Bug #27799 “udev initscript breaks bootlogd by leaving /dev/pts ...” : Bugs : udev package : Ubuntu

Revision history for this message

In Debian Bug tracker #237056, Miquel van Smoorenburg (miquels) wrote on 2004-03-10: Re: Bug#237056: bootlogd: conflicts with udev installed

#1

title 237056 bootlogd: should integrate better with udev ?
severity 237056 wishlist
thanks

On 2004.03.09 15:09, Max Kutny wrote:
> Package: initscripts
> Version: 2.85-9
> Severity: important
>
> Rationale: the boolog doesn't log messages thus unusable.

udev is an experimental package, and bootlogd is turned off by default.
So you have to install an experimental package and fiddle around with
bootlogd to trigger this bug.

>
> Once I have a udev package installed (a replacement of devfs) I've got
> the following botloog message during startup:
> bootlogd: ioctl(/dev/ttyzf, TIOCCONS): bad file descriptor
>
> There is no /dev/ttyzf file under udev. I'm not sure this device
> must be present on the system thus the bug is against bootlog.

Sure, bootlogd is not ready for all production systems and therefor
turned off by default. You just proved again why that is a good idea.
I'm changing the title and setting the severity to wishlist.

Mike.

> udev version is: 0.021-1
>
> -- Max
>
>
> -- System Information:
> Debian Release: testing/unstable
> APT prefers unstable
> APT policy: (500, 'unstable')
> Architecture: i386 (i686)
> Kernel: Linux 2.6.3-1-686
> Locale: LANG=uk_UA.UTF-8, LC_CTYPE=uk_UA.UTF-8
>
> Versions of packages initscripts depends on:
> ii coreutils 5.0.91-2 The GNU core utilities
> ii dpkg 1.10.19 Package maintenance system for Deb
> ii e2fsprogs 1.35-2 The EXT2 file system utilities and
> ii mount 2.12-6 Tools for mounting and manipulatin
> ii util-linux 2.12-6 Miscellaneous system utilities
>
> -- no debconf information
>
>
>

Revision history for this message

In Debian Bug tracker #237056, Max Kutny (mkut) wrote on 2004-03-10:

#2

On Wed, Mar 10, 2004 at 01:26:19PM +0100, Miquel van Smoorenburg wrote:
> > Severity: important
> >
> > Rationale: the boolog doesn't log messages thus unusable.
>
> udev is an experimental package,

I took it from the unstable yesterday :-E
But please notice, it will be used much more extensive than defvs.

> > Once I have a udev package installed (a replacement of devfs) I've got
> > the following botloog message during startup:
> > bootlogd: ioctl(/dev/ttyzf, TIOCCONS): bad file descriptor
> >
> > There is no /dev/ttyzf file under udev. I'm not sure this device
> > must be present on the system thus the bug is against bootlog.
>
> Sure, bootlogd is not ready for all production systems and therefor
> turned off by default. You just proved again why that is a good idea.
> I'm changing the title and setting the severity to wishlist.

Agree. Of cause it doesn't renders initscripts package unusable. I filed
a bug report thinking of bootlogd instead of initscripts. My bad. I'll be
more careful next time.

Thanks.

-- Max

Revision history for this message

In Debian Bug tracker #237056, Miquel van Smoorenburg (miquels) wrote on 2004-03-10: retitle

#3

retitle 237056 bootlogd: should integrate better with udev ?
thanks

Revision history for this message

In Debian Bug tracker #237056, Thomas Hood (jdthood-yahoo) wrote on 2005-12-15: Still a problem?

#4

Submitter: Do you still have this problem with the latest udev?

I use udev and bootlogd works fine for me.
--
Thomas Hood

Revision history for this message

In Debian Bug tracker #237056, Max Kutny (mkut-umc) wrote on 2005-12-17: Re: Bug#237056: Still a problem?

