ifupdown: postinst fails if /dev/shm/network/ does not exist

Bug #14927 reported by Debian Bug Importer
4
Affects Status Importance Assigned to Milestone
ifupdown (Debian)
Fix Released
Unknown
ifupdown (Ubuntu)
Fix Released
High
Adam Conrad

Bug Description

Automatically imported from Debian bug report #302519 http://bugs.debian.org/302519

Revision history for this message
In , Thomas Hood (jdthood-aglu) wrote : Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does not exist

On Fri, 2005-04-01 at 10:45 +0200, Stefan Kluth wrote:
> Setting up ifupdown (0.6.4-4.12) ...
> ifupdown.postinst: Error: The canonical path of /etc/network/run could
> not be determined. Aborting.

What version of the initscripts package do you have?

--
Thomas Hood <email address hidden>

Revision history for this message
In , Stefan Kluth (skluth) wrote : Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does not exist

On Fri, 1 Apr 2005, Thomas Hood wrote:

> On Fri, 2005-04-01 at 10:45 +0200, Stefan Kluth wrote:
> > Setting up ifupdown (0.6.4-4.12) ...
> > ifupdown.postinst: Error: The canonical path of /etc/network/run could
> > not be determined. Aborting.
>
>
> What version of the initscripts package do you have?

skluth@sixpack:~$ apt-cache show initscripts
Package: initscripts
Priority: required
Section: base
Installed-Size: 240
Maintainer: Miquel van Smoorenburg <email address hidden>
Architecture: i386
Source: sysvinit
Version: 2.86.ds1-1
Replaces: sysvinit (<< 2.85-12), libc6, libc6.1, libc0.1, libc0.3
Depends: libc6 (>= 2.3.2.ds1-4), libc6 (>= 2.3.2.ds1-12), dpkg (>= 1.9.0),
mount (>= 2.11l-1), util-linux (>= 2.9t-2), e2fsprogs (>= 1.19-1),
coreutils (>= 5.0.91-2)
Conflicts: kbd (<< 0.95-2), mdutils (<< 0.35-9)
Filename: pool/main/s/sysvinit/initscripts_2.86.ds1-1_i386.deb
Size: 30454
MD5sum: 74882efd4ad50da4dbc9020fe0dd67f6
Description: Standard scripts needed for booting and shutting down
 These scripts are meant for standard Debian/GNU/Linux installations.

I run testing derived some time ago from knoppix. I keep the system
uptodate with regular (every few days) updates.

Stefan

---Stefan Kluth, PhD----------Wissenschaftlicher Mitarbeiter-----------
- MPI fuer Physik - phone: +49 89 32354 468 - OPAL& -
- Foehringer Ring 6 - fax: +49 89 32354 305 - ATLAS -
---D-80805 Munich, Germany----e-mail: <email address hidden>--------------

Revision history for this message
In , Thomas Hood (jdthood-aglu) wrote : Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does not exist

On Fri, 2005-04-01 at 10:45 +0200, Stefan Kluth wrote:
> Setting up ifupdown (0.6.4-4.12) ...

>From which version did you upgrade?

> ifupdown.postinst: Error: The canonical path of /etc/network/run could
> not be determined. Aborting.

Do you have the /lib/init/readlink program? Please send the output of
"ls -l /lib/init/readlink".

Please do "grep PATH /var/lib/dpkg/info/ifupdown.postinst".

It will be helpful if you can reproduce the failure. Please remove the
package with "dpkg --remove --force-depends" and reinstall it. (Make
sure you have the package available locally so you don't need to access
the network in order to reinstall.)

You may have to install the version of ifupdown you had before, and
upgrade to 0.6.4-4.12, in order to reproduce the problem.

If the postinst fails again then please send the output of

    ls -l /etc/network
    ls -l /etc/network/run
    ls -l /dev/shm
    ls -l /dev/shm/network
    /lib/init/readlink -f /etc/network/run
    /bin/readlink -f /etc/network/run
    cat /proc/mounts

> dpkg: error processing ifupdown (--configure):
> subprocess post-installation script returned error exit status 1
> Errors were encountered while processing:
> ifupdown
>
> After googling for a few minutes I found that

Can you give me the URL where you saw information about this problem?

--
Thomas Hood <email address hidden>

Revision history for this message
In , Thomas Hood (jdthood-aglu) wrote : severity serious

severity 302519 serious
thanks

This bug should be fixed for sarge.

In looking for the cause of the bug I have realized that ifupdown is
missing a versioned Dependency on initscripts which supplies a necessary
version of the readlink program; this alone justifies another
sarge-targeted release. However, because the submitter had a
satisfactory version of initscripts installed, I don't yet know where
the bug lies.
--
Thomas Hood <email address hidden>

Revision history for this message
In , Henning Makholm (henning-makholm) wrote : Bug #302519: ifupdown: postinst fails

To the submitter: A better short-term workaround if you prefer not to
edit conffiles that are hopefully going to be fixed soon, is to do

# mkdir /var/run/network
# ln -sf /var/run/network/ /etc/network/run

Otherwise the underlying bug might show up next time you boot, anyway.

The culprit seems to be the following snippet (lines 92ff, reformated
for email) in the postinst of -4.12:

 if RUN_CANONICALDIR="$(readlink -f /etc/network/run)" &&
    [ "$RUN_CANONICALDIR" ] ; then
   if ! mkdir "$RUN_CANONICALDIR" ; then
      report_err "Failure creating directory ${RUN_CANONICALDIR}. Aborting."
      exit 1
   fi
 else
   report_err \
    "The canonical path of /etc/network/run could not be determined. Aborting."
   exit 1
 fi

The problem is that "readlink -f /etc/network/run" returns nothing and
exits with failure when /etc/network/run is a dangling symlink, so the
mkdir that is the point of this code never gets a chance to run.

This behavior is exhibited by the readlink in coreutils 5.2.1-2
(sid+sarge), but not for the readlink in debianutils 1.6.2woody1
(woody).

