Bug#434097: Announce of the upcoming NMU for the debian-edu-install package
Hi, On Wednesday 30 April 2008 08:03, Christian Perrier wrote: Everything committed. I'll commit new translations as they arrive until the deadline. Merci bien! regards, Holger pgpr8JApyVbyX.pgp Description: PGP signature
Bug#465273: doesnt update /etc/readahead/*
Hi, On Friday 02 May 2008 09:15, Petter wrote: I tested this issue in Lenny, and am sure the problem is non-writable /var/run/ at the very start of the boot. The mv to move the pid file fail, and setting RAMRUN=yes solve it. This patch should solve the issue, by moving the pid file to /lib/init/rw/, and should get the profiling to work. Please test it. Readahead works for me on etch since I upgraded to 2.6.22-4, still using 1.20060421.1016-1~bpo+40+1 I've reported this in http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=465273#25 regards, Holger pgpqftmimOYdK.pgp Description: PGP signature
Bug#456640: status of munin ITA
Hi Thijs, On Friday 02 May 2008 13:33, Thijs Kinkhorst wrote: As I use munin quite intensively for work I'm interested in the status of this ITA: it seems good progress was made but no new upload yet. Given the release schedule for lenny, now would be a very good time to upload things. Yup, we (at least Matthias Schmitz and me) are aware of the issue, Matthias has prepared a 1.2.6rc2 package too, which I'll check as soon as he fixed the one issue I already complained about. We plan to get 1.2.6 into lenny and 1.3.x in experimental, though obviously the latter is less pressing atm. Do you need any help? I'd be happy to comaintain or help out in another way if needed. Sure, help is always welcome. Do you have svn upstream access? Cause, this is where we keep the debian packaging too. We also hang out on #munin on freenode. regards, Holger P.S.: Thanks for caring! pgpstZrioXsvS.pgp Description: PGP signature
Bug#478729: ITP: batman-adv-source -- Source for the batman-advanced kernel module.
Hi Simon, On Wednesday 30 April 2008 18:27, Simon Wunderlich wrote: * Package name: batman-adv-source Version : 0.0+svn1041 Cool. Do you really plan to maintain the package or are you looking for someone to package it? If you plan to maintain it, I'd be happy to sponsor your uploads. (Assuming you're not a DD, havent checked.) Also, which version of batman does it need? As you probably know, I maintain batman and lenny+unstable only has 0.2 as you havent done a release since then ;-) (Experimental has some 0.3 svn snapshot..) BTW, I'd also be very happy if you would co-maintain batman ;-) And last question, do you think batman is ready to be part of a stable release? This means, do you think there is a version we can support for 30 months at least? regards, Holger pgpFggyG6XYWG.pgp Description: PGP signature
Bug#469716: Please upload a fix for this release goal issue
Hi, what really concerns me much more than this bug (which impact is almost zero in this special case) is that we will only have very little time to work on getting FAI into shape for Lenny... :( regards, Holger pgpOTh7MczwZZ.pgp Description: PGP signature
Bug#465273: doesnt update /etc/readahead/*
Hi Petter, On Saturday 03 May 2008 09:30, Petter wrote: I know, but I have a hard time to believe that readahead-watch is able to store its pid file before /var/run/ is writable, and thus leaving stop-readahead with no pid file to read to figure out the pid to stop at the end of the boot. Heh, makes sense. I assume here that you mean that 'readahead works' as in a profile run with readahead works. I could profile the boot process (with that kernel commandline param) and after doing this, when booting I see all files are read ahead in the beginning and the boot process is much faster. After the one profiling I have done, I immediatly rebootet the computer after logging in. /etc/readahead/boot was updated, /etc/readahead/profile not. regards, Holger pgpW5AYtVRiCm.pgp Description: PGP signature
Bug#370324: any news?
Hi Stefan, the Lenny freeze is approaching fast, any ETA when you will be able to fix this bug? We would really love to see it fixed in Lenny... also please shout if you need help... regards, Holger pgpEzDLPlUtVF.pgp Description: PGP signature
Bug#370346: any news?
Hi Steve, the Lenny freeze is approaching fast, any ETA when you will be able to fix this bug? We would really love to see it fixed in Lenny... also please shout if you need help... regards, Holger pgplYZ48mE6gA.pgp Description: PGP signature
Bug#473460: any news?
Hi LaMont, the Lenny freeze is approaching fast, any ETA when you will be able to fix this bug? We would really love to see it fixed in Lenny... also please shout if you need help... regards, Holger pgpLBLtAgiR00.pgp Description: PGP signature
Bug#482017: Bug#482015: usplash: should use omitpit feature to avoid getting killed by sendsigs
severity 482015 important severity 482017 important kthxbye Hi, raising severity as this change breaks splashy and usplash. regards, Holger pgpGKXk65aNng.pgp Description: PGP signature
Bug#480856: [Debian-olpc-devel] Bug#480856: Bug#480856: Bug#480856: no activity launcher
severity 480856 wishlist retitle 480856 Please document new activity launcher style kthxbye On Monday 12 May 2008 23:57, Jonas Smedegaard wrote: Perhaps you just are just not looking the right place - I remember now that it took me some time to realize it too, after the switch to the new design: When in the Home zoom level, choose List view in the upper right corner, and select the star symbol for your favorite activities that you want selectable in the Ring view. Ah, that was my problem indeed. Thank a lot for enlightening me! I think it might help to document this in a README.Debian, even though it's a upstream feature/change. Reasoning: many users will know sugar from the install currently deployed by OLPC, which has the old activity launcher style. Basically just put the text I quoted above into a README would be very fine. Or NEWS.Debian ;) I believe upstream wants Sugar to always have some activities in the ring. And I recently learned that they consider the new interface design as pre-alpha, but on the other hand wants me to not roll back to the older one... Okay... I think we have some sugar upstreams reading these bugreports now too :-) Hi! regards, Holger pgpFbr76RneeD.pgp Description: PGP signature
Bug#482386: debian-edu-config is uninstallable
package: debian-edu-config severity: serious version: 1.418 Hi, debian-edu-config is uninstallable in testing because it depends on svk, which is uninstallable on all archs due to #479763 (svk has broken depends) which in turn is blocked by #479698 (perl memory corruption). regards, Holger pgpPqMdVvWhm1.pgp Description: PGP signature
Bug#481871: Softupdate support in simple example
Hi, On Friday 23 May 2008 11:55, Michael Tautschnig wrote: AFAIK disk_var.sh is also available during softupdates Only if you installed with FAI, which is not a requierement to use softupdates. regards, Holger pgpN6ZesHwlfr.pgp Description: PGP signature
Bug#481581: [Debian-olpc-devel] Bug#481581: Bug#481581: sugar: Sugar can power down the computer
Hi, On Sunday 18 May 2008 00:53, [EMAIL PROTECTED] wrote: We could replace shutdown by logout. I've created a quick little patch that add logout to the menu and allow us to return to gdm/kdm/xdm. I would prefer if this would be a configuration option (whether to shutdown or logout), possible system wide, via a (low prio, preseedable) debconf question? regards, Holger pgp3fuxBncJT7.pgp Description: PGP signature
Bug#473618: doesnt work for me
Hi Winnie, attached is my patched for the german keyboard.lst, which applies (if you throw it in to the just uploaded 1.5.17.dfsg1-2 and edit debian/patches/series) nice, but doesnt work: tuxmatch still complains check_needed_unicodes_str() - needed char 'Ä' (Unicode value = 196) not found. Word 'BÄR' not added - contains Unicode chars not in keyboard list check_needed_unicodes_str() - needed char 'Ö' (Unicode value = 214) not found. Word 'FÖN' not added - contains Unicode chars not in keyboard list check_needed_unicodes_str() - needed char 'Ü' (Unicode value = 220) not found. Word 'FÜR' not added - contains Unicode chars not in keyboard list Funnyly ß works. I dont grok the format of keyboard.lst, you obviously do, so I would be glad if you could update the patch :-) Also I wonder if the other patch in this bugreport is still needed, tuxtype didnt crash for me... regards, Holger Index: tuxtype-1.5.17.dfsg1/tuxtype/data/themes/deutsch/keyboard.lst === --- tuxtype-1.5.17.dfsg1.orig/tuxtype/data/themes/deutsch/keyboard.lst 2008-06-25 19:05:36.0 + +++ tuxtype-1.5.17.dfsg1/tuxtype/data/themes/deutsch/keyboard.lst 2008-06-25 19:08:21.0 + @@ -54,7 +54,7 @@ 5| 0|! 0|@ -0|# +9|# 0|1 0|2 0|3 @@ -67,5 +67,14 @@ 0|0 7|, 8|. -9|; +7|; +8|: +9|- +9|_ +9|ö +9|à +9|ü +9|à +9|ö +9|à 9|à pgpXe9lNZcYat.pgp Description: PGP signature
Bug#464949: please support debhelper 7
retitle 464949 please support debhelper compat level 7 thanks Hi Craig, I better retitle this fast, before you close it as done :-) regards, Holger pgpGCwNKzfGgf.pgp Description: PGP signature
Bug#488262: missing depends on rsync
package: cdd-dev severity: important version: 0.5.2 Hi Andreas, while trying to build debian-edu from svn in a clean lenny environment, make dist didnt complete, because rsync wasnt installed. Please add it to cdd-dev's depends. Thanks. regards, Holger pgps7F2l0Hdaq.pgp Description: PGP signature
Bug#480478: ITP: debian-backports-keyring -- GnuPG archive key of the backports.org repository