#5

On Thu, Dec 15, 2005 at 01:26:36PM +0100, Thomas Hood wrote:
> Submitter: Do you still have this problem with the latest udev?
>
> I use udev and bootlogd works fine for me.

It's fine for me too, thanks!

-- Max

Revision history for this message

In Debian Bug tracker #237056, Thomas Hood (jdthood-yahoo) wrote on 2005-12-17:

#6

Max Kutny wrote:
> It's fine for me too, thanks!

Good to hear!

--
Thomas Hood

Revision history for this message

In Debian Bug tracker #237056, Nathanael Nerode (neroden-twcny) wrote on 2005-12-24: I'm getting this bug.

#7

But only on some machines.
Any ideas? :-P

Revision history for this message

In Debian Bug tracker #237056, Thomas Hood (jdthood-yahoo) wrote on 2005-12-26: Re: [Pkg-sysvinit-devel] Bug#237056: I'm getting this bug.

#8

reopen 237056
stop

Nathanael Nerode wrote:
> But only on some machines.
> Any ideas? :-P

Not without some clues.

--
Thomas

Revision history for this message

In Debian Bug tracker #237056, Nathanael Nerode (neroden-twcny) wrote on 2005-12-31: udev breaks bootlogd

#9

Download full text (3.1 KiB)

reassign 237056 udev
severity 237056 grave
retitle 237056 udev initscript breaks bootlogd by leaving /dev/pts unmounted
found 237056 0.076-6
thanks

I've found the source of bug 237056. I think it's udev's responsibility,
but you can discuss that amongst yourselves.

bootlogd calls a routine in glibc which tries to get a UNIX98 PTY. If that
fails, it looks for a BSD (legacy) PTY. If that fails (perhaps because
they're not compiled into the kernel, because nothing much uses them), it
gives up and dies.

The problem lies in the rcS.d init scripts. On my system we have:
S01glibc.sh
S02mountvirtfs
S04udev
S05bootlogd
...
S36mountvirtfs
..

/dev/pts is mounted in mountvirtfs.
S05bootlogd runs as early as possible, but after mountvirtfs, because it
wants /dev/pts (and also /proc).

However, S04udev *unmounts* /dev/pts. And it's not remounted until S36.

Now, bootlogd runs after S04udev, presumably to guarantee that the console
device is available. But this means that /dev/pts isn't available.

As it says in S04udev:
# we need to unmount /dev/pts/ and remount it later over the tmpfs

Except it isn't remounted later, until S36. This is really bad because
it could easily break other programs launched in the initscripts in between,
as well. This is why I assigned this bug to udev at grave severity: it
breaks arbitrary other packages.

The bug might be solved in one of the following ways:
* have udev actually remount /dev/pts (and /dev/shm) at the end of its script
  The natural solution. Creates code duplication, though. Unless udev
  invokes /etc/mountvirtfs directly, in which case it has to watch out
  and make sure that 'set -e' is disabled for the invocation.
* drop yet another invocation of mountvirtfs, between udev and bootlogd
  Requires moving udev's script up to S03 to make room. Ugh.
* split mountvirtfs into multiple scripts, one for /proc and /sys which
  runs before udev, and one for /dev/pts and /dev/shm which runs after.
* move bootlogd before udev
  Requires more static devices, *and* might cause other problems (because
  /dev would vanish under the running bootlogd and reappear); but has the
  advantage of logging udev events. Probably the best long term solution,
  but it's unlikely that it will work right now (anyone want to test it?)

As an aside, none of these solutions are robust to parallelized initscripts
-- but then neither are udev or bootlogd, period. Essentially
nothing can safely run at the same time as udev. Essentially all
initscripts should run after bootlogd for bootlogd to make sense -- but only
if bootlogd is present. And these aren't because the later scripts in any
sense 'require' them.

These two scripts fit very poorly into the parallelized initscripts
concept, and probably deserve special treatment for that reason; worth
thinking about. Perhaps every other initscript should be required to
depend on both udev and bootlogd, and dummy implementations should be
Provided for when the relevant program isn't installed. Or perhaps they
even deserve custom treatment by 'init', though I doubt it.