Code that depends on readlink -f being able to stop at a dangling
symlink also appears in /etc/init.d/{ifupdown,ifupdown-clean}
and will abort the boot initialization on systems where
/etc/network/run refers to a ramdisk. (Which I suppose is the default,
as I do not remember having done anything to redirect /etc/network/run
myself).

Refusing to work with dangling links is the documented behavior in GNU
readlink, so coreutils probably cannot be faulted for this. A
quick-and-dirty workaround that should work in the most common cases
would be to say

   $(readlink -f /etc/network/run || readlink /etc/network/run)

instead of just

   $(readlink -f /etc/network/run)

in all three affected files.

--
Henning Makholm "I ... I have to return some videos."

Revision history for this message
In , Henning Makholm (henning-makholm) wrote :

Scripsit Henning Makholm

> The culprit seems to be the following snippet (lines 92ff, reformated
> for email) in the postinst of -4.12:

Oops, a little further attempt at understanding the bug logs told me
that this is not helpful at all. Sorry for the interruption.

--
Henning Makholm "Jeg har tydeligt gjort opmærksom på, at man ved at
                   følge den vej kun bliver gennemsnitligt ca. 48 år gammel,
               og at man sætter sin sociale situation ganske overstyr og, så
           vidt jeg kan overskue, dør i dybeste ulykkelighed og elendighed."

Revision history for this message
In , Thomas Hood (jdthood-aglu) wrote :
Download full text (3.2 KiB)

Thanks a lot for the input.

On Fri, 2005-04-01 at 23:13 +0100, Henning Makholm wrote:
> To the submitter: A better short-term workaround if you prefer not to
> edit conffiles that are hopefully going to be fixed soon, is to do
>
> # mkdir /var/run/network
> # ln -sf /var/run/network/ /etc/network/run
>
> Otherwise the underlying bug might show up next time you boot, anyway.

That can be done as a workaround. It can't be the default because on
some systems /var/ is network mounted.

> The problem is that "readlink -f /etc/network/run" returns nothing and
> exits with failure when /etc/network/run is a dangling symlink, so the
> mkdir that is the point of this code never gets a chance to run.

The code depends on the presence of a "correctly" working readlink
program. Woody debianutils readlink did the right thing and so does
upstream-unreleased coreutils 5.3.0. To work around the fact that
sarge/sid coreutils 5.2.1 does the wrong thing when the ultimate target
of the symlink chain does not exist, we use the
substitute /lib/init/readlink program included in current initscripts
packages.

jdthood@thanatos:~$ cd /tmp
jdthood@thanatos:/tmp$ ln -s foo bar
jdthood@thanatos:/tmp$ /bin/readlink -f bar
jdthood@thanatos:/tmp$ echo $?
1
jdthood@thanatos:/tmp$ /lib/init/readlink -f bar
/tmp/foo
jdthood@thanatos:/tmp$ echo $?
0

The postinst sets

    PATH=/lib/init:/sbin:/bin:/usr/sbin:/usr/bin

so that the substitute version of the program gets run.

The submitter has an appropriate version of initscripts installed, so he
should be getting the right version of the readlink program.

ifupdown is missing a versioned Dependency on initscripts so we will
need to make a new version for sarge which adds it, but this doesn't
seem to have been the cause of the failure on the submitter's system.

> Code that depends on readlink -f being able to stop at a dangling
> symlink also appears in /etc/init.d/{ifupdown,ifupdown-clean}
> and will abort the boot initialization on systems where
> /etc/network/run refers to a ramdisk. (Which I suppose is the default,
> as I do not remember having done anything to redirect /etc/network/run
> myself).

/etc/init.d/ifupdown also set the PATH in order to get the fixed
readlink program.

/etc/init.d/ifupdown-clean does not set the PATH because it does not
need the fixed readlink program; the current coreutils readlink program
only deviates in its behavior if the ultimate target does not exist,
whereas ifupdown-clean only calls the readlink program if the ultimate
target does exist.

> Refusing to work with dangling links is the documented behavior in GNU
> readlink, so coreutils probably cannot be faulted for this.

Well, upstream changed the behavior back to the old way for 5.3.0.

> A
> quick-and-dirty workaround that should work in the most common cases
> would be to say
>
> $(readlink -f /etc/network/run || readlink /etc/network/run)
>
> instead of just
>
> $(readlink -f /etc/network/run)
>
> in all three affected files.

Interesting idea, but it isn't that simple. For one thing, readlink
without -f prints a relative path rather than a canonical one. Second,
we have to make sure that we don't e...

Read more...

Revision history for this message
In , Thomas Hood (jdthood-aglu) wrote : moreinfo

tags 302519 moreinfo
thanks
--
Thomas Hood <email address hidden>

Revision history for this message
In , Thomas Hood (jdthood-aglu) wrote : unreproducible

I am unable to reproduce the failure, so I speculate that the failure
only occurs when certain combinations of packages are installed.

--
Thomas Hood <email address hidden>

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Automatically imported from Debian bug report #302519 http://bugs.debian.org/302519

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-ID: <email address hidden>
Date: Fri, 1 Apr 2005 10:45:13 +0200 (CEST)
From: Stefan Kluth <email address hidden>
To: Debian Bug Tracking System <email address hidden>
Subject: ifupdown: postinst fails if /dev/shm/network/ does not exist

Subject: ifupdown: postinst fails if /dev/shm/network/ does not exist
Package: ifupdown
Version: 0.6.4-4.12
Severity: important

*** Please type your report below this line ***
after the canocical apt-get update; apt-get upgrade ifupdown fails
like this:

Setting up ifupdown (0.6.4-4.12) ...
ifupdown.postinst: Error: The canonical path of /etc/network/run could
not be determined. Aborting.
dpkg: error processing ifupdown (--configure):
 subprocess post-installation script returned error exit status 1
 Errors were encountered while processing:
  ifupdown

After googling for a few minutes I found that

$ mkdir /dev/shm/network

fixes the problem:

$ dpkg --configure --pending
Setting up ifupdown (0.6.4-4.12) ...