Hi, while I'm actually in favor of adding this package because it makes it a lot easier to obtain a trustpath to the backports.org repo, which is important to our users, it's not true that there isnt a documented trusted path to install the key. It's documented here: http://wiki.debian.org/DebianEdu/Documentation/Etch/HowTo/Administration#head-136bb7e75e07e8b6463e6b30761ac51776c5c27d # add backports.org repo to /etc/apt/sources.list echo deb http://www.backports.org/debian etch-backports main contrib non-free /etc/apt/sources.list # install the debian-keyring securily: aptitude install debian-keyring # fetch the backports.org key insecurily: gpg --keyserver pgpkeys.pca.dfn.de --recv-keys 16BA136C # check securily if the key is correct and add it to root's keyring if it is: gpg --keyring /usr/share/keyrings/debian-keyring.gpg --check-sigs 16BA136C gpg --export 16BA136C | apt-key add - # update the list of available packages: aptitude update But it's really quite complicated and a lot to type :) So I would definitly prefer a package, optionally with a low-priority debconf question (for preseeding mostly) to also edit to sources.list :-) regards, Holger pgptkkrwhEKxd.pgp Description: PGP signature
Bug#480478: ITP: debian-backports-keyring -- GnuPG archive key of the backports.org repository
On Saturday 28 June 2008 02:48, Holger Levsen wrote: It's documented here: http://wiki.debian.org/DebianEdu/Documentation/Etch/HowTo/Administration#head-136bb7e75e07e8b6463e6b30761ac51776c5c27d now also with the correct order of commands :-) regards, Holger (see, it ain't easy :-D pgpsQSpEhNSCA.pgp Description: PGP signature
Bug#487747: debian-edu-install: [INTL:pt] Updated Portuguese translation for debconf messages
Hi Christian, On Tuesday 24 June 2008 07:08, Christian Perrier wrote: ../debian-edu-install.templates:2001 msgid Workstation msgstr Estação-de-Trabalho Actually, that's up to you, but indeed nothing needs that you *have* to use these ugly hyphens. So, if you prefer having Estação de Trabalho, you can very well use that. To debian-edu-install maintainers, this is indeed something which is a good use case for the Choices-C trick in debconf templates so that you can add an English translation directly in the templates: __Choices: Main Server, Workstation... Choices-C: Main-Server, Workstation Thus, users will show the choicees with hyphens while what will be without passed to debconf is what you did put in Choices-C. Good way to show friendly strings to suers while your scripts use more cryptic ones. Actually, I'm a bit hesistant to do this, not only will that make all translations fuzzy, it will also make preseeding files not work anymore, as one has to use localised preseed files, right? Or can I preseed an installation to be in, say, german, and use Choices-C preseed values? (It's probably also because I dont consider hyphens to be that ugly :) Yet another tip for a possible Tips and tricks writing debconf templates talk/BOFH/course at DebConf8..:-) Please, also create a written document...! :) regards, Holger pgpq4imzqhEEm.pgp Description: PGP signature
Bug#487678: debian-edu-install: Inconsistent usage of \t in debconf template
tags 487678 +pending thanks On Monday 23 June 2008 22:37, Nicolas François wrote: The problem is in the debian/debian-edu-profile-udeb.templates file, which contain tabulations. Tabulations should be expanded there. Here is a patch for the template file. Thank you, Nicolas. Petter committed this to svn on tuesday, so it will be included in the next upload. regards, Holger pgp4uekzTDJvx.pgp Description: PGP signature
Bug#293749: please reopen if still exists
please reopen if still exists pgpo3a93Agssn.pgp Description: PGP signature
Bug#488386: please provide an udeb
package: laptop-detect version: 0.13.6 x-debbugs-cc: [EMAIL PROTECTED] severity: wishlist Hi, currently laptop-detect doesnt provide an udeb. For historic reasons the debian-edu-profile-udeb (binary) package includes a code copy of laptop-detect, so that we can configure d-i differently depending on whether the system is a laptop or not. I'd like to get rid of this code copy (#487404), so I'm asking you to provide a laptop-detect-udeb. regards, Holger pgpw5g9HPe8EV.pgp Description: PGP signature
Bug#488386: please provide an udeb
block 487404 by 488386 thanks Hi Frans, On Saturday 28 June 2008 16:55, Frans Pop wrote: A udeb was recently removed. Do you mean a year ago or was it recently removed again? In our opinion there should be no reason why the _installer_ needs to be aware it's installing to a laptop or not. I tried to explain that in the bug report. We (Debian Edu) use it to modify the profile which gets installed... Also, Otavio, the laptop-detect maintainer stated in 426608 that he wouldnt have a problem to reintroduce the udeb. Any reason you cannot just chroot into /target and run laptop-detect there? Because /target is not set up, when we need it. You'll have to provide a much better/clearer use-case scenario before we'll allow a udeb to be reintroduced. I hope I did :) regards, Holger pgp7BO2N8MsVj.pgp Description: PGP signature
Bug#488386: please provide an udeb
Hi, On Saturday 28 June 2008 17:43, Frans Pop wrote: Yes, looks like that was about a year ago (even if the changelog does not mention it). Ah, ok. I tried to explain that in the bug report. We (Debian Edu) use it to modify the profile which gets installed... Sorry, but that is Greek to me. What is a profile, what does it do? In a normal Debian Edu installation the user is prompted to choose one out of four profiles, which define how the machine will be used, that is, install a certain set of packages and apply configurations. For those four profiles certain normal d-i steps (like for example task+package selection) are skipped. Using expert installation you can choose out of seven profiles, including one which has task and package selection :) Some of these profile can mixed, some cant. And their is an eighth profile, Debian Edu laptop, which is added dynamically, if laptop-detect succeeds. How/why exactly does it need to be different for a laptop than for a desktop? I believe it only differs in that it installs some more packages - implemented as described above. Also, Otavio, the laptop-detect maintainer stated in 426608 that he wouldnt have a problem to reintroduce the udeb. So, why hasn't this been a problem for debian-edu in the past year? Because nobody noticed. I only filed #487404 seven days ago... Why does this suddenly need to happen when Lenny is about to be frozen? It doesnt (absolutly) need to happen now. I've just upgraded our code copy to the one of laptop-detect 0.13.6. But not shipping code copies is a release goal... Also, adding a udeb is not up to Otavio as laptop-detect maintainer. Any new udeb needs to be approved by the D-I team and thus discussed on the debian-boot list. So I'm adding debian-boot to cc: now :) Any reason you cannot just chroot into /target and run laptop-detect there? Because /target is not set up, when we need it. OK, that could be a valid reason. But you've not yet explained why you need it _before_ /target is set up. We modified d-i so that most interactive questions are at the beginning (before there is /target), to provide a more pleasant installation experience :-) One of these questions is the profile question, which needs to be modified if a laptop is detected. You'll have to provide a much better/clearer use-case scenario before we'll allow a udeb to be reintroduced. I hope I did :) Not really, or at least, not yet in sufficient detail to satisfy me personally. Now? regards, Holger pgpkVoJEZuwax.pgp Description: PGP signature
Bug#488262: missing depends on rsync
Hi, On Sunday 29 June 2008 16:20, Andreas Tille wrote: What do you think (Suggests / Depends / replacing rsync in the Makefile) to solve this bug? I'd prefer a recommends on rsync or a rewrite to make rsync not needed. regards, Holger pgp8wejcrlvSo.pgp Description: PGP signature
Bug#488644: refers to non-existing /usr/share/doc/mailman/README.POSTFIX
package: mailman version: 2.1.9-7 Hi, thanks for packaging mailman! /usr/share/doc/postfix/README.Debian for configuration with postfix refers to /etc/mailman/postfix-to-mailman.py which in turn tells me to configure /etc/mailman/mm_cfg.py, where I should set #MTA = None # No MTA alias processing required OTOH, /etc/mailman/mm_cfg.py tells me to set MTA=postfix (not None!) and that I should read /usr/share/doc/mailman/README.POSTFIX, which doesnt exist. Now I'm a bit confused :) I assume that MTA=None is right when following the advice in /etc/mailman/postfix-to-mailman.py but I'm not fully sure. It would be nice if you could cleanup /etc/mailman/mm_cfg.py to reflect what should be done. Thanks. regards, Holger pgpWTZGUKrLtx.pgp Description: PGP signature
Bug#489132: lenny release notes, upgrade dpkg first
Hi, On Thursday 03 July 2008 12:25, Raphael Hertzog wrote: Given that the new dpkg also supports triggers, we should probably also recommend to upgrade apt/aptitude at the same time otherwise those tools might be confused by the new package status... #464559 (in apt) is also a problem for etch to sid/lenny upgrades, so IMHO this is a sensible idea anyway... regards, Holger pgpJOkXT1OKX7.pgp Description: PGP signature
Bug#487404: code copy updated
Hi, FYI: in the upcoming 0.670 version the laptop-detect code copy has been updated to the code in 0.13.5... regards, Holger pgprtal4A3G8M.pgp Description: PGP signature
Bug#464559: fixed in 0.7.11
