice as 5 minutes before the start the room was completely full, and they prevented more people to enter.

Full room for Bdale

He explained that when the LCA organizer contacted him for a talk in 2013, he was still sorting issues folliwng the fire disaster which occured to his house and wasn’t really thinking to a talk for LCA. But they convinced him to just talk about that, as they really wanted to have him onboard. So that was indeed not a technical talk on IT, rather a technical talk on life during and after such as disaster. That was a shoking story of course, and Bdale, despite the humour he always keep in his talks, was touched even months after the event when he told us the story. He also gave some hints in order to keep important papers outside of the house, or scanning them to keep them on a mobile computer you can get with you. Surprisingly (for me) he insisted on insurance contracts and the good choice he made without really realizing it when he was younger. Maybe insurance company explain more to us in Europe consequences of such choices when we do it. As I’m a bit paranoid on valuable data, I already have scanned docs of the most important ones, and the backup copies I do regularly on a laptop nd on a server to always have many versions of the file for both backup and disaster recovery purposes. Maybe my work oN MondoRescue made me that way :-) Bdale received a lot of help from the community during this period and he concluded by saying that people were awesome and he was really meaning at the tone of his voice. Pretty moving presentation I must confess.

Bdale Garbee

I didn’t attend the last session of the day due to a call with my daughter and later I attended the Dinner for Professionals, where I passed most of the time discussing with Thomas Petazzoni again. Robert Collins was HP’s representative to introduce the session and he did it that shortly and with lots of energy !

Robert Collins

Waiting now for the last day !

Day 3 at Linux.conf.au

2014/01/14

Those interested by the previous part of the story can read the previous article on LCA14.

I really had to sleep a bit that night (5 hours is a minimum these days) so I missed the lightning talks of the morning, but it doesn’t seem I missed a major announce there. But I didn’t want to miss the start of the main conferences.

UWA

As a followup of the previous day, and in order to know better what we, HP, do for the OpenStack project, I chose to attend Elisabeth Krumbach Joseph who was describing the management of the project infrastructure. She made a good description of the tools used, including some navigation on the various Web interfaces they are using. She insisted on the fact that all the infrssctructure is managed the same way as the rest of the OpenSTack project, which means in the open, and using gated trunk as well ,which is much less usual for an infrstucture than for a project. It sounded to me as a risk of adding too much bureaucracy, and was maybe more suited for an established infra not moving too much, but they seem to deal with it without problem. Definitely an interesting approach that to have peer review of infrastructure configuration changes before applying them. Maybe my Solution Center could get some inspiration out of it. Will see when I’m back. The ~6 people are managing around 35 servers for the OpenStack infra (all virtual, so they do not deal with HW issues).
As she had time left, I think nxt time she could demo in more details some of the tools she mentioned, and give a bit more details around sizing (which I gathered from questions I asked).
Elisabeth Krumbach Joseph

As I’m always interested by build or execution environment, I then attended the presentation from Jay Coles on Linception: playing with conatiners under Linux. It took me nearly the full presentation to understand what it really was about, as I first thought it would be an explanation of LXC or something similar, but it was in fact again a new implementation of containers for Linux, using various capabilities provided by the kernel. (Thanks to the audience member who asked for a demo at the, which made it completely clear to me then !). And re-reading the description of the talk, I think the content wasn’t completely what was described. Anyway, was interesting, but do we really need again a new solution in that area, after LXC, Libvirt-lxc and systemd-spawn ? Look yourself at http://www.doger.io/ to make your own idea. I’m not completely convinced.
Jay Cole

After lunch, Mark Nottingham explained to us what HTTP 2.0 was all about. Plenty of good reasons to improve HTTP for sure, with lots of graphs demonstrating the issues. But guys, please stop making stuff so administrator unfriendly ! Compressing header may be a win for resources, but for sure a loss for us. We won’t be able to make a simple telnet host 80 and type GET / HTTP/1.0 e.g to check rapidly stuff. I don’t think that helps at the end. Of course modifications are required to solve real issues. But be smart and invent something more friendly for us.
Mark Nottingham.