Cheers, Stefan

-- System Information:
Debian Release: 3.1
  APT prefers testing
  APT policy: (990, 'testing')
Architecture: i386 (i686)
Kernel: Linux 2.4.22-xfs
Locale: LANG=en_US, LC_CTYPE=en_US (charmap=ISO-8859-1)

Versions of packages ifupdown depends on:
ii debconf [debconf-2.0] 1.4.30.11 Debian configuration management sy
ii libc6 2.3.2.ds1-20 GNU C Library: Shared libraries an
ii net-tools 1.60-10 The NET-3 networking toolkit

-- debconf information:
  ifupdown/convert-interfaces: true

---Stefan Kluth, PhD----------Wissenschaftlicher Mitarbeiter-----------
- MPI fuer Physik - phone: +49 89 32354 468 - OPAL& -
- Foehringer Ring 6 - fax: +49 89 32354 305 - ATLAS -
---D-80805 Munich, Germany----e-mail: <email address hidden>--------------

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-Id: <1112348366.12352.113.camel@thanatos>
Date: Fri, 01 Apr 2005 11:39:26 +0200
From: Thomas Hood <email address hidden>
To: Stefan Kluth <email address hidden>, <email address hidden>
Subject: Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does
 not exist

On Fri, 2005-04-01 at 10:45 +0200, Stefan Kluth wrote:
> Setting up ifupdown (0.6.4-4.12) ...
> ifupdown.postinst: Error: The canonical path of /etc/network/run could
> not be determined. Aborting.

What version of the initscripts package do you have?

--
Thomas Hood <email address hidden>

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-ID: <email address hidden>
Date: Fri, 1 Apr 2005 12:07:05 +0200 (CEST)
From: Stefan Kluth <email address hidden>
To: Thomas Hood <email address hidden>
Cc: <email address hidden>
Subject: Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does
 not exist

On Fri, 1 Apr 2005, Thomas Hood wrote:

> On Fri, 2005-04-01 at 10:45 +0200, Stefan Kluth wrote:
> > Setting up ifupdown (0.6.4-4.12) ...
> > ifupdown.postinst: Error: The canonical path of /etc/network/run could
> > not be determined. Aborting.
>
>
> What version of the initscripts package do you have?

skluth@sixpack:~$ apt-cache show initscripts
Package: initscripts
Priority: required
Section: base
Installed-Size: 240
Maintainer: Miquel van Smoorenburg <email address hidden>
Architecture: i386
Source: sysvinit
Version: 2.86.ds1-1
Replaces: sysvinit (<< 2.85-12), libc6, libc6.1, libc0.1, libc0.3
Depends: libc6 (>= 2.3.2.ds1-4), libc6 (>= 2.3.2.ds1-12), dpkg (>= 1.9.0),
mount (>= 2.11l-1), util-linux (>= 2.9t-2), e2fsprogs (>= 1.19-1),
coreutils (>= 5.0.91-2)
Conflicts: kbd (<< 0.95-2), mdutils (<< 0.35-9)
Filename: pool/main/s/sysvinit/initscripts_2.86.ds1-1_i386.deb
Size: 30454
MD5sum: 74882efd4ad50da4dbc9020fe0dd67f6
Description: Standard scripts needed for booting and shutting down
 These scripts are meant for standard Debian/GNU/Linux installations.

I run testing derived some time ago from knoppix. I keep the system
uptodate with regular (every few days) updates.

Stefan

---Stefan Kluth, PhD----------Wissenschaftlicher Mitarbeiter-----------
- MPI fuer Physik - phone: +49 89 32354 468 - OPAL& -
- Foehringer Ring 6 - fax: +49 89 32354 305 - ATLAS -
---D-80805 Munich, Germany----e-mail: <email address hidden>--------------

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-Id: <1112366464.12352.150.camel@thanatos>
Date: Fri, 01 Apr 2005 16:41:04 +0200
From: Thomas Hood <email address hidden>
To: <email address hidden>
Subject: severity serious

severity 302519 serious
thanks

This bug should be fixed for sarge.

In looking for the cause of the bug I have realized that ifupdown is
missing a versioned Dependency on initscripts which supplies a necessary
version of the readlink program; this alone justifies another
sarge-targeted release. However, because the submitter had a
satisfactory version of initscripts installed, I don't yet know where
the bug lies.
--
Thomas Hood <email address hidden>

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-Id: <1112362244.12352.125.camel@thanatos>
Date: Fri, 01 Apr 2005 15:30:43 +0200
From: Thomas Hood <email address hidden>
To: <email address hidden>
Subject: Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does
 not exist

On Fri, 2005-04-01 at 10:45 +0200, Stefan Kluth wrote:
> Setting up ifupdown (0.6.4-4.12) ...

>From which version did you upgrade?

> ifupdown.postinst: Error: The canonical path of /etc/network/run could
> not be determined. Aborting.

Do you have the /lib/init/readlink program? Please send the output of
"ls -l /lib/init/readlink".

Please do "grep PATH /var/lib/dpkg/info/ifupdown.postinst".

It will be helpful if you can reproduce the failure. Please remove the
package with "dpkg --remove --force-depends" and reinstall it. (Make
sure you have the package available locally so you don't need to access
the network in order to reinstall.)

You may have to install the version of ifupdown you had before, and
upgrade to 0.6.4-4.12, in order to reproduce the problem.

If the postinst fails again then please send the output of

    ls -l /etc/network
    ls -l /etc/network/run
    ls -l /dev/shm
    ls -l /dev/shm/network
    /lib/init/readlink -f /etc/network/run
    /bin/readlink -f /etc/network/run
    cat /proc/mounts

> dpkg: error processing ifupdown (--configure):
> subprocess post-installation script returned error exit status 1
> Errors were encountered while processing:
> ifupdown
>
> After googling for a few minutes I found that

Can you give me the URL where you saw information about this problem?