Hi Michael, thanks for the information. On Tuesday 17 June 2008 12:36, Michael Vogt wrote: this bug is fixed in version 0.7.11 of apt (rev. 1585..1586 in the debian-sid branch). If this is a problem for etch-lenny upgrades, this change needs to be either backported to etch or a workaround needs to be applied. There is some information in http://launchpad.net/bugs/217435 what can be done to work around the problem. The only workaround I saw there was to change+relax the dependencies of packages which trigger the bug in apt, I dont think this is a good fix/workaround. (Because we might miss some packages.) But then, a mandatory apt upgrade before upgrading to lenny is also optimal, even though this has been done before. regards, Holger pgpKnAPDsn6Es.pgp Description: PGP signature
Bug#482687: 0.4.0-1 build fine on sparc
notfound 482687 0.4.0-1 close 482687 0.4.0-1 kthxbye Hi Frank, sugar-hulahop 0.4.0-1 built fine on sparc, so I'm closing this bug. regards, Holger pgpvq2Wz9XKTu.pgp Description: PGP signature
Bug#485817: please include a Select CDD config option
Hi Frans, On Wednesday 11 June 2008 19:02, Frans Pop wrote: Well, if you want to use the settings infrastructure I introduced in the thread you're replying to, you'll at least have to wait until that's actually implemented and available. Heh. Point taken. Do you assume that this will happen before Lenny? BTW: your use of x-debbugs-cc is incorrect. No :) (Or maybe it is, but it worked as intended.) regards, Holger pgpWpZuMlVk1q.pgp Description: PGP signature
Bug#482687: [Debian-olpc-devel] Bug#482687: 0.4.0-1 build fine on sparc
http://buildd.debian.org/fetch.cgi?pkg=sugar-hulahop;ver=0.4.0-1;arch=sparc;stamp=1213041124 forgot to include that link... pgpZfxTIPi5Gv.pgp Description: PGP signature
Bug#486868: #486868: munin-node: upgrade also broke exim_mailstats
Hi, On Friday 20 June 2008 09:54, Damyan Ivanov wrote: While adding tail_close call, don't forget to remove the bare close() call, otherwise tail_close will fail. Thanks. Can I help for this to be fixed before the freeze? Thanks for your offer, but I assume we can handle this :) It would be very bad to release with broken exim plugins. Exim is the default mailer... True, but the freeze doesnt mean we cant fix stuff anymore. Quite the opposite: the freeze means, we can only fix stuff, not introduce new features :-) Also true is, that every fix needs to be approved by the release team and we should try to keep their load low :-) regards, Holger pgpa8wDlOULsU.pgp Description: PGP signature
Bug#486868: #486868: munin-node: upgrade also broke exim_mailstats
Hi, on re-reading this: On Friday 20 June 2008 11:13, Holger Levsen wrote: Can I help for this to be fixed before the freeze? Thanks for your offer, but I assume we can handle this :) one should never turn help down :) Of course you can help by providing and testing patches and sending them to the bts :-) regards, Holger pgpkS5Zwz0MjK.pgp Description: PGP signature
Bug#487404: dont copy code from laptop-detect
package: debian-edu-install severity: important version: 0.669 Hi, ./debian-edu-profile contains code from laptop-detect refering to #426608 which has been fixed more than a year ago. Please remove that code and use the one from laptop-detect. Filing this as severity important as no code duplication from other packages is a release-goal. Not fixing this myself as I'm not familar with the code in question. regards, Holger pgp93GUgWoZdP.pgp Description: PGP signature
Bug#487404: Acknowledgement (dont copy code from laptop-detect)
Hi, additionally our code copy checks for /proc/sys/acpi to detects laptops, which has been deprecated in 2.6.24. laptop-detect (0.13.5) correctly checks for the /sys power interface (see #463662). regards, Holger pgpaFbOaoS2B1.pgp Description: PGP signature
Bug#487413: no handling of nagios2 config files on upgrades
package: debian-edu-config severity: important version: 1.419 pere h01ger: saw you switched to nagios3. Remember that all conffiles need special handling when removed, and all files in /etc/ are conffiles. h01ger that only affects upgrades, right? pere h01ger: yes, it only affects upgrades. regards, Holger pgpdmZhKZSxcI.pgp Description: PGP signature
Bug#480478: ITP: debian-backports-keyring -- GnuPG archive key of the backports.org repository
Hi, On Saturday 21 June 2008 15:52, Alexander Wirt wrote: I'm still not that sure if its a good idea to add a non-offical debian repo keyring into the archive... Nobody is forced to install it?! And AFAICS we regulary recommend backports.org to users, who need newer software. So I think it should be in. regards, Holger pgpyDvLsaISRi.pgp Description: PGP signature
Bug#487419: debian-edu-install: [INTL:de] updated German debconf translation
Hi Christian, On Saturday 21 June 2008 20:01, Christian Perrier wrote: Hmmm, Is this the consequence of a call for translation updates ? yes. For .fr it has been sent to Last-Translator: Cyril Brulebois [EMAIL PROTECTED]\n and Language-Team: French [EMAIL PROTECTED]\n regards, Holger pgpETW9IbjrpP.pgp Description: PGP signature
Bug#487488: marking your translations invalid for now
tags 487419 +invalid tags 487488 +invalid tags 487500 +invalid thanks Hi, I'm marking your bug reports as invalid for the reasons stated in my third call for translations. Sorry again about that. Please send in new translations to those bugs, I'll pick them up almost immediatly :) (And then I'll tag the bugs again..) regards thanks for your work, Holger pgpuFkALg746R.pgp Description: PGP signature
Bug#440622: closed by Holger Levsen [EMAIL PROTECTED] (Bug#440622: fixed in munin 1.2.6-1)
found 440622 1.2.6-1 kthxbye pgpjYpIHouKbQ.pgp Description: PGP signature
Bug#487486: debian-edu-install: [INTL:fr] French debconf templates translation update
Hi Christian, On Monday 23 June 2008 06:55, Christian Perrier wrote: A stupid spelling error spotted by proofreaders..:-) profil comes oout of my fingers with a final e. I definitely write too much English now...:-) Hehe :) Comitted to svn. BTW, when unfuzzying you also missed that I changed bugs.skolelinux.no to bugs.skolelinux.org in the template ;-) (SCNR to point this out, your help with improving the english templates was *great*, so I was kinda happy to find this mistake ;) Though I would probably have been my job to update that in the translations and in fact, it's on my agenda... regards, Holger pgpllfbk87wLh.pgp Description: PGP signature
Bug#487678: debian-edu-install: Inconsistent usage of \t in debconf template
tags 487678 -patch reassign 487678 po-debconf thanks Hi Helge, On Monday 23 June 2008 16:51, Helge Kreutzmann wrote: While updating the German Debconf translation, I noticed that you sometimes (!) use \t. Actually, I dont. It's present in the .po files, but not in the templates, so sometime po-debconf adds it there, so reassigning and removing the patch tag. (As the .po files are autogenerated/updated (and that adds the \t, I cannot apply that patch anyway.) regards, Holger, subscribing to this bug pgpWV4Fhn3k1L.pgp Description: PGP signature
Bug#479248: does not install lintian overrides
package: cdd-dev version: 0.5.0 Hi, the debian-edu source package contains debian/lintian.override but this is not installed by /usr/share/cdd-dev/rules thus filing this bug. regards, Holger pgpspJPArgBMf.pgp Description: PGP signature
Bug#479249: does not install README.source
package: cdd-dev version: 0.5.0 Hi, http://lists.debian.org/debian-devel-announce/2008/04/msg00016.html mentioned that it's recommended now, to include a file README.source which describes how to get/modify the source package if it's not straightforward. cdd-dev currently only checks if a file README exists and adds this as a param to dh_installdocs. Please also check for and add README.source as well. regards, Holger pgpcXiVxuucTG.pgp Description: PGP signature
Bug#479248: does not install lintian overrides
Hi Andreas, On Sunday 04 May 2008 00:34, you wrote: Nice feature request. When writing cdd-dev I had the same policy as debhelper in mind which does intentionally not provide dh_installlintianoverrides (or something like that - I remember a bug about this, but I'm to tired now to verify the bug number). Me too, but I'd like to know :) But I agree that we might perhaps add such a feature to cdd-dev. On the other hand we need a means to ensure that the overrides will be put into the right meta package. Sure. But knowing Joey Hess, I doubt that was the reason :) This will probably not work with the corrent lintian.override in the debian-edu source. I think it will, using approach 3: I see three chances: 2. Use separate override files and use a naming sheme like task.override. Why use 10 files when you can use 1? 3. Grep the single lintian.override file for metapackage-names and put the relevant entry into the according meta package. I'd suggest to do this. regards, Holger pgp3zK5JdIZUR.pgp Description: PGP signature
Bug#466690: (no subject)
Hi, On Sunday 04 May 2008 11:03, Thomas Lange wrote: IMO your problem is the cause of the different kernel. I also like to keep the order of disk, that is created by the kernel. AFAIK this order depends on the order the kernel loads the drivers. If we sort the disks by their names, we will lost this information. Isnt nowadays this order random? regards, Holger pgp9M3JF5heRD.pgp Description: PGP signature
Bug#449235: Intel wireless firmware in firmware-nonfree
