RFC: Moving /tmp/target to /target

Hi all, I need your opinions about changing the directory /tmp/target to /target. d-i is using /target. FAI ist using /tmp/target, because in the past we could not make /target writeable for the clients, which is needed when mounting the file systems from every partition to a target destination.

Re: New features, and bugfix for faimond

On Fri, 24 Aug 2007 15:34:51 +0200, Jonas Eriksson [EMAIL PROTECTED] said: faimond, bad clients-patch: * Add a timeout, so that bad clients that are locking up the faimond will be disconnected after a given timeout. I see this as a bugfix since bad clients may lock up

Re: Problem with script 'make-fai-nfsroot'

On Fri, 24 Aug 2007 16:25:02 +0200, Michael Tautschnig [EMAIL PROTECTED] said: Thomas, I strongly suggest either reverting the patch of dropping the dependency on libapt-pkg-perl or properly working around the missing library. I think, there is also no note on that in the

Re: [Fai-commit] r4552 - in trunk: bin man

On Thu, 6 Sep 2007 14:26:23 +0200, Michael Tautschnig [EMAIL PROTECTED] said: I've no idea about the specifics of Ubuntu, but can't we just detect, whether the system we are running on is Ubuntu? (something similar to /etc/debian_version maybe) We discussed that yesterday on

Re: [Fai-commit] r4552 - in trunk: bin man

On Thu, 06 Sep 2007 16:30:26 +0200, Henning Sprang [EMAIL PROTECTED] said: The nfsroot on etch is large anyway(last time I looked at it was over 500MB, right?!), additional 20MB(just a guess) are not really doing much No, it's only 330MB. I will not add another 20MB or so (IIRC more

Re: make-fai-nfsroot should check for live-initramfs at the begining

On Sat, 17 Nov 2007 23:20:30 +0100, Henning Sprang [EMAIL PROTECTED] said: Hmm, with the newest fai 3.2beta10, this even doesn't change when live-initramfs is installed - output at the end of make-fai-nfsroot -v is: ERROR: live-initramfs is not installed. Aborting

Re: gpl header copyright bugfix for my old contributions ready to be merged

On Wed, 4 Jun 2008 11:27:33 +0200, Henning Glawe [EMAIL PROTECTED] said: @Thomas: could you merge the people/eartoast/fix-copyright branch? Hi Henning, in my opinion it's difficult to say when a patch is worth that a copyright message should be included. All people that are listed in the

Re: stable-proposed-updates?

On Fri, 20 Feb 2009 11:46:28 +0100, Holger Levsen hol...@layer-acht.org said: Ok, thanks. Added to svn/trunk and svn/branches/lenny as I pretty much assume Thomas is fine with that and if not I can easily revert ;-) The patch is ok. holger, thanks for doing the s-p-u stuff. --

Re: Huge number of lintian messages

On Fri, 1 May 2009 16:19:12 +0200, Michael Tautschnig m...@debian.org said: E: fai-client: dir-or-file-in-var-run var/run/fai/ that one needs investigation. /var/run is not guaranteed to keep files after a reboot as it may be on a ramdisk. E:

Re: Huge number of lintian messages

On Fri, 1 May 2009 14:49:10 +0200, Holger Levsen hol...@layer-acht.org said: W: fai source: package-uses-deprecated-debhelper-compat-version 4 uhm, yes, should be set to 5 (to ease backporting to etch) Already set to 7. W: fai source: out-of-date-standards-version 3.8.0

Cleanup of FAI svn directory

Hi FAI developers, I ask you to have a look at your branches in the FAI svn repository and do a cleanup of old unused directories (e.g. eartoast/fix-copyright, faiguy/bugfix/480830,...). It's easier for others to find relevant information if obsolete data is removed. You should also have a look

FAI Workshop 2010

Es ist wieder mal Zeit fuer einen FAI Workshop. Von Freitag 2.Juli bis zum Sonntag 4.Juli 2010 koennen wir im Linuxhotel in Essen wieder eifrig an FAI arbeiten. Zielgruppe fuer dieses Treffen sind Leute, die schon Erfahrung mit FAI gemacht haben und bereit sind an dem Wochenende an FAI

FAI workshop 2010 wiki page

Hi all, I've set up the wiki page for the next FAI workshop. I tried to keep the information short, because noone likes to read very long text. But feel free to add more information. The wiki page is http://faiwiki.informatik.uni-koeln.de/index.php/DeveloperWorkshopJuly2010 Currently we do not

debootstrap for RPM based distributions

Just for the archive. There's a tool called rinse (part of xen-tools), which can created a minimal chroot environment for RPM based distributions. This may be usefull for creating the basefiles for FAI. http://xen-tools.noone.org/software/rinse/ -- regards Thomas