--
Thomas Hood <email address hidden>

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-ID: <email address hidden>
Date: Fri, 1 Apr 2005 23:13:02 +0100
From: Henning Makholm <email address hidden>
To: <email address hidden>
Cc: Thomas Hood <email address hidden>
Subject: Bug #302519: ifupdown: postinst fails

To the submitter: A better short-term workaround if you prefer not to
edit conffiles that are hopefully going to be fixed soon, is to do

# mkdir /var/run/network
# ln -sf /var/run/network/ /etc/network/run

Otherwise the underlying bug might show up next time you boot, anyway.

The culprit seems to be the following snippet (lines 92ff, reformated
for email) in the postinst of -4.12:

 if RUN_CANONICALDIR="$(readlink -f /etc/network/run)" &&
    [ "$RUN_CANONICALDIR" ] ; then
   if ! mkdir "$RUN_CANONICALDIR" ; then
      report_err "Failure creating directory ${RUN_CANONICALDIR}. Aborting."
      exit 1
   fi
 else
   report_err \
    "The canonical path of /etc/network/run could not be determined. Aborting."
   exit 1
 fi

The problem is that "readlink -f /etc/network/run" returns nothing and
exits with failure when /etc/network/run is a dangling symlink, so the
mkdir that is the point of this code never gets a chance to run.

This behavior is exhibited by the readlink in coreutils 5.2.1-2
(sid+sarge), but not for the readlink in debianutils 1.6.2woody1
(woody).

Code that depends on readlink -f being able to stop at a dangling
symlink also appears in /etc/init.d/{ifupdown,ifupdown-clean}
and will abort the boot initialization on systems where
/etc/network/run refers to a ramdisk. (Which I suppose is the default,
as I do not remember having done anything to redirect /etc/network/run
myself).

Refusing to work with dangling links is the documented behavior in GNU
readlink, so coreutils probably cannot be faulted for this. A
quick-and-dirty workaround that should work in the most common cases
would be to say

   $(readlink -f /etc/network/run || readlink /etc/network/run)

instead of just

   $(readlink -f /etc/network/run)

in all three affected files.

--
Henning Makholm "I ... I have to return some videos."

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-ID: <email address hidden>
Date: Fri, 1 Apr 2005 23:19:18 +0100
From: Henning Makholm <email address hidden>
To: Thomas Hood <email address hidden>
Cc: <email address hidden>
Subject: Re: Bug #302519: ifupdown: postinst fails

Scripsit Henning Makholm

> The culprit seems to be the following snippet (lines 92ff, reformated
> for email) in the postinst of -4.12:

Oops, a little further attempt at understanding the bug logs told me
that this is not helpful at all. Sorry for the interruption.

--
Henning Makholm "Jeg har tydeligt gjort opm�som p�at man ved at
                   f�den vej kun bliver gennemsnitligt ca. 48 �gammel,
               og at man s�er sin sociale situation ganske overstyr og, s� vidt jeg kan overskue, d�dybeste ulykkelighed og elendighed."

Revision history for this message
Debian Bug Importer (debzilla) wrote :
Download full text (3.5 KiB)

Message-Id: <1112421770.21883.23.camel@thanatos>
Date: Sat, 02 Apr 2005 08:02:50 +0200
From: Thomas Hood <email address hidden>
To: Henning Makholm <email address hidden>
Cc: <email address hidden>
Subject: Re: Bug #302519: ifupdown: postinst fails

Thanks a lot for the input.

On Fri, 2005-04-01 at 23:13 +0100, Henning Makholm wrote:
> To the submitter: A better short-term workaround if you prefer not to
> edit conffiles that are hopefully going to be fixed soon, is to do
>
> # mkdir /var/run/network
> # ln -sf /var/run/network/ /etc/network/run
>
> Otherwise the underlying bug might show up next time you boot, anyway.

That can be done as a workaround. It can't be the default because on
some systems /var/ is network mounted.

> The problem is that "readlink -f /etc/network/run" returns nothing and
> exits with failure when /etc/network/run is a dangling symlink, so the
> mkdir that is the point of this code never gets a chance to run.

The code depends on the presence of a "correctly" working readlink
program. Woody debianutils readlink did the right thing and so does
upstream-unreleased coreutils 5.3.0. To work around the fact that
sarge/sid coreutils 5.2.1 does the wrong thing when the ultimate target
of the symlink chain does not exist, we use the
substitute /lib/init/readlink program included in current initscripts
packages.

jdthood@thanatos:~$ cd /tmp
jdthood@thanatos:/tmp$ ln -s foo bar
jdthood@thanatos:/tmp$ /bin/readlink -f bar
jdthood@thanatos:/tmp$ echo $?
1
jdthood@thanatos:/tmp$ /lib/init/readlink -f bar
/tmp/foo
jdthood@thanatos:/tmp$ echo $?
0

The postinst sets

    PATH=/lib/init:/sbin:/bin:/usr/sbin:/usr/bin

so that the substitute version of the program gets run.

The submitter has an appropriate version of initscripts installed, so he
should be getting the right version of the readlink program.

ifupdown is missing a versioned Dependency on initscripts so we will
need to make a new version for sarge which adds it, but this doesn't
seem to have been the cause of the failure on the submitter's system.

> Code that depends on readlink -f being able to stop at a dangling
> symlink also appears in /etc/init.d/{ifupdown,ifupdown-clean}
> and will abort the boot initialization on systems where
> /etc/network/run refers to a ramdisk. (Which I suppose is the default,
> as I do not remember having done anything to redirect /etc/network/run
> myself).

/etc/init.d/ifupdown also set the PATH in order to get the fixed
readlink program.

/etc/init.d/ifupdown-clean does not set the PATH because it does not
need the fixed readlink program; the current coreutils readlink program
only deviates in its behavior if the ultimate target does not exist,
whereas ifupdown-clean only calls the readlink program if the ultimate
target does exist.