Hi, On Saturday 03 May 2008 23:57, Per Olofsson wrote: Bastian Blank wrote: We decided that we can't ship it with this license in non-free. Where did you document this decission? On what is it based? But if you have already decided that the firmware can't be packaged, I guess this bug should be closed or marked wontfix. Or rather taken to the tech CTTE or whatever... [1] http://ipw2100.sourceforge.net/firmware_faq.php Looks perfectly fine for nonfree to me. regards, Holger pgpm1bvgPUnJh.pgp Description: PGP signature
Bug#449235: Intel wireless firmware in firmware-nonfree
Hi, On Sunday 04 May 2008 17:23, Waldi wrote: | Your rights to redistribute the Software shall be contingent upon your | installation of this Agreement in its entirety in the same directory as | the Software. This was interpreted by at least one person that the license needs to pushed in the same directory on the mirror than the package. And because there is one person interpreting this differently than the licencing party itself, Debian shall not do what the licencing party and the users would like to see? I have every reason to believe Intel wants us to distribute the firmware! The question where this discussion is documented is still open. And the question whether you as the maintainers agree with that one person or if you going to change this... regards, Holger pgpkX6ukiqrsl.pgp Description: PGP signature
Bug#449235: Intel wireless firmware in firmware-nonfree
Hi, On Sunday 04 May 2008 19:12, Bastian Blank wrote: This was interpreted by at least one person that the license needs to pushed in the same directory on the mirror than the package. And because there is one person interpreting this differently than the licencing party itself, Debian shall not do what the licencing party and the users would like to see? Who wants to define that someone is not allowed to define a binary dump as source and publish it under the GPL and we are therefor not allowed to distribute such things? Even after reading this three times and having my second coffee I fail to understand what this paragraph has to do with the ipw2x00 firmware. It's not GPLed. Maybe I need a third coffee though. The question where this discussion is documented is still open. Somewhere in the IRC logs. Update: Just found it, was between aba and me, in #debian.de, 25.5.2006. So there is no public log. (And it was probably also in the wrong language :) And the question whether you as the maintainers agree with that one person or if you going to change this... If the ftp-team acks it, I have no problem with that. Cool. So let's get their confirmation. regards, Holger pgp6yX0cjbch8.pgp Description: PGP signature
Bug#478989: Bug#479248: does not install lintian overrides
Hi Andreas, thanks for fixing this bug and 479249 so fast! On Sunday 04 May 2008 09:32, Andreas Tille wrote: See bug #478989 as quotet above. I completely agree with Raphael Geissert and I really hope that Joey will implement this soon. /me too. CC:ing the bug to voice our support :) Even if I agree with you that the single override file is the better way to go, I would prefer to use common debhelper tools as they are and don't invent code which might be duplicated later in debhelper. Would this be acceptable for you? Totally. :) BTW, I think using dh_lintian seems to need a versioned dependency from debhelper 6.x? Or if you have quick access to a stable system running debhelper 5.x, could you please verify the existence of dh_lintian? It's not available in stable lintian. regards, Holger pgp1JSnxEmUG6.pgp Description: PGP signature
Bug#434097: Announce of the upcoming NMU for the debian-edu-install package
Hi Christian, On Thursday 08 May 2008 09:08, Christian Perrier wrote: OK, Holger. The deadline is now reached. No. Quoting you in http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=434097#12 --- Otherwise the following will happen (or already has): Friday, April 18, 2008 : send the first intent to NMU notice to the package maintainer. Tuesday, April 29, 2008 : send this notice Friday, May 09, 2008 : deadline for receiving translation updates Saturday, May 10, 2008 : build the package and upload it to DELAYED/2-day send the NMU patch to the BTS Monday, May 12, 2008 : NMU uploaded to incoming --- And I still plan to upload on saturday :-) May I suggest an upload now ? Sure :-D But I'll wait :-) regards, Holger pgpQbR5BXI6t2.pgp Description: PGP signature
Bug#480461: please support udebs in dget
package: devscripts severity: wishlist version: 2.10.27 tags: +dget Hi, on a sid system: $ dget debian-edu-install-udeb W: Unable to locate package debian-edu-install-udeb dget: debian-edu-install-udeb has no installation candidate While the udeb is available in sid. please support udebs in dget. regards, Holger pgpCYBCZ6ju1J.pgp Description: PGP signature
Bug#480461: please support udebs in dget
Hi Adam, On Saturday 10 May 2008 11:42, Adam D. Barratt wrote: That's because the udebs appear in a different packages file from deb packages. All you need to do is ensure that you have the udeb packages file available to apt. (i.e. dget doesn't support or not support udebs - it uses whatever information is available to apt) $ egrep install /etc/apt/sources.list deb ftp://192.168.0.12/mirrors/ftp.debian.org/ unstable main/debian-installer Ah, thanks. Leaving this bug open so you can decide whether you just want to close it or add a hint in the documentation ;-) regards, Holger pgpxmjyYvqXUa.pgp Description: PGP signature
Bug#362765: missing binary package: nvidia-kernel-2.4.27-3-k7
package: nvidia-modules-i386 Hi Randall, I just noticed nvidia-kernel-2.4.27-3-k7 is missing in all distributions, could you please take care of getting it in?! kernel-image-2.4.27-i386 now builds -3- packages, and nvidia-modules should do the same. (Especially in sarge.) Thanks regards, Holger pgp6tGNnxTN73.pgp Description: PGP signature
Bug#363701: don't ignore gpg-authentification for apt-repositories
package: fai version: 2.10.1 priority: wishlist Hi, in the default configuration, fai ignores gpg-authentification for apt-repositories. As this is a very handy and long-awaited feature which will be available in etch, I would like to see this fixed until etch is released. see http://faiwiki.debian.net/index.php/Using_gpg-authenticated_debian-archives for more infos. regards, Holger pgpQCrPycpvGV.pgp Description: PGP signature
Bug#353535: define classes on the fai commandline
package: fai version: 2.9.1 severity: wishlist Hi, it would be really useful (esp. for softupdates of systems), if one could add/define additional classes on the fai commandline, e.g. fai -N --define-classes FOO,BAR regards, Holger pgp8qspHvE4IQ.pgp Description: PGP signature
Bug#353536: classes which includes classes
package: fai version: 2.9.1 severity: wishlist Hi, I would be really useful, if classes could define sub-classes. For example with a sub-directory in /class, let's call it subclasses. If the class FOO is defined, and a file /class/subclasses/FOO exists, with the following contents: ---begin- BAR BAZ ---end--- the classes BAR and BAZ are also defined for that system. regards, Holger pgp3cKjBi2SRv.pgp Description: PGP signature
Bug#348664: test compile with CONFIG_NFSMOUNT in config-udeb-linux
Hi, I just did a test compile with CONFIG_NFSMOUNT enabled in config-udeb-linux, to see how much bigger it gets: Roughly 3k for the udeb and the busybox binary, are there any other sizes to check ? matrix:~$ ls -latr busybox-udeb_1.01-*udeb -rw-r--r-- 1 holger holger 133082 Feb 19 12:30 busybox-udeb_1.01-4_i386.udeb -rw-r--r-- 1 holger holger 136892 Feb 19 12:39 busybox-udeb_1.01-5_i386.udeb matrix:~$ ls -la busybox-1.01-4/bin/busybox -rwxr-xr-x 1 holger holger 230248 Feb 19 12:30 busybox-1.01-4/bin/busybox matrix:~$ ls -la busybox-1.01-5/bin/busybox -rwxr-xr-x 1 holger holger 233224 Feb 19 12:39 busybox-1.01-5/bin/busybox I've also not tested yet, if this is sufficient to mount nfs-shares. I want this enabled in the busybox udeb, so it's possible to use d-i boot-media to mount an fai nfsroot, to do fai installs... regards, Holger pgp8vsnPtgcjC.pgp Description: PGP signature
Bug#290259: Update regarding the keyboard
Hi Joerg, On Saturday 22 January 2005 16:12, Joerg Dorchain wrote: As I have sent a Console keymap already, I supply only the files for X in this mail you didn't :-) regards, Holger pgpjsQbVzFpsk.pgp Description: PGP signature
Bug#291954: The kdebase templates and the German translation
Hi, On Wednesday 26 January 2005 05:21, Adeodato Simó wrote: OK, so this is something you have to discuss among yourselves. I'm not very familiarized with the l10n process in Debian, but: do you think it'd make sense to X-Debbugs-CC: the appropriate -l10n- list upon submitting INTL bugs? Yes, I think this makes sense. regards, Holger pgpQoUTptwhUo.pgp Description: PGP signature
Bug#274649: Bug#292530: Should configure inittab for sercon if network-console is used
Hi, On Friday 28 January 2005 23:09, Frans Pop wrote: While installing with a serial console, it would be fine to have the matching inittab line enabled, so that I'm able to use the machine later on. I've committed a patch to SVN for this issue. The patch is fairly straightforward and I have tested both a normal installation (i386) and an installation over serial console using network-console (sparc U10). In both cases the /etc/inittab turned out as expected. I think this patch could be considered for Sarge. Wouldn't this patch (see 292530) fix 274649 as well ? regards, Holger pgpfqFFQ4Xsj2.pgp Description: PGP signature
Bug#348350: make-fai-nfsroot -K should remove the patches file too.