--
Nathanael Nerode <email address hidden>

Make sure your vote will count.
http://www.verifiedvoting.org...

reassign 237056 udev
severity 237056 grave
retitle 237056 udev initscript breaks bootlogd by leaving /dev/pts unmounted
found 237056 0.076-6
thanks

I've found the source of bug 237056.  I think it's udev's responsibility,
but you can discuss that amongst yourselves.

bootlogd calls a routine in glibc which tries to get a UNIX98 PTY.  If that
fails, it looks for a BSD (legacy) PTY.  If that fails (perhaps because
they're not compiled into the kernel, because nothing much uses them), it
gives up and dies.

The problem lies in the rcS.d init scripts.  On my system we have:
S01glibc.sh
S02mountvirtfs
S04udev
S05bootlogd
...
S36mountvirtfs
..

/dev/pts is mounted in mountvirtfs.
S05bootlogd runs as early as possible, but after mountvirtfs, because it
wants /dev/pts (and also /proc).

However, S04udev *unmounts* /dev/pts.  And it's not remounted until S36.

Now, bootlogd runs after S04udev, presumably to guarantee that the console
device is available. But this means that /dev/pts isn't available.

As it says in S04udev:
# we need to unmount /dev/pts/ and remount it later over the tmpfs

Except it isn't remounted later, until S36.  This is really bad because
it could easily break other programs launched in the initscripts in between,
as well.  This is why I assigned this bug to udev at grave severity: it
breaks arbitrary other packages.

The bug might be solved in one of the following ways:
* have udev actually remount /dev/pts (and /dev/shm) at the end of its script
  The natural solution.  Creates code duplication, though.  Unless udev
  invokes /etc/mountvirtfs directly, in which case it has to watch out
  and make sure that 'set -e' is disabled for the invocation.
* drop yet another invocation of mountvirtfs, between udev and bootlogd
  Requires moving udev's script up to S03 to make room.  Ugh.
* split mountvirtfs into multiple scripts, one for /proc and /sys which
  runs before udev, and one for /dev/pts and /dev/shm which runs after.
* move bootlogd before udev
  Requires more static devices, *and* might cause other problems (because
  /dev would vanish under the running bootlogd and reappear); but has the
  advantage of logging udev events.  Probably the best long term solution,
  but it's unlikely that it will work right now (anyone want to test it?)

As an aside, none of these solutions are robust to parallelized initscripts
-- but then neither are udev or bootlogd, period.  Essentially
nothing can safely run at the same time as udev.  Essentially all
initscripts should run after bootlogd for bootlogd to make sense -- but only
if bootlogd is present.  And these aren't because the later scripts in any
sense 'require' them.

These two scripts fit very poorly into the parallelized initscripts 
concept, and probably deserve special treatment for that reason; worth 
thinking about.  Perhaps every other initscript should be required to 
depend on both udev and bootlogd, and dummy implementations should be 
Provided for when the relevant program isn't installed.  Or perhaps they
even deserve custom treatment by 'init', though I doubt it.

-- 
Nathanael Nerode  <neroden@twcny.rr.com>

Make sure your vote will count.
http://www.verifiedvoting.org/

Revision history for this message

In Debian Bug tracker #237056, Marco d'Itri (md) wrote on 2005-12-31:

#10

found 237056 0.050-1 # this means I can ignore this bug forever
thanks

On Dec 31, Nathanael Nerode <email address hidden> wrote:

> Except it isn't remounted later, until S36. This is really bad because
By design, because it was hard to do it reliably and Miquel assured me
that this would cause no problems (and indeed it did no, except for
bootlogd).