> Refusing to work with dangling links is the documented behavior in GNU
> readlink, so coreutils probably cannot be faulted for this.

Well, upstream changed the behavior back to the old way for 5.3.0.

> A
> quick-and-dirty workaround that should work in the most common cases
> would be to say
>
> $(readlink -f /etc/network/run || readlink /etc/network/run)
>
> instead o...

Read more...

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-Id: <1112440342.21883.79.camel@thanatos>
Date: Sat, 02 Apr 2005 13:12:21 +0200
From: Thomas Hood <email address hidden>
To: <email address hidden>
Subject: moreinfo

tags 302519 moreinfo
thanks
--
Thomas Hood <email address hidden>

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-Id: <1112516007.21883.94.camel@thanatos>
Date: Sun, 03 Apr 2005 10:13:27 +0200
From: Thomas Hood <email address hidden>
To: <email address hidden>
Subject: unreproducible

I am unable to reproduce the failure, so I speculate that the failure
only occurs when certain combinations of packages are installed.

--
Thomas Hood <email address hidden>

Revision history for this message
In , Stefan Kluth (skluth) wrote : Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does not exist

On Fri, 1 Apr 2005, Thomas Hood wrote:

> On Fri, 2005-04-01 at 10:45 +0200, Stefan Kluth wrote:
> > Setting up ifupdown (0.6.4-4.12) ...
>
>
> >From which version did you upgrade?
>
>
> > ifupdown.postinst: Error: The canonical path of /etc/network/run could
> > not be determined. Aborting.
>
>
> Do you have the /lib/init/readlink program? Please send the output of
> "ls -l /lib/init/readlink".

skluth@sixpack:~$ ls -l /lib/init/readlink
ls: /lib/init/readlink: No such file or directory
skluth@sixpack:~$ which readlink
/bin/readlink

> Please do "grep PATH /var/lib/dpkg/info/ifupdown.postinst".

skluth@sixpack:~$ grep PATH /var/lib/dpkg/info/ifupdown.postinst
# /lib/init/ can be removed from the PATH.
PATH=/lib/init:/sbin:/bin:/usr/sbin:/usr/bin

>
> It will be helpful if you can reproduce the failure. Please remove the
> package with "dpkg --remove --force-depends" and reinstall it. (Make
> sure you have the package available locally so you don't need to access
> the network in order to reinstall.)

I'll do that later.

> > After googling for a few minutes I found that
>
>
> Can you give me the URL where you saw information about this problem?

I am not sure I can reproduce the page anymore ... I didn't keep the link
... sorry.

Cheers, Stefan

---Stefan Kluth, PhD----------Wissenschaftlicher Mitarbeiter-----------
- MPI fuer Physik - phone: +49 89 32354 468 - OPAL& -
- Foehringer Ring 6 - fax: +49 89 32354 305 - ATLAS -
---D-80805 Munich, Germany----e-mail: <email address hidden>--------------

Revision history for this message
Matt Zimmerman (mdz) wrote :

It's not entirely clear whether this bug would affect woody->warty or
warty->hoary upgrades; please investigate

Revision history for this message
In , Anthony Towns (ajt) wrote : Bug#302519: fixed in ifupdown 0.6.5
Download full text (25.2 KiB)

Source: ifupdown
Source-Version: 0.6.5

We believe that the bug you reported is fixed in the latest version of
ifupdown, which is due to be installed in the Debian FTP archive:

ifupdown_0.6.5.dsc
  to pool/main/i/ifupdown/ifupdown_0.6.5.dsc
ifupdown_0.6.5.tar.gz
  to pool/main/i/ifupdown/ifupdown_0.6.5.tar.gz
ifupdown_0.6.5_i386.deb
  to pool/main/i/ifupdown/ifupdown_0.6.5_i386.deb

A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed. If you
have further comments please address them to <email address hidden>,
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Anthony Towns <email address hidden> (supplier of updated ifupdown package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing <email address hidden>)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Format: 1.7
Date: Mon, 4 Apr 2005 23:41:06 +1000
Source: ifupdown
Binary: ifupdown
Architecture: source i386
Version: 0.6.5
Distribution: unstable
Urgency: low
Maintainer: Anthony Towns <email address hidden>
Changed-By: Anthony Towns <email address hidden>
Description:
 ifupdown - high level tools to configure network interfaces
Closes: 54814 57731 57830 62542 62633 63071 67743 67796 68869 69723 70030 71779 72872 74312 74563 75279 76086 76142 76198 76985 79683 79822 81143 81150 81611 82339 82604 83496 83542 83557 83804 84602 84697 85206 86410 86895 86902 88946 88947 88948 92993 94656 96265 97782 98448 99444 101204 103868 105342 107701 108857 108876 110198 112012 112652 113338 113620 114429 121755 122422 129003 135502 138403 138694 141634 148666 150773 151465 151932 152853 152893 154517 156789 157698 160918 164823 169194 171981 172147 173101 174764 175679 178226 178630 180000 186316 196366 196865 198841 200786 203636 204468 207727 208607 208726 208811 209006 210436 213068 213723 216716 224468 229503 229527 231404 231910 231995 232347 232594 235993 239142 241248 242314 242527 242537 242607 242867 245067 245067 247772 248717 249233 250713 254388 255218 255228 255574 258965 259609 263913 264134 265165 266282 266479 283649 284123 286155 286166 287172 294970 296273 297762 297898 300937 302519
Changes:
 ifupdown (0.6.5) unstable; urgency=low
 .
   * The Gernot Heiser release -- aged like a fine South Australian wine.
 .
   * Switch to Debian-native versioning / source packaging for the time being.
     (Closes: Bug#84697)
   * Change Section: to base, to match overrides.
 .
   * Thanks to Michael Weber, Javier Fernandez-Sanguino Pena, Marc
     Haber, and Thomas Hood for NMUs. (Closes: Bug#150773, Bug#151465,
     Bug#152893, Bug#208726, Bug#209006, Bug#242314, Bug#263913,
     Bug#266282, Bug#297762)
 .
   * debian/rules: Don't re-build unnecessarily. Thanks to Michael Banck
     (Closes: Bug#296273)
 .
   * Add post-up, pre-down as aliases for "up" and "down". (Closes: Bug#62633)
 .
   * Deprecate (undocument) "--exclude" option.
 .
   * Add support for "allow-*" lines for systems such as hotplug or ifplugd.
     Usage is "ifup --allow=hotplug <interfaces>"...

