On Tue, Jun 03, 2014 at 07:49:08PM +0200, maximilian attems wrote:
> thank you very much, pushed out to initramfs repo:
> http://anonscm.debian.org/gitweb/?p=kernel/initramfs-tools.git
>
> will upload soonish.
Hi. It's two months later and people (e.g. me!) are being bitten by
this bug when up
Hi,
> You are misunderstanding. If /usr is on a separate file system and we
> mount that from within the initramfs [1], we don't need to worry about
> using resources from /usr
>
> > Apart from that, there is no /usr in the initramfs.
>
> I didn't say that.
Ah. I misread your idea as bind-mount
Am 04.06.2014 07:54, schrieb Dominik George:
>> While the attached patch looks correct, I'd very much prefer if we could
>> just mount /usr from the initramfs, which would make this issue obsolete.
>
> Uh? Not really… remember, we are just goin to check whether we got the
> *rootfs* mounted, we c
> While the attached patch looks correct, I'd very much prefer if we could
> just mount /usr from the initramfs, which would make this issue obsolete.
Uh? Not really… remember, we are just goin to check whether we got the
*rootfs* mounted, we can hardly go mount something to /usr for that.
Apart
Am 03.06.2014 22:10, schrieb Sven Joachim:
> I guess the following should work (untested):
>
> --8<---cut here---start->8---
> diff --git a/scripts/nfs b/scripts/nfs
> index 967e67f..3b7ade2 100644
> --- a/scripts/nfs
> +++ b/scripts/nfs
> @@ -67,7 +67,7 @@ mou
On 2014-06-03 21:30 +0200, maximilian attems wrote:
> On Tue, Jun 03, 2014 at 08:03:16PM +0200, Sven Joachim wrote:
>>
>> Is this test going to work for systems with a separate /usr ?
>>
>> ,
>> | $ which test
>> | /usr/bin/test
>> `
>>
>
> dash or bash or POSIX shells have builtin test
On Tue, Jun 03, 2014 at 08:03:16PM +0200, Sven Joachim wrote:
>
> Is this test going to work for systems with a separate /usr ?
>
> ,
> | $ which test
> | /usr/bin/test
> `
>
dash or bash or POSIX shells have builtin test too.
so yes it should just work.
--
maks
--
To UNSUBSCRIBE,
On 2014-06-03 19:49 +0200, maximilian attems wrote:
> Hello Michael!
>
> On Tue, Jun 03, 2014 at 06:19:23PM +0200, Michael Biebl wrote:
>>
>> On Tue, Jun 03, 2014 at 11:36:25AM +0200, maximilian attems wrote:
>> > in any case, did you test it with klibc chroot too?
>> > meaning disabling BUSYBOX
Hello Michael!
On Tue, Jun 03, 2014 at 06:19:23PM +0200, Michael Biebl wrote:
>
> On Tue, Jun 03, 2014 at 11:36:25AM +0200, maximilian attems wrote:
> > in any case, did you test it with klibc chroot too?
> > meaning disabling BUSYBOX for the initramfs.
>
> I quickly tested /usr/lib/klibc/bin/ch
Hi maks,
On Tue, Jun 03, 2014 at 11:36:25AM +0200, maximilian attems wrote:
> in any case, did you test it with klibc chroot too?
> meaning disabling BUSYBOX for the initramfs.
I quickly tested /usr/lib/klibc/bin/chroot, and that worked fine.
So if the udeb versions aren't fundametally different,
On Tue, Jun 03, 2014 at 12:00:32PM +0200, Dominik George wrote:
> > no need of nmu games, I am following this.
>
> Huh? I build a local version by using dch -n, which appends an NMU
> version, and then I use debdiff to create a patch.
>
> If you feel offended by a standard workflow, then bring in
> no need of nmu games, I am following this.
Huh? I build a local version by using dch -n, which appends an NMU
version, and then I use debdiff to create a patch.
If you feel offended by a standard workflow, then bring in your own fix
:)!
--
Dominik George (Vorstandsvorsitzender)
Teckids e.V. (
On Tue, Jun 03, 2014 at 11:17:40AM +0200, Dominik George wrote:
> Control: tags -1 + patch
>
> > initramfs-tools could just use a check like
> >
> > chroot /root test -x /sbin/init
> >
> > Then it doesn't matter whether it's a relative or absolute symlink.
>
> I made the needed fix in the scrip
Control: tags -1 + patch
> initramfs-tools could just use a check like
>
> chroot /root test -x /sbin/init
>
> Then it doesn't matter whether it's a relative or absolute symlink.
I made the needed fix in the script and tested in our terminal server
environment. It works as expected and I do not
Processing control commands:
> tags -1 + patch
Bug #750360 [initramfs-tools] initramfs-tools: nfs mount script check for
/sbin/init incomplete
Added tag(s) patch.
--
750360: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=750360
Debian Bug Tracking System
Contact ow...@bugs.debian.org with pro
Hi,
> initramfs-tools could just use a check like
>
> chroot /root test -x /sbin/init
>
> Then it doesn't matter whether it's a relative or absolute symlink.
Not bad, for a systemd maintainer ;) ... *SCNR*
(don't take me serious, I like systemd :)
So I will send a patch with that easy fix; gr
Am 03.06.2014 08:47, schrieb Dominik George:
> Control: reassign -1 initramfs-tools 0.115
> Control: retitle -1 initramfs-tools: nfs mount script check for /sbin/init
> incomplete
>
> https://www.debian.org/doc/debian-policy/ch-files.html#fr91
>
> Ok, the explanation here changes things a bit. I
Processing control commands:
> reassign -1 initramfs-tools 0.115
Bug #750360 [systemd-sysv] systemd-sysv: breaks NFS root systems
Bug reassigned from package 'systemd-sysv' to 'initramfs-tools'.
No longer marked as found in versions systemd/204-10.
Ignoring request to alter fixed versions of bug #
18 matches
Mail list logo
