Package: debootstrap
Version: 1.0.87
Followup-For: Bug #817236
(Not CCing people on purpose because this doesn't add to the current thread's
discussion:)
Just wanted to add for the record (e.g. online readers) that Ansgar's
workaround of replacing the current fstab /dev/pts line does works to unb
Package: schroot
Version: 1.6.10-2
Severity: important
debootstrap recently replaced the /dev/ptmx device node with a symlink
/dev/ptmx -> pts/ptmx[1]. This changed the default permissions from
world-writable (0666) for /dev/ptmx to no-access () in chroots[2].
This causes build failures as, f
On Mar 09, Ansgar Burchardt wrote:
> in the sbuild profile's fstab. To accommodate old chroots that still
> have a /dev/ptmx device node, an additional bind mount
>
> ${chroot}/dev/pts/ptmx /dev/ptmx none rw,bind 0 0
>
> is needed to make sure only the devpts' ptmx device is used. I don
On Wed, 2016-03-09 at 11:41 +0100, Marco d'Itri wrote:
> On Mar 09, Ansgar Burchardt wrote:
>
> > in the sbuild profile's fstab. To accommodate old chroots that
> > still
> > have a /dev/ptmx device node, an additional bind mount
> >
> > ${chroot}/dev/pts/ptmx /dev/ptmx none rw,bind 0 0
On Wed, 09 Mar 2016 10:02:14 +0100 Ansgar Burchardt
wrote:
> Package: schroot
> Version: 1.6.10-2
> Severity: important
>
> debootstrap recently replaced the /dev/ptmx device node with a symlink
> /dev/ptmx -> pts/ptmx[1]. This changed the default permissions from
> world-writable (0666) for /dev
Ben Hutchings (2016-11-07):
> On Wed, 09 Mar 2016 10:02:14 +0100 Ansgar Burchardt
> wrote:
> > Package: schroot
> > Version: 1.6.10-2
> > Severity: important
> >
> > debootstrap recently replaced the /dev/ptmx device node with a symlink
> > /dev/ptmx -> pts/ptmx[1]. This changed the default perm
reassign 817236 debootstrap 1.0.85
reassign 841935 debootstrap
forcemerge 817236 841935
affects 817236 pbuilder sbuild schroot
stop
On Sun, Nov 20, 2016 at 08:03:25AM +0100, Cyril Brulebois wrote:
> Ben Hutchings (2016-11-07):
> > On Wed, 09 Mar 2016 10:02:14 +0100 Ansgar Burchardt
> > > deboots
On Sun, 26 Feb 2017 at 14:31:31 +, Simon McVittie wrote:
> * schroot/1.6.10-3:
> - sbuild profile: bind-mounts host /dev/pts only
> + in the real device node case, we're fine
Actually no - running script inside schroot inside an lxc container on a
jessie kernel, with lxc from jessie-back
On Sun, 26 Feb 2017 at 14:31:31 +, Simon McVittie wrote:
> On Wed, 15 Feb 2017 at 11:13:39 +, Simon McVittie wrote:
> > If debootstrap inside a container is meant to work, then this would seem
> > like a job for an autopkgtest. I'll try to write one if someone can tell me
> > which containe
I have filed #856877 against schroot suggesting mounting a new instance
of /dev/pts, effectively making schroot behave less like chroot and more
like a container manager in this particular respect.
However, that causes a nasty regression for interactive use (ttyname()
won't report that the interac
On Sun, 05 Mar 2017 at 19:41:14 +, Simon McVittie wrote:
> I have filed #856877 against schroot suggesting mounting a new instance
> of /dev/pts, effectively making schroot behave less like chroot and more
> like a container manager in this particular respect.
>
> However, that causes a nasty
Hi,
Simon McVittie (2017-03-06):
> On Sun, 05 Mar 2017 at 19:41:14 +, Simon McVittie wrote:
> > I have filed #856877 against schroot suggesting mounting a new instance
> > of /dev/pts, effectively making schroot behave less like chroot and more
> > like a container manager in this particular
On Mar 09, Cyril Brulebois wrote:
> If nobody (esp. Ben & Marco) objects to them, I guess you could push
> them to master once alioth is back?
I think Simon did a great job analyzing this, so I fully support merging
his patch.
--
ciao,
Marco
signature.asc
Description: PGP signature
Marco d'Itri (2017-03-09):
> On Mar 09, Cyril Brulebois wrote:
>
> > If nobody (esp. Ben & Marco) objects to them, I guess you could push
> > them to master once alioth is back?
> I think Simon did a great job analyzing this, so I fully support merging
> his patch.
Thanks, I've just done so. H
On Fri, 10 Mar 2017 at 13:21:37 +0100, Cyril Brulebois wrote:
> Marco d'Itri (2017-03-09):
> > I think Simon did a great job analyzing this, so I fully support merging
> > his patch.
>
> Thanks, I've just done so. Hopefully the changelog entry I added is
> neither too short nor too inaccurate.
On Sat, 26 Nov 2016 at 10:27:38 +, Ben Hutchings wrote:
> On Fri, 2016-11-25 at 18:54 +0100, Marco d'Itri wrote:
> > setup_devices_simple() could be easily modified to create /dev/ptmx with
> > mknod instead of the current symlink.
That's presumably:
- ln -s pts/ptmx $TARGET/dev/ptmx
+
On Wed, 2017-02-15 at 11:13 +, Simon McVittie wrote:
> On Sat, 26 Nov 2016 at 10:27:38 +, Ben Hutchings wrote:
[...]
> > The temporary workaround with /dev/ptmx could be made optional. It's
> > not OK to break the previously working configurations.
>
> If I'm understanding the situation c
Control: tags 817236 + patch
On Wed, 15 Feb 2017 at 11:13:39 +, Simon McVittie wrote:
> If debootstrap inside a container is meant to work, then this would seem
> like a job for an autopkgtest. I'll try to write one if someone can tell me
> which containerization technologies (systemd-nspawn?
On Nov 20, Cyril Brulebois wrote:
> > This is not needed at all from Linux 4.7. The open operation on
> > /dev/ptmx automatically looks up the sibling pts/ directory. (Also,
> > every mount of devpts is a 'new instance'.)
> >
> > It seems to me that the change in debootstrap ought to be revert
On Fri, 2016-11-25 at 18:54 +0100, Marco d'Itri wrote:
> > On Nov 20, Cyril Brulebois wrote:
>
> > > This is not needed at all from Linux 4.7. The open operation on
> > > /dev/ptmx automatically looks up the sibling pts/ directory. (Also,
> > > every mount of devpts is a 'new instance'.)
> > >
20 matches
Mail list logo