Revision history for this message
In , Thomas Hood (jdthood-aglu) wrote : Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does not exist

On 1 Apr 2005 12:07:05, Stefan Kluth wrote:
> On Fri, 1 Apr 2005, Thomas Hood wrote:
> > What version of the initscripts package do you have?
>
> skluth@sixpack:~$ apt-cache show initscripts
[...]
> Version: 2.86.ds1-1

Please do "dpkg -s initscripts|grep Version".

On Mon, 2005-04-04 at 17:38 +0200, Stefan Kluth wrote:
> skluth@sixpack:~$ ls -l /lib/init/readlink
> ls: /lib/init/readlink: No such file or directory
> skluth@sixpack:~$ which readlink
> /bin/readlink

Thanks for the additional information. This confirms that the problem
was an outdated initscripts package. Current initscripts contains
a /lib/init/readlink program which works around the problem in the
current coreutils readlink program.

This will be fixed when I can find someone to sponsor an upload.

--
Thomas Hood <email address hidden>

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-ID: <email address hidden>
Date: Mon, 4 Apr 2005 17:38:03 +0200 (CEST)
From: Stefan Kluth <email address hidden>
To: Thomas Hood <email address hidden>, <email address hidden>
Cc: <email address hidden>,
 Debian BTS <email address hidden>
Subject: Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does
 not exist

On Fri, 1 Apr 2005, Thomas Hood wrote:

> On Fri, 2005-04-01 at 10:45 +0200, Stefan Kluth wrote:
> > Setting up ifupdown (0.6.4-4.12) ...
>
>
> >From which version did you upgrade?
>
>
> > ifupdown.postinst: Error: The canonical path of /etc/network/run could
> > not be determined. Aborting.
>
>
> Do you have the /lib/init/readlink program? Please send the output of
> "ls -l /lib/init/readlink".

skluth@sixpack:~$ ls -l /lib/init/readlink
ls: /lib/init/readlink: No such file or directory
skluth@sixpack:~$ which readlink
/bin/readlink

> Please do "grep PATH /var/lib/dpkg/info/ifupdown.postinst".

skluth@sixpack:~$ grep PATH /var/lib/dpkg/info/ifupdown.postinst
# /lib/init/ can be removed from the PATH.
PATH=/lib/init:/sbin:/bin:/usr/sbin:/usr/bin

>
> It will be helpful if you can reproduce the failure. Please remove the
> package with "dpkg --remove --force-depends" and reinstall it. (Make
> sure you have the package available locally so you don't need to access
> the network in order to reinstall.)

I'll do that later.

> > After googling for a few minutes I found that
>
>
> Can you give me the URL where you saw information about this problem?

I am not sure I can reproduce the page anymore ... I didn't keep the link
... sorry.

Cheers, Stefan

---Stefan Kluth, PhD----------Wissenschaftlicher Mitarbeiter-----------
- MPI fuer Physik - phone: +49 89 32354 468 - OPAL& -
- Foehringer Ring 6 - fax: +49 89 32354 305 - ATLAS -
---D-80805 Munich, Germany----e-mail: <email address hidden>--------------

Revision history for this message
Debian Bug Importer (debzilla) wrote :
Download full text (25.4 KiB)

Message-Id: <email address hidden>
Date: Mon, 04 Apr 2005 12:02:25 -0400
From: Anthony Towns <email address hidden>
To: <email address hidden>
Subject: Bug#302519: fixed in ifupdown 0.6.5

Source: ifupdown
Source-Version: 0.6.5

We believe that the bug you reported is fixed in the latest version of
ifupdown, which is due to be installed in the Debian FTP archive:

ifupdown_0.6.5.dsc
  to pool/main/i/ifupdown/ifupdown_0.6.5.dsc
ifupdown_0.6.5.tar.gz
  to pool/main/i/ifupdown/ifupdown_0.6.5.tar.gz
ifupdown_0.6.5_i386.deb
  to pool/main/i/ifupdown/ifupdown_0.6.5_i386.deb

A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed. If you
have further comments please address them to <email address hidden>,
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Anthony Towns <email address hidden> (supplier of updated ifupdown package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing <email address hidden>)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Format: 1.7
Date: Mon, 4 Apr 2005 23:41:06 +1000
Source: ifupdown
Binary: ifupdown
Architecture: source i386
Version: 0.6.5
Distribution: unstable
Urgency: low
Maintainer: Anthony Towns <email address hidden>
Changed-By: Anthony Towns <email address hidden>
Description:
 ifupdown - high level tools to configure network interfaces
Closes: 54814 57731 57830 62542 62633 63071 67743 67796 68869 69723 70030 71779 72872 74312 74563 75279 76086 76142 76198 76985 79683 79822 81143 81150 81611 82339 82604 83496 83542 83557 83804 84602 84697 85206 86410 86895 86902 88946 88947 88948 92993 94656 96265 97782 98448 99444 101204 103868 105342 107701 108857 108876 110198 112012 112652 113338 113620 114429 121755 122422 129003 135502 138403 138694 141634 148666 150773 151465 151932 152853 152893 154517 156789 157698 160918 164823 169194 171981 172147 173101 174764 175679 178226 178630 180000 186316 196366 196865 198841 200786 203636 204468 207727 208607 208726 208811 209006 210436 213068 213723 216716 224468 229503 229527 231404 231910 231995 232347 232594 235993 239142 241248 242314 242527 242537 242607 242867 245067 245067 247772 248717 249233 250713 254388 255218 255228 255574 258965 259609 263913 264134 265165 266282 266479 283649 284123 286155 286166 287172 294970 296273 297762 297898 300937 302519
Changes:
 ifupdown (0.6.5) unstable; urgency=low
 .
   * The Gernot Heiser release -- aged like a fine South Australian wine.
 .
   * Switch to Debian-native versioning / source packaging for the time being.
     (Closes: Bug#84697)
   * Change Section: to base, to match overrides.
 .
   * Thanks to Michael Weber, Javier Fernandez-Sanguino Pena, Marc
     Haber, and Thomas Hood for NMUs. (Closes: Bug#150773, Bug#151465,
     Bug#152893, Bug#208726, Bug#209006, Bug#242314, Bug#263913,
     Bug#266282, Bug#297762)
 .
   * debian/rules: Don't re-build unnecessarily. Thanks to Michael Banck
     (Closes: Bug#296273)
 .
   * Add post-up, pre-down as aliases for "up" and "do...