> * have udev actually remount /dev/pts (and /dev/shm) at the end of its script
> The natural solution. Creates code duplication, though. Unless udev
> invokes /etc/mountvirtfs directly, in which case it has to watch out
> and make sure that 'set -e' is disabled for the invocation.
See above. I am not inclined to do this again (it adds complexity to
work around a misfeature in a different package).

> * split mountvirtfs into multiple scripts, one for /proc and /sys which
> runs before udev, and one for /dev/pts and /dev/shm which runs after.
This is what me and Miquel agreed to do, but when I tried discussing it
with the current maintainer he was not very interested.

> * move bootlogd before udev
This is what I do on my system and it works fine, but I'd rather split
mountvirtfs which is the correct solution.

> As an aside, none of these solutions are robust to parallelized initscripts
> -- but then neither are udev or bootlogd, period. Essentially
> nothing can safely run at the same time as udev. Essentially all
> initscripts should run after bootlogd for bootlogd to make sense -- but only
> if bootlogd is present. And these aren't because the later scripts in any
> sense 'require' them.
I have been saying this for a while, now you only need to persuade the
initscripts maintainer.

--
ciao,
Marco

Revision history for this message

Debian Bug Importer (debzilla) wrote on 2005-12-31:

#11

Automatically imported from Debian bug report #237056 http://bugs.debian.org/237056

Revision history for this message

Debian Bug Importer (debzilla) wrote on 2005-12-31:

#12

Message-Id: <E1B0hv8-0000Xl-Q7@localhost>
Date: Tue, 09 Mar 2004 16:09:22 +0200
From: Max Kutny <email address hidden>
To: Debian Bug Tracking System <email address hidden>
Subject: bootlogd: conflicts with udev installed

Package: initscripts
Version: 2.85-9
Severity: important

Rationale: the boolog doesn't log messages thus unusable.

Once I have a udev package installed (a replacement of devfs) I've got
the following botloog message during startup:
bootlogd: ioctl(/dev/ttyzf, TIOCCONS): bad file descriptor

There is no /dev/ttyzf file under udev. I'm not sure this device
must be present on the system thus the bug is against bootlog.

udev version is: 0.021-1

-- Max

-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: i386 (i686)
Kernel: Linux 2.6.3-1-686
Locale: LANG=uk_UA.UTF-8, LC_CTYPE=uk_UA.UTF-8

Versions of packages initscripts depends on:
ii coreutils 5.0.91-2 The GNU core utilities
ii dpkg 1.10.19 Package maintenance system for Deb
ii e2fsprogs 1.35-2 The EXT2 file system utilities and
ii mount 2.12-6 Tools for mounting and manipulatin
ii util-linux 2.12-6 Miscellaneous system utilities

-- no debconf information

Revision history for this message

Debian Bug Importer (debzilla) wrote on 2005-12-31:

#13

Message-ID: <email address hidden>
Date: Wed, 10 Mar 2004 13:26:19 +0100
From: Miquel van Smoorenburg <email address hidden>
To: Max Kutny <email address hidden>
Cc: <email address hidden>, <email address hidden>
Subject: Re: Bug#237056: bootlogd: conflicts with udev installed

title 237056 bootlogd: should integrate better with udev ?
severity 237056 wishlist
thanks

On 2004.03.09 15:09, Max Kutny wrote:
> Package: initscripts
> Version: 2.85-9
> Severity: important
>
> Rationale: the boolog doesn't log messages thus unusable.

udev is an experimental package, and bootlogd is turned off by default.
So you have to install an experimental package and fiddle around with
bootlogd to trigger this bug.

>
> Once I have a udev package installed (a replacement of devfs) I've got
> the following botloog message during startup:
> bootlogd: ioctl(/dev/ttyzf, TIOCCONS): bad file descriptor
>
> There is no /dev/ttyzf file under udev. I'm not sure this device
> must be present on the system thus the bug is against bootlog.

Sure, bootlogd is not ready for all production systems and therefor
turned off by default. You just proved again why that is a good idea.
I'm changing the title and setting the severity to wishlist.

Mike.