Hi, why are those kernel files removed with rm in the first place, and not with dpkg -r ?? 2nd: /boot/patches* belongs to the linux|kernel-image package and exists, if the kernel was build using patches. (and make-kpkg obviously). regards, Holger pgpxkAkQjw6Jt.pgp Description: PGP signature
Bug#286855: still opses on 1.10.2/2.6.15
Hi Björn, On Friday 03 February 2006 11:09, Björn Torkelsson wrote: With fai-kernels 1.10.2 it still opses for me in xfs. However xfs works with a home built kernel based on 2.6.15-rc4, with a completely different config though. Thanks for reporting this. I have no idea atm, why the kernel config should matter, but then I have never used xfs myself so far... 1.10.2 was build against 2.6.15-3. regards, Holger pgpfQwUBBTC4p.pgp Description: PGP signature
Bug#230217: seriously wrong pathes in FAI and #230217
Hi, Manoj is Manoj Srivastava [EMAIL PROTECTED], h01ger is me. from #debian-devel on OFTC: h01ger Manoj, any comment on #230217 ? Manoj h01ger: well, we need to come up with language that spell out which (perhaps optional) parts of FHS 2.3 are to be ignored by debian packages, and then change the recommendation. No one is stopping you from using FHS 2.3 today, really -- dancerjBO has quit (Remote host closed the connection) h01ger Manoj, but how should one read current policy that a newer version of FHS is right to use ? and more importantly, doesnt that bug list all problems with FHS 2.3 ? Manoj h01ger: whenever things in policy are to change, some packages do so in advance of policy h01ger ic Manoj indeed, in general, that is good, since then we can learn of any errors in a proposed scheme, and policy can then recommend the best practice, one which has actually been known to work Manoj h01ger: the only time not sticking with policy makes a difference, in practical terms, is release tiem, and you should talk to policy editors and release people to see when being ahead of the times is OK. I think it is, I should have uplaoded a new version of policy earlier, but my health has not been what it ought to be Manoj but since policy is team maintained, it has been less well maintained, this is the weakness of team maintained packages :) dato heh. huggie Eep, I thought we left the trolls on lists? h01ger Manoj, i'll forward this conversation to the fai maintainer (who wants to wait with fixing his bugs til this is decided) and tell him he should go for FHS2.3 (releaseteam people have spoken in favor of this regarding those fai bugs) - thanks. Manoj h01ger: thanks. If someome could read the bug report, look at my objections, and the conclusions we reached there, and come up with a paragraph that says that, for one thing, debian package need to not pay attention to the dotfil optional stuff, and we are going to ignore the name of the X configuration file, etc, it would help regards, Holger pgp3mH4MHiR3q.pgp Description: PGP signature
Bug#364696: /usr/local/share/fai - $FAI_LOCATION
Hi, On Tuesday 25 April 2006 03:57, [EMAIL PROTECTED] wrote: I believe that /usr/local/share/fai/hooks from above should be replaced by $FAI_LOCATION/hooks. Thanks for your bug report. It's actually correct at the moment - within fai's buggy default of FAI_LOCATION=/usr/local/share/fai. (too lazy to look the bugnumbers, Thomas knows which.) Said that, I'm still in favor of your suggestion. regards, Holger pgpjUDELzXklF.pgp Description: PGP signature
Bug#354551: discard all future messages from from:-address
package: listadmin severity: wishlist version 2.27-2 Hi, listadmin is pretty cool, but it lacks the feature to discard all future message from a specific from:-address - this is only possible with mailmans webinterface. It would be great, if you could add this feature. regards, Holger pgpnpBtvAsQXa.pgp Description: PGP signature
Bug#302929: Improper copyright file
Hi Justin, On Thursday 02 March 2006 03:28, Justin Pryzby wrote: Can I expect to see this bug fixed for etch? Yes. (etch as in stable, (maybe) not (soon) as in testing.) It was tagged 'pending' for over 6 months, and the package hasn't been updated in nearly 4 years. According do my calender, January 22nd 2006 is merely 6 weeks ago, not six month. Upstream is working on it, but as there a numerous licence issues (soundfiles fonts) this takes some time. An upload with only some of the issues fixed isn't sensible. regards, Holger pgpPNWbSoPmKm.pgp Description: PGP signature
Bug#356087: yaird error: no link to block device
package: yaird version: 0.12-4 Hi, matrix:~# apt-get install linux-image-2.6.15-1-686 Reading package lists... Done Building dependency tree... Done linux-image-2.6.15-1-686 is already the newest version. 0 upgraded, 0 newly installed, 0 to remove and 1 not upgraded. 1 not fully installed or removed. Need to get 0B of archives. After unpacking 0B of additional disk space will be used. Setting up linux-image-2.6.15-1-686 (2.6.15-8snapshot.6094) ... Running depmod. Finding valid ramdisk creators. Using mkinitrd.yaird to build the ramdisk. yaird error: no link to block device in /sys/devices/pci:00/:00:1f.1/ide0/0.0 (fatal) mkinitrd.yaird failed to create initrd image. Failed to create initrd image. dpkg: error processing linux-image-2.6.15-1-686 (--configure): subprocess post-installation script returned error exit status 9 Errors were encountered while processing: linux-image-2.6.15-1-686 E: Sub-process /usr/bin/dpkg returned an error code (1) matrix:~# iF linux-image-2. 2.6.15-8snapsh Linux kernel 2.6.15 image on PPro/Celeron/PI ii yaird 0.0.12-4 Yet Another mkInitRD regards, Holger pgpPjWaatLfgk.pgp Description: PGP signature
Bug#309917: please mention Skolelinux next to debian-edu on http://www.debian.org/devel/
Hi, www.debian.org/devel still only mentions debian-edu: the page says: ---schnipp--- A number of these projects aim to create Custom Debian Distributions for a particular group of users while working fully within the Debian system. These include: *Debian Jr. Project *Debian-Med Project *Debian-Edu Project *Debian Desktop Project *Debian-Lex Project *Debian Non-Profit Project *Debian Accessibility Project *Debian GIS Project *Agnula DeMuDi ---schnapp--- I would like to get consensus on this one now and close this bug soon. My proposal still is to s/Debian-Edu Project/Debian-Edu\/Skolelinux Project/ - do you agree or have better proposals ? Or do you think that page should remain as it is ? regards, Holger pgpnyoSHn2wQr.pgp Description: PGP signature
Bug#314991: add code to BTS
Hey Thomas, On Saturday 25 March 2006 23:11, Thomas Lange wrote: please add you svn checkout code for the config space to the BTS so we can have a look at it. I'm still only using my quick hack from last year, which you can find in svn://people/h01ger/fai-sarge and there in share/subroutines and conf/fai.conf In subroutines I just modified get_fai_cvs, without even bothered to rename the funtion :-/ I think it's a good idea to use a abstract name like FAI_SVNREPO for the fai-configdir-repository in fai.conf and also to define the checkout-command (FAI_CHECKOUT_CMD) and options (FAI_CHECKOUT_OPTIONS) there. regards, Holger pgpxuMSBfUnF7.pgp Description: PGP signature
Bug#342467: please implement something like this
Hi Thomas, I think it would be very useful to have this feature in fai. Björn, thanks for the patch! Though instead of FAI_HOOK_DIR=/etc/fai/hooks I would rather propose /etc/fai/make-fai-nfsroot-hooks/ or /etc/fai/nfsroot-hooks. regards, Holger pgpxNmqQYv4pe.pgp Description: PGP signature
Bug#361524: empty LOGSERVER var doesnt work as expected
package: fai version: 2.8.4 I'm forwarding this to the BTS because the fai-guide says, if LOGSERVER is not set, the installserver will be used for logging as well. As Ingo has reported, there is a case when this isn't true. regards, Holger -- Forwarded Message -- Subject: $LOGSERVER and $SERVER vs. softupdate: bug? Date: Saturday 25 March 2006 23:56 From: Ingo Wichmann [EMAIL PROTECTED] To: linux-fai linux-fai@uni-koeln.de Hi! In my /etc/fai/fai.conf on the installserver LOGSERVER= is empty, so as expected the installclient uses the installserver instead. When fai softupdate was run on this client, the logs did not appear on the installserver. I've fixed it by setting LOGSERVER to the name of the installserver. But why is this necessary? In /usr/lib/fai/fai-savelog, function save_log_remote() the variable LOGSERVER defaults to $SERVER : [ -z $LOGSERVER ] LOGSERVER=$SERVER I think this is a bug in the case of softupdates, because $SERVER is not set during softupdates. Ingo --- pgpzxbRyO2t57.pgp Description: PGP signature
Bug#361559: please document the commandline-parameter HOST=demohost
package: fai severity: wishlist version: 2.10 Hi, On Sunday 26 February 2006 14:38, Andreas Schockenhoff wrote: On Sunday 26 February 2006 11:26, Darryl Luff wrote: Hi all. I'm new to FAI. I've seen a few requests in the archives about setting the hostname on the built machine using other than the DHCP config, but none really did what I need to do. Try: append HOST=demohost Woot! This question has popped up so many times, but I've never seen that answer. We should definitly document this in the fai guide! regards, Holger pgpT2jdOayVw2.pgp Description: PGP signature
Bug#361786: problem with security updates and fai-cd
package: fai
Hi Thomas,
we talked about the issue yesterday morning, as you had no time to work on
this yesterday, I decided to forward it to the BTS so that this issue doesn't
get forgotten.