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-Id: <1112637434.19313.116.camel@thanatos>
Date: Mon, 04 Apr 2005 19:57:13 +0200
From: Thomas Hood <email address hidden>
To: <email address hidden>
Subject: Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does
 not exist

On 1 Apr 2005 12:07:05, Stefan Kluth wrote:
> On Fri, 1 Apr 2005, Thomas Hood wrote:
> > What version of the initscripts package do you have?
>
> skluth@sixpack:~$ apt-cache show initscripts
[...]
> Version: 2.86.ds1-1

Please do "dpkg -s initscripts|grep Version".

On Mon, 2005-04-04 at 17:38 +0200, Stefan Kluth wrote:
> skluth@sixpack:~$ ls -l /lib/init/readlink
> ls: /lib/init/readlink: No such file or directory
> skluth@sixpack:~$ which readlink
> /bin/readlink

Thanks for the additional information. This confirms that the problem
was an outdated initscripts package. Current initscripts contains
a /lib/init/readlink program which works around the problem in the
current coreutils readlink program.

This will be fixed when I can find someone to sponsor an upload.

--
Thomas Hood <email address hidden>

Revision history for this message
In , Anthony Towns (aj-azure) wrote : Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does not exist

Stefan Kluth wrote:
> Setting up ifupdown (0.6.4-4.12) ...
> ifupdown.postinst: Error: The canonical path of /etc/network/run could
> not be determined. Aborting.

> After googling for a few minutes I found that
> $ mkdir /dev/shm/network

I believe I had some weird, unreproducible, behaviour from
/lib/init/readlink when trying some of this stuff out; so afaics the
above should never happen, but I'm not sure it actually doesn't. The
obvious explanation would've been that /dev/shm didn't exist, let alone
/dev/shm/network, though that seems odd too.

Anyway, new ifupdown in unstable doesn't use "readlink -f", instead
having its own sh script fragment, so should work more reliably.

This bug's already closed. FYI only.

Cheers,
aj

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-ID: <email address hidden>
Date: Tue, 05 Apr 2005 11:55:05 +1000
From: Anthony Towns <email address hidden>
To: Stefan Kluth <email address hidden>, <email address hidden>
Subject: Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does
 not exist

Stefan Kluth wrote:
> Setting up ifupdown (0.6.4-4.12) ...
> ifupdown.postinst: Error: The canonical path of /etc/network/run could
> not be determined. Aborting.

> After googling for a few minutes I found that
> $ mkdir /dev/shm/network

I believe I had some weird, unreproducible, behaviour from
/lib/init/readlink when trying some of this stuff out; so afaics the
above should never happen, but I'm not sure it actually doesn't. The
obvious explanation would've been that /dev/shm didn't exist, let alone
/dev/shm/network, though that seems odd too.

Anyway, new ifupdown in unstable doesn't use "readlink -f", instead
having its own sh script fragment, so should work more reliably.

This bug's already closed. FYI only.

Cheers,
aj

Revision history for this message
In , Thomas Hood (jdthood-aglu) wrote : Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does not exist

On Tue, 2005-04-05 at 11:55 +1000, Anthony Towns wrote:
> I believe I had some weird, unreproducible, behaviour from
> /lib/init/readlink when trying some of this stuff out

On Mon, 2005-04-04 at 17:38 +0200, Stefan Kluth wrote:
> skluth@sixpack:~$ ls -l /lib/init/readlink
> ls: /lib/init/readlink: No such file or directory

The submitter has _no_ /lib/init/readlink. That, and not a bug
in /lib/init/readlink, was the basic problem.

--
Thomas Hood <email address hidden>

Revision history for this message
In , Stefan Kluth (skluth) wrote : Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does not exist

On Mon, 4 Apr 2005, Thomas Hood wrote:

> On 1 Apr 2005 12:07:05, Stefan Kluth wrote:
> > On Fri, 1 Apr 2005, Thomas Hood wrote:
> > > What version of the initscripts package do you have?
> >
> > skluth@sixpack:~$ apt-cache show initscripts
> [...]
> > Version: 2.86.ds1-1
>
>
> Please do "dpkg -s initscripts|grep Version".

skluth@sixpack:~$ dpkg -s initscripts|grep Version
Package `initscripts' is not installed and no info is available.
Use dpkg --info (= dpkg-deb --info) to examine archive files,
and dpkg --contents (= dpkg-deb --contents) to list their contents.

skluth@sixpack:~$ apt-cache policy initscripts
initscripts:
  Installed: (none)
  Candidate: 2.86.ds1-1
  Version Table:
     2.86.ds1-1 0
        990 http://ftp.de.debian.org testing/main Packages

As I said, I originally installed using Knoppix 3.4, then fixed apt to
testing and surfed the wave ever since. Knopper seems to have fiddled
with the initscripts somehow. Should I install initscripts?

So I guess part of the problem lies with me for not running a stock debian
system ... but it seems other problems with ifupdown were resolved in the
process, so there there was some benefit.

Now that we have a really decent installer I should probably install from
scratch to avoid problems like this in the future.