Re: make /var/lib/dpkg a ramdisk :(

On Tue, 18 May 2010 17:28:38 +0200, Juergen Kahnert juergen.kahn...@desy.de said: any idea why this: # make /var/lib/dpkg a ramdisk mkramdisk $target/var/lib/dpkg was added into /usr/lib/fai/subroutines? This makes the installation much faster (on machine which have

FAI workshop

Hi all, again this year, we will held a FAI workshop meeting from july 2-4 at the Linuxhotel (http://www.linuxhotel.de) in Essen, Germany. People which already use FAI meet there to work the whole weekend on improving the FAI software like we did in the past. If you have experiences with FAI

should fai-client depend on fai-setup-storage?

I'm wondering if fai-client should really depend on fai-setup-storage? Instead, I would prefer it recommends fai-setup-storage and maybe fai-nfsroot depends on fai-setup-storage. Does this make sense? -- regards Thomas

FAI roadmap

I've created a little roadmap for the FAI development. It includes some topics I like to work on for the next release and a brief timeline. Comments are welcome. http://www.informatik.uni-koeln.de/fai/roadmap -- regards Thomas

Re: FAI roadmap

On Tue, 20 Jul 2010 16:20:53 +0100, Michael Tautschnig m...@debian.org said: - Please make that DHCP-transition-bug top priority. We really shouldn't be blocking transitions. (Well, they'll ignore FAI anyway, but it's just such a small change that only requires proper

Re: FAI roadmap

On Tue, 20 Jul 2010 16:43:00 +0100, Joel Merrick joel.merr...@gmail.com said: Looks good Thomas, one thing I'd like to ask about though is IPv6 support. I guess this is not a FAI problem. I realise that there's a lack of PXE booting in IPv6 (very Yep. My guess is that even in three

Re: FAI roadmap

On Tue, 20 Jul 2010 17:37:06 +0100, Michael Tautschnig m...@debian.org said: I'd propose something as simple as checking for -d /etc/dhcp3 or -d /etc/dhcp (just writing that off the top of my head, paths may be wrong) and act accordingly!? I might be missing something and

Re: FAI roadmap

On Tue, 20 Jul 2010 17:42:13 -0400, Holger Levsen hol...@layer-acht.org said: IMHO it's - as Julien said - quite risky to not have fai in testing until mid september, that could very well be after the freeze. And it's quite likely that new packages will not be allowed into

Re: FAI roadmap

On Wed, 21 Jul 2010 09:23:56 +0100, Michael Tautschnig m...@debian.org said: While technically possible, it's not a very nice attitude to plainly ignore the freeze. There is no freeze yet which I ignore. And during the next two weeks I will work a lot on FAI. I just need to

Re: FAI roadmap

On Fri, 23 Jul 2010 10:43:42 +0200, Michael Tautschnig m...@debian.org said: fai-nfsroot.install: etc/dhcp3/dhclient-fai-script fai-nfsroot.install: etc/dhcp3/dhclient-fai.conf But, well, has anybody out there actually looked at what truly will be broken with the

Re: FAI roadmap

On Fri, 23 Jul 2010 11:20:16 +0200, Michael Tautschnig m...@debian.org said: Sure - but the DHCP issue wasn't/isn't on your roadmap. But if you accept the proposed patch it needn't go there, it's done already :-) Not really. If the patch is in trunk, there's still a package upload

Re: FAI roadmap

On Fri, 23 Jul 2010 11:51:51 +0200, Michael Tautschnig m...@debian.org said: @Thomas: If I get your ACK I will, after coordinating with Mika, push it into trunk. Maybe you could also voice your opinion about the desired strategy for the simple examples, as I proposed two in

Re: Bug#595365: Intent to 0-day NMU FAI 3.4.1 to fix #595365

On Wed, 15 Sep 2010 17:29:37 +0200, Julien BLACHE jbla...@debian.org said: I intend to prepare, test and upload an NMU of FAI 3.4.1 tomorrow to finally beat FAI into shape for Squeeze (fixing #595365). I see no reason why to do an NMU. Mika is on vacation until tomorrow, and he will

FAI $HOME has moved

For more than 11 years now, FAI (Fully Automatic Installation) lived happily at http://www.informatik.uni-koeln.de/fai. Now that it's grown up, it finally deserves its own home at http://fai-project.org. This also means less typing for all of FAI's friends. Homepage old:

Re: TEST: fai libeatmydata

On Tue, 23 Nov 2010 18:13:33 +0100, Julien BLACHE jbla...@debian.org said: We went from 45 minutes down to somewhere around 22-25 minutes. Wow! This was done by adding libeatmydata to LD_PRELOAD in the instsoft hook Great idea. Currently FAI has only the ramdisk on top of

Re: Patch for fcopy

On Sat, 4 Dec 2010 10:57:06 +0100, Michael Tautschnig m...@debian.org said: Thank you very much for this contribution and sorry for taking so long to respond. Thanks Michael for adding this patch. I also like to add the patch to FAI, but had no time in the past. I took the

Re: device2grub

On Thu, 10 Feb 2011 23:15:43 +0100, Michael Goetze mgoe...@mgoetze.net said: Therefore, I propose to make device2grub independent of $target. An example implementation can be found below. I checked this patch on all my machines. The new version of device2grub work fine for all of

FAI workshop at Linuxhotel

Hi all, the next FAI workshop will be held from Friday 21 October to Sunday 23 October 2011 at the Linuxhotel in Essen, Germany. This year, we will focus on preparing the 4.0 release. The necessary work focusses on testing, updating documentation, fixing bugs, and getting the Ubuntu package in a

Re: reporting a documentation bug and couple of questions

On Mon, 30 Jan 2012 16:35:11 +0100, Thomas Neumann blacky+...@fluffbunny.de said: I found a documentation bug in http://fai-project.org/fai-guide/ar01s07.html . (What is the proper way to report this bug?) Use the Debian BTS. (I LOVE THE HOOKS! THANK YOU FOR IMPLEMENTING

Re: Error: at /usr/share/fai/setup-storage//Parser.pm line 382

On Wed, 04 Sep 2013 09:29:16 +0200, Karl Koch cypher_h...@gmx.net said: Good Morning, is the same Problem with 'C': No, it not the same problem. (CMD) parted -s /dev/sda unit TiB print 1 /tmp/0MMJ46TdOt 2 /tmp/ig5grBuJds Executing: parted -s /dev/sda unit TiB print (STDOUT)

fai-project package repository is now signed only with my new gpg key

-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 As I announced in november, I have a new gpg key. https://lists.uni-koeln.de/pipermail/linux-fai/2013-November/010220.html Since last week the fai-project package repository is signed only with my new key. Before I signed it with my old and new

FAI release for Debian Jessie

Hi folks, currently we are working on FAI 4.3 which should be uploaded to unstable until mid of october, so it will enter testing before the release freeze of Debian Jessie. If you have any wishes for this next release, please feel free to post it to this mailing list. We also welcome any

Re: Origin location of pxelinux.0 inside fai-make-nfsroot

On Mon, 06 Oct 2014 13:10:22 -0700, Ronan Keryell ro...@keryell.fr said: I have an issue with fai-make-nfsroot around lines 492--508 : It is not clear to me why pxelinux.0 is fetched from $NFSROOT instead of the local file system? Yes, that's not clean. - it is more

Re: RFC: How to improve Ubuntu support

On Tue, 04 Nov 2014 09:37:28 +0100, Moritz Struebe moritz.stru...@informatik.uni-erlangen.de said: Currently I see the following ways of improving things: * Get the FAI-Project at launchpad[1] up to date. Nice. I will contact someone, that can give you write access to the launchpad

Re: RFC: How to improve Ubuntu support

On Thu, 06 Nov 2014 10:16:30 +0100, Moritz Struebe moritz.stru...@informatik.uni-erlangen.de said: I think we can take care of almost all issues at packaging time as described in [3]. Nice link. I didn't know that this was possible. What are the current diffs in the packaging files

Re: RFC: How to improve Ubuntu support

On Fri, 07 Nov 2014 17:44:32 +0100, Moritz Struebe moritz.stru...@informatik.uni-erlangen.de said: What are the current diffs in the packaging files for Ubuntu? For my PPA none and the packages build and install just fine - the only But isn't it much more comfortable, if the Ubuntu

Re: RFC: How to improve Ubuntu support

On Fri, 07 Nov 2014 17:44:32 +0100, Moritz Struebe moritz.stru...@informatik.uni-erlangen.de said: described in [3]. And if that is not sufficient rather add a configuration file that is adjusted before building. What about a ubuntu/patches subdir, which contains the patches needed

Re: [PATCH] Tiny build patch

On Thu, 13 Nov 2014 13:44:42 +0100, Moritz Struebe moritz.stru...@cs.fau.de said: Here's a tiny patch fixing the debhelper-but-no-misc-depends warning. Thanks. Already applied. I push this soon. -- regards Thomas