Then was a conference from someone I really wanted to see, if only for his vcs-pkg.org initiative. And more over because Martin Krafft was talking about configuration management for system administration, which is something I’m also working on for my systems, even if I don’t have the final solution at the moment (ansible looking really close to what I like the most). And his presentation was in the line of what I expected: he prefers pets to cattle, and I must confess so do I ;-) He has strong opinion (he is a Debian advocate ;-)) on system management, and most of them sound pretty acurate to me. Martin developed a tool called reclass to help him encapsulate Salt and Ansible and do what he wants them to do. Definitely worth exploring. Even if it seems that latest versions of Salt have similar features, making the tool less useful. Anyway, I prefer ansible (for the SSH communication aspect), so added to my TODO list, which never gets empty !
Martin Krafft

After that talk, I could not miss the only one made by a french hacker ! Especially as I know Thomas Petazzoni for a long time now, have loved his Kernel presentations he was making in the past, and was interested to see a new tool such as BuildRoot, which could be relevant in the MondoRescue context. And it seems I wasn’t the only one as the room was pretty packed for his presentation, which is good for his first LCA.

Thomas Petazzoni

Is it chauvinism ? I found his presentation one of the best of the week. He gave a very detailed view on the tool features, giving concrete example of usage corresponding to the various configurations he was describing. But all that remaining cristal clear. The presentation was easily understood by the audience. The tool itself is pretty impressive, recompiling in order all what is needed to make a standalone Linux from scratch system ready to be embedded on any system. Of course, Thomsa knows the subject extremely well as he does that for a living for free electron, which was appearing on Jon Corbets’s stats as a major ARM kernel contributors company.

He even mentioned the help that is provided with regards to licenses of the software contained in the build, which is a topic too few developers take seriously, and it was great to see him mentioning that.

My only regret: I’d have loved to have a short demo of on of his existing system, but honestly he had no additional time to do it, so could be worth adding for a one hour time slot. Anyway, well done Thomas, and looking forward to attend more of your presentation, as I always learn stuff when I do, which makes my day :-)

https://lh5.googleusercontent.com/-clI2-vU4hKk/Us2ovNL9oNI/AAAAAAAAB3U/TBXFPTxkNjA/w466-h704-no/DSC_9956.JPG

And finally, I finished the day with another of my favorite speakers in such event, Ric Wheeler from Red Hat who was talking about File systems and storage systems of course ! But this time he was exploring the concerns that bring to the Linux Kernel community the new storage systems arriving: Flash components used on memory slots, post-flash components, with low lantency and high storage capabilities (such HP’s own memristor e.g.). But even if it creates some difficulties to the kernel hackers to keep with the pace of storage technology, these are really good news, as first storage technology will now also make a giant leap, as was done for network, CPU in the past. We will have TB of data in a very small form factor soon, providing bandwidth and latency optimal accesses. Then the Linux community is part of the developments made around these new devices, instead of running behind as it was the case in its enfancy. And finally it will change a lot the way we architect solutions in the future, especially thinking about optic introduction to interconnect all these new components. So the future 10 years are a as bright as were the past 20 ones at least.

Ric Wheeler

That days was also the day of the Penguin Dinner, which was at a walking distance from the University.

Penguin Dinner

And I must confess I passed that dinner as a lzay guy speaking french with Thomas, and also 2 other french persons from Neo Caledonia ! We were near the Swan River and beloeve me, it was pretty cold compared to the Monday, so I ended up having a small flu :-( But that was a nice time remaking the world again and discussing technology.

French discussion at LCA

The place was really neat with a view over the Swan River up to the city, and we stayed there a couple of hours, before going back to our hotels, tired, but happy of the day, and just waiting for the next one !

Swan River view

BTW all videos are now online so you can easily make your own opinion by looking yourself at them, as if you were there (you just lack the sun, the wine, and the nice talks ;-)) Cf: http://mirror.linux.org.au/linux.conf.au/2014/

Perth by night

Day 2 at Linux.conf.au

2014/01/09

Those interested by the start of the story can read the previous article on LCA14.

Kate Chapman
Day 2 at Linux.conf.au started with another keynote.