regards,
Holger
-- Forwarded Message --
Subject: Re: debootstrap: security updates
Date: Monday 30 January 2006 20:46
From: Christoph Mitasch [EMAIL PROTECTED]
To: linux-fai@uni-koeln.de
Hi!
On Tue, 29 Nov 2005 15:58:06 +0100, Christoph Mitasch [EMAIL
PROTECTED] said:
When calling make-fai-nfsroot debootstrap installs some base debian
packages but does not take security updates into account.
...debootstrap $FAI_DEBOOTSTRAP_OPTS $1 $NFSROOT $2...
Is there a way to get debootstrap to include security updates from
security.debian.org?
debootstrap itself can not use security.debian.org, because it only
uses one URI. But make-fai-nfsroot is doing and update;upgrade, so it
should work if you include security in /etc/fai/sources.list.
--
regards Thomas
security.debian.org was already included in /etc/fai/sources.list.
The security updates for the packages fetched via debootstrap are not
included.
I checked $NFSROOT/var/tmp/base.tgz. My impression was, that it does not
include packages from security.debian.org.
Now I have a new problem that I think is related to my security updates
troubles. Today I created a new fai-mirror and afterwards a new fai-cd.
When I try to install with the new FAI-CD it fails because there is a
conflict between a package from the base repository (without security
updates) and a package from security.debian.org that is trying to be
installed.
Any ideas?
Thank you!
Christoph
---
-- Forwarded Message --
Subject: Re: debootstrap: security updates
Date: Sunday 09 April 2006 03:25
From: Holger Levsen [EMAIL PROTECTED]
To: linux-fai@rrz.uni-koeln.de
Hi,
On Tuesday 31 January 2006 17:42, Christoph Mitasch wrote:
Here are the details:
The problem was that the function delete_base_packages in the
fai-mirror script removes all packages that are already included in
base.tgz. This function removed a security-updated package from the
fai-mirror.
Now when task_instsoft tried to install a package that depends on the
security-updated one, it failed because of a dependency to the newer
version of the updated package existed.
That updated package was not installed through task_extrbase and not
updated through task_updatebase since the fai-mirror didn't contain
the security-updated package.
Ok and now the concrete case.
When installing with fai-cd it fails at task_instsoft with the
following error:
perl: Depends: perl-base (= 5.8.4-8sarge3) but 5.8.4-8 is to be
installed
The reason for that is that the perl-base 5.8.4-8 package was already
installed through base.tgz and that perl-base 5.8.4-8sarge3 was
removed through delete_base_packages from the fai-mirror.
I think the delete_base_packages function should be reworked so it
does not remove security updates.
At the moment it removes packages with that command, where $p is a name
from base-pkgs.lis (e.g. perl-base)
# rm $archivedir/${p}_*.deb
There is a option -afor fai-mirror that solved my problem.
-a Add packages included in base.tgz and packages defined in
make-fai-nfsroot.conf to the mirror.
And now I'm just happy that it works again ;)
Christoph
On Mon, 2006-01-30 at 21:20 +0100, Michael Tautschnig wrote:
What kind of conflict is it? Could you be so kind and provide some logs
or the exact error message?
Thomas, you have not replied to this at all :( To me this looks like a bug
worth fixing - shall I commit it to the BTS tomorrow ?!
regards,
Holger
---
pgpvlZxpj4vW9.pgp
Description: PGP signature
Bug#361817: problem with fai_chboot and multi-interfaces on the server
package: fai
On Monday 10 April 2006 15:53, Thomas Lange wrote:
On Sun, 9 Apr 2006 03:03:51 +0200, Holger Levsen
[EMAIL PROTECTED] said:
Hi,
On Wednesday 15 March 2006 09:47, Andreas Jobs wrote:
Problem:
task_chboot does not disable the correct pxe configuration.
Description:
If you have more than one NIC, your hostname may not map to the
adress used during install. In my case (installing some Compaq Blade
Servers) the Installation uses eth0 but the production interface
is eth2. So, hostname resolves to the address bind to eth2. During
task_chboot fai does fai_chboot -r $hostname and quits with a file
not found message.
Solution:
Use the address that was used during installation. My
quick-and-dirty patch currently is:
--- subroutines-linux.orig 2006-01-12 17:49:49.0 +0100
+++ subroutines-linux 2006-03-10 09:41:23.0 +0100
@@ -185,7 +185,8 @@
local frsh
local doexit=0
-local hostname=$(hostname)
+local hostname=$(cat $LOGDIR/boot.log | grep IPADDR | cut -d\'
-f2) +[ -z $hostname ] hostname=$(hostname)
frsh=$FAI_REMOTESH -l $LOGUSER ${SERVER}
Maybe there is a better (cleaner) solution.
I'd like to file a bug in BTS. Should I file it to fai or
fai-client?
Please file a bug for this. Then I will send a comment on it. IMO
using the ip address is a good solution since fai-chboot can use
them. But I may implement this a little bit different.
But first file a bug for this.
pgpPIIIiLZyXu.pgp
Description: PGP signature
Bug#349492: better documentation for other fb drivers
package: splashy severity: wishlist Hi, thanks for packaging splashy, it works great here :-) The FAQ on splashy.alioth.debian.org says - How can I use another fb driver other than VESA (or, My BIOS doesn't support vesafb) Instead of vga=0×314 use video=MYFB (where MYFB can be any of your kernel’s framebuffer drivers) in your bootloader’s configuration file. --- This (the instead) is not true here. I have to set both vga= and radeonfb= to get a working splash-screen while booting. video=radeonfb:[EMAIL PROTECTED] vga=795 is what I have in my /boot/grub/menu.lst. And, please put this information into /usr/share/doc/splashy/README(.Debian) as it's inconvinient to browse the web to learn about how to properly configure the package. (BTW, the README.Debian could also be cleaned up for better readability :) Also it would be nice if you would mention splashy_config in the README(.Debian) regards, Holger pgpHDoHewlJh0.pgp Description: PGP signature
Bug#349492: better documentation for other fb drivers
tag 349492 patch Hey, On Monday 23 January 2006 11:10, you wrote: Provide a patch for it, please. I think it's your duty to provide good documentation as you know the package best. (Our priorities are our users...) Nonetheless I've tried to come up with a patch, though I won't be surprised if you don't take it as I might have done mistakes and you know the package much better. Then my time would have been wasted (only 5min. but still) - that's why I think it's unreasonable to ask users of a software to provide documentation fixes. I hope we can move splashy to unstable as soon as possible. It miss some important bugs fixes yet but this should change fast. If those bugs are *important*, then please file them into the BTS! As this package is in experimental, of course I have looked into the BTS to see if there are important or serious problems. But I found none - should I worry now ? The social contract also states: we won't hide problems. regards, Holger --- splashy.README.Debian.h01ger2 2006-01-23 15:35:24.0 +0100 +++ splashy.README.Debian 2006-01-23 15:42:25.0 +0100 @@ -5,6 +5,17 @@ A boot splash program that doesn't require patching the Linux kernel. It paints graphic images directly to framebuffers using libdirectfb. +You might want to look at http://splashy.alioth.debian.org/wiki/doku.php?id=faq + + +Using other fb drivers +--- +For other fb drivers you need to both specify vga= and video=, for example like this: + +video=radeonfb:[EMAIL PROTECTED] vga=795 + + + Known Bugs: --- Image is too little work around: pgpJBixSF2zOk.pgp Description: PGP signature
Bug#315236: tag bugs pending
Hi, On Wednesday 25 January 2006 08:51, you wrote: You intend to adopt tuxtype? (This is just a ping, I'm not interested in adopting it myself.) Yes. I've tagged 8 bugs in tuxtype pending for a reason :-) The 1.5.3 package is almost ready, but I still want to polish it a bit, an upload (to mentors.d.n first) should happen on friday I would guess. I've also contacted upstream and got some feedback. regards, Holger pgpt13CofspKh.pgp Description: PGP signature
Bug#350333: raw1394 does not work on i386
package: linux-2.6 version: 2.6.15-3 severity: important Hi, raw1394 does not work with 2.6.15-3 on a Thinkpad R51, while it works with 2.6.12-1. The command to reproduce is dvgrab --size 0 -v foo (grab from a camera) which is followed by Error: no camera exists regards, Holger pgpt5VZzNRlXU.pgp Description: PGP signature
Bug#230217: should packages begin using /srv ?