Thanks a lot for your help!

---Stefan Kluth, PhD----------Wissenschaftlicher Mitarbeiter-----------
- MPI fuer Physik - phone: +49 89 32354 468 - OPAL& -
- Foehringer Ring 6 - fax: +49 89 32354 305 - ATLAS -
---D-80805 Munich, Germany----e-mail: <email address hidden>--------------

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-Id: <1112685361.15045.4.camel@thanatos>
Date: Tue, 05 Apr 2005 09:16:00 +0200
From: Thomas Hood <email address hidden>
To: <email address hidden>, <email address hidden>
Subject: Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does
 not exist

On Tue, 2005-04-05 at 11:55 +1000, Anthony Towns wrote:
> I believe I had some weird, unreproducible, behaviour from
> /lib/init/readlink when trying some of this stuff out

On Mon, 2005-04-04 at 17:38 +0200, Stefan Kluth wrote:
> skluth@sixpack:~$ ls -l /lib/init/readlink
> ls: /lib/init/readlink: No such file or directory

The submitter has _no_ /lib/init/readlink. That, and not a bug
in /lib/init/readlink, was the basic problem.

--
Thomas Hood <email address hidden>

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-ID: <email address hidden>
Date: Tue, 5 Apr 2005 10:41:55 +0200 (CEST)
From: Stefan Kluth <email address hidden>
To: Thomas Hood <email address hidden>, <email address hidden>
Cc: <email address hidden>,
 Debian BTS <email address hidden>
Subject: Re: Bug#302519: ifupdown: postinst fails if /dev/shm/network/ does
 not exist

On Mon, 4 Apr 2005, Thomas Hood wrote:

> On 1 Apr 2005 12:07:05, Stefan Kluth wrote:
> > On Fri, 1 Apr 2005, Thomas Hood wrote:
> > > What version of the initscripts package do you have?
> >
> > skluth@sixpack:~$ apt-cache show initscripts
> [...]
> > Version: 2.86.ds1-1
>
>
> Please do "dpkg -s initscripts|grep Version".

skluth@sixpack:~$ dpkg -s initscripts|grep Version
Package `initscripts' is not installed and no info is available.
Use dpkg --info (= dpkg-deb --info) to examine archive files,
and dpkg --contents (= dpkg-deb --contents) to list their contents.

skluth@sixpack:~$ apt-cache policy initscripts
initscripts:
  Installed: (none)
  Candidate: 2.86.ds1-1
  Version Table:
     2.86.ds1-1 0
        990 http://ftp.de.debian.org testing/main Packages

As I said, I originally installed using Knoppix 3.4, then fixed apt to
testing and surfed the wave ever since. Knopper seems to have fiddled
with the initscripts somehow. Should I install initscripts?

So I guess part of the problem lies with me for not running a stock debian
system ... but it seems other problems with ifupdown were resolved in the
process, so there there was some benefit.

Now that we have a really decent installer I should probably install from
scratch to avoid problems like this in the future.

Thanks a lot for your help!

---Stefan Kluth, PhD----------Wissenschaftlicher Mitarbeiter-----------
- MPI fuer Physik - phone: +49 89 32354 468 - OPAL& -
- Foehringer Ring 6 - fax: +49 89 32354 305 - ATLAS -
---D-80805 Munich, Germany----e-mail: <email address hidden>--------------

Revision history for this message
Adam Conrad (adconrad) wrote :

Fixed for us in 0.6.4-4.10ubuntu3. Fixed more intrusively in the new
Debian version, so it won't come back to haunt us.

Revision history for this message
In , Thomas Hood (jdthood-aglu) wrote : reopen and tag "sarge"

reopen 302519
retitle 302519 ifupdown: missing versioned Dependency on initscripts (>= 2.85-15)
tags 302519 sarge
thanks
--
Thomas Hood <email address hidden>

Revision history for this message
In , Thomas Hood (jdthood-aglu) wrote : tags - moreinfo

tags 302519 - moreinfo
thanks
--
Thomas Hood <email address hidden>

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-Id: <1113248611.22196.12.camel@thanatos>
Date: Mon, 11 Apr 2005 21:43:30 +0200
From: Thomas Hood <email address hidden>
To: <email address hidden>
Subject: reopen and tag "sarge"

reopen 302519
retitle 302519 ifupdown: missing versioned Dependency on initscripts (>= 2.85-15)
tags 302519 sarge
thanks
--
Thomas Hood <email address hidden>

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-Id: <1113248657.22196.14.camel@thanatos>
Date: Mon, 11 Apr 2005 21:44:13 +0200
From: Thomas Hood <email address hidden>
To: <email address hidden>
Subject: tags - moreinfo

tags 302519 - moreinfo
thanks
--
Thomas Hood <email address hidden>

Revision history for this message
In , Thomas Hood (jdthood-aglu) wrote : t-p-u upload?

If bug #304188 in ifupdown 0.6.6 won't be fixed in time for sarge then I
have a 0.6.4-4.13 version ready which fixes #302519 in ifupdown
0.6.4-4.12.
--
Thomas Hood <email address hidden>

Revision history for this message
Debian Bug Importer (debzilla) wrote :

Message-Id: <1113765290.21496.56.camel@thanatos>
Date: Sun, 17 Apr 2005 21:14:50 +0200
From: Thomas Hood <email address hidden>
To: <email address hidden>
Subject: t-p-u upload?

If bug #304188 in ifupdown 0.6.6 won't be fixed in time for sarge then I
have a 0.6.4-4.13 version ready which fixes #302519 in ifupdown
0.6.4-4.12.
--
Thomas Hood <email address hidden>

Revision history for this message
In , Steve Langasek (vorlon) wrote : Re: ifupdown: missing versioned Dependency on initscripts (>= 2.85-15)

ifupdown 0.6.7 has been accepted into testing, so it appears this bug is
fixed.

Thanks,
--
Steve Langasek
postmodern programmer

Changed in ifupdown:
status: Unknown → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.