This time we had a presentation of a subproject of OpenStreetmap called HOT and showed how they’re using and promoting Open Street Map in humanitarian situations such as earth quake or floods. Typically was intereting that they now use it proactively in some part of the world such as Indonesia to *prevent* disasters, rather than reacting afterwards.
Kate Chapman

Listening to that, I thought that each country should propose to their young people in age of conscription (when they still have such thing in place) to contribute freely during a 3/6 months period to such an initiative: They don’t need high skilled IT engineers, but just people able to make correct maps of a given area in their neighbourhood. We could end up having a large Open Data collection in Open Street Map, regularly up to date.

Anyway, I was impressed by the generosity of this project, and to see how concretely useful some of the work which is done by hthe Open Source community is used.

The rest of the day was again dedicated to Miniconfs. I passed my time in OpenStack one. The track was organized by Michael Still of Rackspace who was last year Conference Director and made probably the most interesting session of the week in introduction, which consisted into giving an OpenStack T-shirt to all attendees ;-)
Michael Still

There were then some changes in the set of sessions due to infortunate problems for some speakers. So James E Blair from the OpenStack Foundation and Michael Still did the presentation on Open Stack governance on behalf of Tristan Goode.
James Blair & Michael Still
Followed by Paul Holland’s presentation on the move to a foundation made by the Open Stack community to manage their brand and governance model just explained, which is quite unique in its willingness to be egalitarian, which is probably a big concern when you gather thousand+ developers and dozens+ companies, and you want everybody plays the game. So the whole election mechanism, to board, to PTL roles, the gated trunk, … are probably mandatory now for such huge project with such a young history. I’d not necessarily recommend every project to adopt it as it will IMO be overkill for most of them,even if continuous testing (for them with Jenkins) is for sure a good idea (I wish they would do continuous packaging as well ;-)). However, for the projects I’m leading, it’s still an objective rather than a reality.
Paul Holland

In the afternoon, there were presentation on an incubated Open Stack project by Bruno Lago for billing called Artifice and based on the recently introduced telemetry OpenStack Ceilometer. There are different proposals around this billing need, and time will say which solution is retained as a project.

The following presentation was more interestig for me as it was on improving Network support in OpenStack by Iain Robertson from Brocade. It’s interesting to observe how this area of Network Functions Virtualzation (NFV) and Software Defined Network (SDN) is moving so fast, which is surely only possible due to the rapid evolutions made on lots of Open SOurce components to support them. I think there will be more and more overlap betwee nthe functions provided by the major Frameworks such as OpenStack (python based) OpenDaylight (java based) if only for language eason (API, interfaces, reuse of code) …

Then we had a very good presentation of Erasure Code for Open Stack Swift by John Dickinson, Swift PTL. which explained how this sort of RAID technology for Swift, the object storage brick of OpenStack, will be provided, and made a demo which helped understanding the concepts exposed.
John Dickinson

Anita Kuno, HP, came back on stage for a short presentation on some unknown aspects of OpenStack around the CI tooling (grab her presentation when it will be published to ge tthe links).
Anita Kuno

Then again Robert Collins, HP, took over Joe Gordon’s presentation on the evolution of OpenStack between Diablo and Havanna. Which on one hand looks very similar with the same core components, and also lots of differences with three times the number of lines of code and many additional bricks added. Of course, while not the original speaker, Robert’s extremely large knowledge of OpenStack in a whole, allowed him to very precisely show the dynamism of this project and the progresses made in the last 2 years. He also explained how Nova lost most of the features it had initialy to create separate components instead.
Robert Collins

Overall remarks about LCA2014: timing are respected very strictly, and they provide enough time in order to move from one conference to another (which could take 5 minutes crossing the university), they have timekeeper in each room, video in each room (at least all the one I was in). The first day they forgot to print banner to indicate where the rooms were, so it was a bit difficult to find its way (despite online maps, but regular readers know I’m old(fashioned) and also like signs !). Day 3 was thus perfect with regards to that !
Signs on Wednesday !

I had a very nice dinner that second day at the Friends Restaurant, with a Grilled barramundi fillet, quinoa risotto, roasted almonds and confit summer vegetable in extra virgin olive oil ! Of course, with such a title, It could only be a french touch, and indeed the chef was french. But the fish was local and very good !
Dinner

That was a long evening full of mail, and some coding. More pictures are available at https://plus.google.com/photos/112434061686721373729/albums/5966194867275136465

Next day is the start of the main conferences !


Follow

Get every new post delivered to your Inbox.

Join 99 other followers