block 340609 by 230217 block 340608 by 230217 block 311524 by 230217 block 315080 by 230217 block 336650 by 230217 thanks Hi, On Wednesday 21 December 2005 19:05, Adam D. Barratt wrote: On Wed, 2005-12-21 at 13:13 +0100, Holger Levsen wrote: [...] Anyway, shall I file a bug against policy now, to include FHS 2.3 instead of 2.1 ? I think I should. Otherwise it would look to me like the release team can simply overwrite policy decissions. Three such bugs already exist, one with quite a lot of discussion on the issue. These are merged and the most relevant one today is #230217, see http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=230217 The bug was opened two years ago, while the last post is four month old. Besides some, IMHO, smaller, problems, nobody objected and quite a few people seconded upgrading policy to FHS 2.3 - and _today_ we still have some time left til etch. So how about upgrading the severity or take some other measures (usertags for example), so that this bug will not be forgetten but fixed in time ? BTW, the fai maintainer said he will wait til this has been resolved before fixing #340609, #340608, #311524, #315080, #336650. (Which are serious (or should be) as they are against policy.) regards, Holger pgpRiJTgTPU4L.pgp Description: PGP signature
Bug#350967: patch doesn't apply to 2.6.15
package: kernel-patch-vserver version: 2:2.0.1-1 Hi, The patches in this packages dont apply to linux-source-2.6.15 :( Upstream has patches for 2.6.15 at http://vserver.13thfloor.at/Experimental/ ii kernel-patch-vserv 2.0.1-1context switching virtual private servers - kernel p ii linux-tree-2.6.15 2.6.15-3 Linux kernel source tree for building Debian kernel This is sid. regards, Holger pgpLW6uvu5ww7.pgp Description: PGP signature
Bug#350967: patch doesn't apply to 2.6.15
Hi Micah, On Thursday 02 February 2006 02:59, you wrote: tag 350967 +unreproducible I can reproduce it with make-kpkg clean time nice make-kpkg--append-to-version vserver --revision 01 --config menuconfig --initrd kernel_image though /# cd /usr/src /usr/src# tar -jxf linux-source-2.6.15.tar.bz2 /usr/src# cd linux-source-2.6.15 /usr/src/linux-source-2.6.15# cp /boot/config-2.6.15-rc4-k7 ./.config /usr/src/linux-source-2.6.15# export PATCH_THE_KERNEL=YES /usr/src/linux-source-2.6.15# make-kpkg --rootcmd fakeroot \ --revision custom01 \ --added-patches vserver \ --append-to-version +vserver \ --initrd \ binary-arch this (without make-kpkg clean) works perfectly :-/ Is this expected behaviour ? regards, Holger pgpxzIpb2Sl1T.pgp Description: PGP signature
Bug#230217: should packages begin using /srv ?
Hi, On Monday 30 January 2006 18:31, Holger Levsen wrote: BTW, the fai maintainer said he will wait til this has been resolved before fixing #340609, #340608, #311524, #315080, #336650. (Which are serious (or should be) as they are against policy.) FWIW those bugs are also present against FHS 2.1, but the maintainer thinks it's not good for his users, if he changes those paths now compliance with FHS 2.1 and then in some month again, for FHS 2.3. regards, Holger pgpCaxKE3Svdq.pgp Description: PGP signature
Bug#258437: more info
Hi, In the list A quick list of what to do if a package doesn't build on a buildd: I didn't mention to (to me) useful resources, because they are not official, do you think they should be included anyway: http://people.debian.org/~igloo/ http://buildd.debian.net regards, Holger pgpaTgtIDdQr8.pgp Description: PGP signature
Bug#258437: correct patch
Hi, sorry, this is the right patch. The first one has the order of changes reversed, that is added lines have a - and removed lines a +. regards, Holger diff -Nur developers-reference-3.3.7/common.ent developers-reference-3.3.7_258437-fixed/common.ent --- developers-reference-3.3.7/common.ent 2005-12-26 11:13:32.0 + +++ developers-reference-3.3.7_258437-fixed/common.ent 2006-06-17 16:24:57.0 + @@ -67,6 +67,7 @@ !ENTITY url-cvsweb http://cvs.debian.org/; !ENTITY url-devel-machines http://db.debian.org/machines.cgi; !ENTITY url-buildd http://buildd.debian.org/; +!ENTITY url-buildd-doc http://www.debian.org/devel/buildd/; !ENTITY url-lintian http://lintian.debian.org/; !ENTITY url-debian-qa http://qa.debian.org/; !ENTITY url-debian-qa-orphaned http://qa.debian.org/orphaned.html; diff -Nur developers-reference-3.3.7/developers-reference.sgml developers-reference-3.3.7_258437-fixed/developers-reference.sgml --- developers-reference-3.3.7/developers-reference.sgml 2006-04-04 21:41:26.0 + +++ developers-reference-3.3.7_258437-fixed/developers-reference.sgml 2006-06-17 16:27:16.0 + @@ -1550,7 +1550,14 @@ you don't forget any open bug, and so that you don't forget which packages are under your responsibility. -sect id=aliothDebian *Forge: Alioth +sect id=build-infoInformation about the buildd network + p +A package is usually uploaded as a binary package built for one architecture. +Once it's moved into the archive it gets build for those architectures which +are defined in ttdebian/control/tt. More details about what happens then +can be found in ref id=buildd. + + sect id=aliothDebian *Forge: Alioth p Alioth is a fairly new Debian service, based on a slightly modified version of the GForge software (which evolved from SourceForge). This software @@ -2863,22 +2870,14 @@ headingpackagebuildd/package/heading p The packagebuildd/package system is used as a distributed, -client-server build distribution system. It is usually used in +client-server build distribution system, which is described in +more detail at url id=url-buildd-doc;. It is usually used in conjunction with emauto-builders/em, which are ``slave'' hosts which simply check out and attempt to auto-build packages which need to be ported. There is also an email interface to the system, which allows porters to ``check out'' a source package (usually one which cannot yet be auto-built) and work on it. p -packagebuildd/package is not yet available as a package; however, -most porting efforts are either using it currently or planning to use -it in the near future. The actual automated builder is packaged as -packagesbuild/package, see its description in ref id=sbuild. -The complete packagebuildd/package system also collects a number of as yet unpackaged -components which are currently very useful and in use continually, -such as prgnandrea/prgn and -prgnwanna-build/prgn. - p Some of the data produced by packagebuildd/package which is generally useful to porters is available on the web at url id=url-buildd;. This data includes nightly updated information @@ -2894,8 +2893,17 @@ p The buildds admins of each arch can be contacted by the mail address [EMAIL PROTECTED] - - sect1 id=packages-arch-specificWhen your package is emnot/em portable + p +packagebuildd/package is not yet available as a package; however, +most porting efforts are either using it currently or planning to use +it in the near future. The actual automated builder is packaged as +packagesbuild/package, see its description in ref id=sbuild. +The complete packagebuildd/package system also collects a number of as yet unpackaged +components which are currently very useful and in use continually, +such as prgnandrea/prgn and +prgnwanna-build/prgn. + sect1 id=packages-arch-specificWhen your package is emnot/em + portable or does emnot/em build p Some packages still have issues with building and/or working on some of the architectures supported by Debian, and cannot be ported at all, @@ -2941,7 +2949,34 @@ If in the past some binary packages were uploaded on unsupported architectures, request their removal by filing a bug against packageftp.debian.org/package - + p +A quick list of what to do if a package doesn't build on a buildd: + list + item + p +Read the buildd logs, if they are available at url id=url-buildd; +(if there was no build attempt yet, there won't be any logs) + p + item +Wait a few days for the build attempt + p + item +Check the build statistics at url id=http://buildd.debian.org/stats/;; to see +if a port is falling behind + p + item +Check if missing build depends are going to be built later + p + item +Check if missing build depends are failing to build + p + item +Ask in the appropriate port mailing list for help/information + p + item +Try to build the package yourself on the arch which is
Bug#374392: detect and configure WPA networks
package: whereami severity: wishlist version: 0.3.28 Hi Andrew, it would be cool, if it was possible to detect and configure WPA networks with whereami too. Besides that, whereami still rocks my world! btw, there is a typo in the changelog entry for 0.3.28: unconfiugured regards, Holger pgpcJWvUBXOyd.pgp Description: PGP signature
Bug#369304: Quik uses OF input-device/output-device as console; on oldword this results in no visible outpu