Re: dracut in ubuntu (Was: RFC: How to improve Ubuntu support)

On Mon, 17 Nov 2014 10:13:51 +0100, Moritz Struebe moritz.stru...@cs.fau.de said: Hmm, but only in testing. And vivid is there, too: https://launchpad.net/dracut So, for Ubuntu I see two options. Add a ppa or contact Harald Hoyer [1]. But's that's probably something you

Re: Add new package: fai-make-nfsroot

On Wed, 26 Nov 2014 09:15:22 +0100, Moritz Struebe moritz.stru...@cs.fau.de said: I'd like to add a new package fai-make-nfsroot. Rationale: In our system we have one central tftp and nfs server. For the sake of clean systems we run fai-make-nfsroot on a different machine.

Re: New Ubuntu PPAs

On Tue, 10 Mar 2015 10:12:38 +0100, Moritz Struebe moritz.stru...@cs.fau.de said: I updated/created two PPAs at Launchpad. The first contains Launchpad-builds of the latest FAI-releases. https://launchpad.net/~fai/+archive/ubuntu/ppa Hi morty, thanks for this great work. --

Re: Ubuntu, Vivid: Dracut

On Wed, 29 Apr 2015 11:10:39 +0200, Moritz Struebe moritz.stru...@cs.fau.de said: We are currently trying to getting Ubuntu installed, but getting the dracut package installed seems a PITA. I already made some adjustments[2], but I'm neither familiar with dracut nor a packaging

Re: Ubuntu support / Redesign

On Wed, 07 Jan 2015 09:16:55 +0100, Moritz Struebe moritz.stru...@cs.fau.de said: did you have time to have a look at my changes [1]? In the long run they should also save you from having to make separate releases, e.g. for wheezy. [1] https://github.com/cmorty/fai/tree/dev Hi

FAI 5.0 release candidate

We are very happy to announce the release candidate of FAI 5.0! We plan to have a two week period of testing before the final release will be made. We still work on the FAI guide, but it has yet changed a lot and was updated for the new FAI version. Please give FAI 5.0 a try and give us feedback

Re: Do you want patches here or github or some other way?

> On Fri, 01 Jan 2016 16:27:18 -0800, Ian Kelling > said: > Re: Do you want patches here or github or some other way? You can do both. For small patches, use an email to this list, or to f...@fai-project.org. If a patch fixes a bug, please file a bug report to the

Re: [PATCH] Improve fai-cd docs

Hi Ian, thanks for the two document patches. I've applied both and will also look into you bug report soon. -- regards Thomas

Re: split example configspace to submodule

> On Tue, 18 Jul 2017 18:44:43 +0200, Juri Grabowski > said: > it's no needs to split config to separate debian package. I mean > something like I don't understand what this does. But, I'm not a git expert and never used the submodule command of git. --

Re: split example configspace to submodule

Hi Juri, sorry, aber ich verstehe immer noch nicht was du mit den git submodulen willst. Du kannst mich gerne mal anrufen, vielleicht ist das am Telefon besser zu erklaeren. -- viele Gruesse Thomas -- Thomas Lange

Re: PATCH: loading vars for fai-diskimage

> On Fri, 26 May 2017 12:24:04 +0200 (CEST), Ondrej Jombik > said: > I noticed when using fai-diskimage: variables from $hostname.var are not > properly loaded. > Variables from $CLASSNAME.var are loaded -- this works OK. > This small patch fixes that.

Re: linux-fai-devel Digest, Vol 86, Issue 2

> On Wed, 13 Sep 2017 18:35:29 +0200, Emmanuel Kasper > said: > I suppose we also need to rebuild the STRETCH64 base in case you haven't > done it yet Just done it and put it onto the web. Thanks to reminding me. -- regards Thomas

Re: fai-diskimage and STRETCH64.tar.xz

> On Tue, 12 Sep 2017 11:39:49 +0200, Emmanuel Kasper > said: > in > https://github.com/faiproject/fai/blob/master/examples/simple/basefiles/mk-basefile#L98 > we delete the /etc/machine-id file for debian based VMs It's now fixed. > the question: >

Re: hooks/updatebase.UBUNTU break fai softupdate

> On Sun, 24 Jun 2018 16:50:01 +0200, Achim Bohnet > said: > the script hooks/updatebase.UBUNTU uses curl & netselect. > Unfortunately netselect package is not available in the ubuntu repos and > therefore fai softupdate creates an invalid sources.list (without repo URL)

Re: [PATCH] fai-cd: fix empty expansion causing superflous 1 exit

Hi Ian, I've applied your patches to the FAI sources. Thanks for your work. -- regards Thomas