Hi, On Friday 23 June 2006 18:01, Daniel Dickinson wrote: On Fri, Jun 23, 2006 at 11:42:59AM +0200, [EMAIL PROTECTED] wrote: I know. I think the best solution is to have the installer modify the input-device and output-device variables to point to keyboard/screen. I was thinking of doing this; I wanted to see if there was an easier way first. I'll be hacking on quik-installer next week to see about doing this. Another option would be, quik reads those values from /etc/quik.conf and sets them, while quik-installer writes quik.conf. Obviously this requires a version of quik first, that supports this. AIUI quik needs boot-device to be eg. /bandit/ohare/mesh/[EMAIL PROTECTED]:0 to boot the master boot record anyway (maybe this is model dependant as well?). At least my mac-clone doesn't boot the mbr unless I specify the boot-device. AIUI its OF which needs those settings. Anyway, also we need to keep in mind that not only that different models require different settings, but also they might have scsi or ide discs. That's a good idea. I also need a big scary warning with option to abort before modifying boot-device. I'm thinking that at the most detailed debconf priority there could be input fields for the user to specify input-device, output-device, and boot-device themselves (with defaults to what normally would happen). What do you think? Basically I agree. I would like to have sensible/possible choices in the input fields instead of free text fields :) Also we (d-i team) need to keep in mind, that some users dont want to use quik and modify OF values, as they already have a working BootX setup. But this surely doesn't belong into this bugreport :) regards, Holger pgpU8hPNPhcZ8.pgp Description: PGP signature
Bug#369304: settings for 4400
Hi, http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=274115 shows the OF settings an powermac 4400 needs. regards, Holger pgpmczQ6CyCAx.pgp Description: PGP signature
Bug#375208: ppc oldworld miboot-floppies daily-build status
package: installation-report Hi, I tested the daily-build from http://people.debian.org/~cjwatson/d-i/powerpc-miboot/2006-06-21/powerpc/floppy/ on an oldworld powermac 4400 with lowmen (48mb IIRC) and noticed some problems and a failed installation. The etch floppies from http://ftp.nl.debian.org/debian/dists/testing/main/installer-powerpc/beta2/images/ didnt work (boot-floppy didn't boot), I guess because they are build without the non-free miboot tool. It would be nice, if for beta3 we could point people to working images. (Which I haven't found for beta2.) After booting the boot.img from june 21st (which worked fine so #345467 can be closed) I noticed that I need to eject it manually, before I could insert the root-floppy. IIRC this is a regression, and we had ejecting boot-floppies. After writing a floppy image on a floppy, I compare them with /usr/bin/cmp. cmp: EOF on root.img is known for a while (#295996), whats new is that I get the same error with the cd-drivers.img now as well. The net-drivers.img is too big to fit on a floppy! :-( I downloaded and used the CD image I used from http://cdimage.debian.org/cdimage/daily-builds/daily/arch-latest/powerpc/iso-cd/debian-testing-powerpc-netinst.iso Next problem was the auto-partioning, in Partition disks: it defaults to floppy :( So far I avoided LVM and choose the non LVM setup. (So I cannot say if #267992 and #279118 are fixed) Then the installation runs without problems until quik-installer, this ends with an error because /boot is on a seperate partition (though quik supports it, the error-msg (typed not copied) is: the quik bootloader requires /boot to be on the same partition as / (#264815 are #266123 (essentially fixed, otherwise the above would not happen :) are related to this. ) Even when this is fixed, the machine will not boot, cause some OF variables need to be set, see #274115. So my todo-list is: 1. close #345467 (root.img doesnt work) 2. clone this bug to have one as a reminder to have beta3 miboot images somewhere - should I also reassign/usertag it? 3. clone this bug and rename to net-drivers.img to big to fit on a floppy 4. send mail to #295996 indicating that cd-drivers.img is know also affected 5. clone this bug, reassign to ??? and rename to partioning defaults to floppy drive when booting from floppy 6. clone this bug, reassign to quik-installer and rename to /boot on a seperate partition is ok, try to fix it 7. add powerpc usertag for user debian-boot@lists.debian.org to this bug and the cloned ones. 8. monitor #267992, #279118 (both lvm related) and #264815, #266123 (quik ext3 related), and #274115, #369304 (quik OF related) regards, Holger pgpMzWToQuOHf.pgp Description: PGP signature
Bug#370393: debian-edu doesnt use amanda anymore
Hi, closing this bug as debian-edu-config doesnt use amanda anymore. Sorry for the noise :-) (since debian-edu-config 0.402+svn6656) regards, Holger pgpxzrO2RVnzk.pgp Description: PGP signature
Bug#375653: fai-kernels: SATA driver AHCI is missing
Hi, On Tuesday 27 June 2006 14:22, Thomas Lange wrote: The AHCI SATA driver is missing in fai-kernels config. Please add CONFIG_SCSI_SATA_AHCI=m Thanks, will do in the next upload. (Which I plan to do before the fai developer weekend, where I plan to make fai-kernels deprecated :) regards, Holger pgplIkLHoA55S.pgp Description: PGP signature
Bug#274115: configure OF via quik.conf?
Hi, #369304 proposes to configure OF via quik.conf... regards, Holger pgpZscFGzBMkW.pgp Description: PGP signature
Bug#321820: console-switching
Hi, I cannot confirm your first bug, that is I can switch to another console and back without problems. So I would say this was a temporary problem. As you said, a seperate /boot partition is ok for quik, as long as it's on the same disk as / - retitling the bug and assigning to quik-installer. regards, Holger pgpMK5OPXOrUo.pgp Description: PGP signature
Bug#375208: ppc oldworld miboot-floppies daily-build status
Hi, On Saturday 24 June 2006 15:02, Holger Levsen wrote: So my todo-list is: 1. close #345467 (root.img doesnt work) done. Also closed #296782 2. clone this bug to have one as a reminder to have beta3 miboot images somewhere - should I also reassign/usertag it? I will retitle this on instead :) I'm not sure if important is an appropriate severity for this... reassigned to debian-installer also. About the actual problem, that we need miboot for floppy booting atm: #323182 - is the miboot bug with the most info in it. #315393 #315308 and #252426 (the only one of the four with severity important) can also be closed when this is fixed. (There are two proposed solutions for it: implement a free miboot alternative (in progress) or extend quik-funcionality (already implemented, but still buggy, doesnt work yet.) 3. clone this bug and rename to net-drivers.img to big to fit on a floppy done. 4. send mail to #295996 indicating that cd-drivers.img is know also affected Actually the bugs already stated that, so I just retitled that bug, as the other issues in this bugs are either solved or reported elsewhere. 5. clone this bug, reassign to ??? and rename to partioning defaults to floppy drive when booting from floppy I guess its partman-auto ?! (Not yet cloned.) 6. clone this bug, reassign to quik-installer and rename to /boot on a seperate partition is ok, try to fix it #321820 describes this behaviour as well (I cannot reproduce the first big problem the submitter mentioned, so I expect it to be a temporary problem and ignore it), so I retitle this bug and reassigned it to quik-installer. 7. add powerpc usertag for user debian-boot@lists.debian.org to this bug and the cloned ones. Done. 8. monitor #267992, #279118 (both lvm related) and #264815, #266123 (quik ext3 related), and #274115, #369304 (quik OF related) re: the first two bugs: If LVM works on ppc, they can be closed. The rest still needs more monitoring So I usertagged them as well... 9. Even when this is fixed, the machine will not boot, cause some OF variables need to be set, see #274115. #369304 proposes to fix this via storing the OF settings in quik.conf regards, Holger pgpKxwcE1PUvH.pgp Description: PGP signature
Bug#296782: root.img fits work now
Hi, http://people.debian.org/~cjwatson/d-i/powerpc-miboot/2006-06-21/powerpc/floppy/ works, so I close this bug. regards, Holger pgpCZs1PHPVts.pgp Description: PGP signature
Bug#345467: root.img fits work now
Hi, http://people.debian.org/~cjwatson/d-i/powerpc-miboot/2006-06-21/powerpc/floppy/ works, so I close this bug. regards, Holger pgpjfz6tZw4Z3.pgp Description: PGP signature
Bug#375697: cloning and usertagging in the same mail doesn't work
package: bugs.debian.org severity: wishlist Hi, If I clone a bug, I cannot usertag the clone in the same mail to [EMAIL PROTECTED], though this would be logical and useful :) From control: On Tuesday 27 June 2006 17:48, you wrote: Processing commands for [EMAIL PROTECTED]: clone 375208 -1 Bug#375208: provide beta3 (and rc1++) miboot images til they're obsolete Bug 375208 cloned as bug 375684. user debian-boot@lists.debian.org Setting user to debian-boot@lists.debian.org (was [EMAIL PROTECTED]). usertag 375208 + powerpc There were no usertags set. Usertags are now: powerpc. usertag -1 + powerpc There were no usertags set. Usertags are now: powerpc. But 375684 wasn't usertagged :( (I did that manuelly now.) regards, Holger pgpjr0zVin8lE.pgp Description: PGP signature
Bug#370350: fixed
Hi, from debian-edu-config (0.402+svn6656) changelog: * Remove obsolete cf.udev as we are configuring the rights for the sound device via pam (Closes: #370350) regards, Holger pgp15tuOkn4mt.pgp Description: PGP signature