> udev version is: 0.021-1
>
> -- Max
>
>
> -- System Information:
> Debian Release: testing/unstable
> APT prefers unstable
> APT policy: (500, 'unstable')
> Architecture: i386 (i686)
> Kernel: Linux 2.6.3-1-686
> Locale: LANG=uk_UA.UTF-8, LC_CTYPE=uk_UA.UTF-8
>
> Versions of packages initscripts depends on:
> ii coreutils 5.0.91-2 The GNU core utilities
> ii dpkg 1.10.19 Package maintenance system for Deb
> ii e2fsprogs 1.35-2 The EXT2 file system utilities and
> ii mount 2.12-6 Tools for mounting and manipulatin
> ii util-linux 2.12-6 Miscellaneous system utilities
>
> -- no debconf information
>
>
>

Message-ID: <20040310122619.GE8798@traveler.cistron.net>
Date: Wed, 10 Mar 2004 13:26:19 +0100
From: Miquel van Smoorenburg <miquels@cistron.nl>
To: Max Kutny <mkut@umc.ua>
Cc: 237056@bugs.debian.org, control@bugs.debian.org
Subject: Re: Bug#237056: bootlogd: conflicts with udev installed

title 237056 bootlogd: should integrate better with udev ?
severity 237056 wishlist
thanks

On 2004.03.09 15:09, Max Kutny wrote:
> Package: initscripts
> Version: 2.85-9
> Severity: important
> 
> Rationale: the boolog doesn't log messages thus unusable.

udev is an experimental package, and bootlogd is turned off by default.
So you have to install an experimental package and fiddle around with
bootlogd to trigger this bug.

> 
> Once I have a udev package installed (a replacement of devfs) I've got
> the following botloog message during startup:
> bootlogd: ioctl(/dev/ttyzf, TIOCCONS): bad file descriptor
> 
> There is no /dev/ttyzf file under udev. I'm not sure this device 
> must be present on the system thus the bug is against bootlog.

Sure, bootlogd is not ready for all production systems and therefor
turned off by default. You just proved again why that is a good idea.
I'm changing the title and setting the severity to wishlist.

Mike.

> udev version is: 0.021-1
> 
> -- Max
> 
> 
> -- System Information:
> Debian Release: testing/unstable
>   APT prefers unstable
>   APT policy: (500, 'unstable')
> Architecture: i386 (i686)
> Kernel: Linux 2.6.3-1-686
> Locale: LANG=uk_UA.UTF-8, LC_CTYPE=uk_UA.UTF-8
> 
> Versions of packages initscripts depends on:
> ii  coreutils                     5.0.91-2   The GNU core utilities
> ii  dpkg                          1.10.19    Package maintenance system for Deb
> ii  e2fsprogs                     1.35-2     The EXT2 file system utilities and
> ii  mount                         2.12-6     Tools for mounting and manipulatin
> ii  util-linux                    2.12-6     Miscellaneous system utilities
> 
> -- no debconf information
> 
> 
>

Revision history for this message

Debian Bug Importer (debzilla) wrote on 2005-12-31:

#14

Message-ID: <email address hidden>
Date: Wed, 10 Mar 2004 14:46:51 +0200
From: Max Kutny <email address hidden>
To: Miquel van Smoorenburg <email address hidden>
Cc: <email address hidden>
Subject: Re: Bug#237056: bootlogd: conflicts with udev installed

On Wed, Mar 10, 2004 at 01:26:19PM +0100, Miquel van Smoorenburg wrote:
> > Severity: important
> >
> > Rationale: the boolog doesn't log messages thus unusable.
>
> udev is an experimental package,

I took it from the unstable yesterday :-E
But please notice, it will be used much more extensive than defvs.

> > Once I have a udev package installed (a replacement of devfs) I've got
> > the following botloog message during startup:
> > bootlogd: ioctl(/dev/ttyzf, TIOCCONS): bad file descriptor
> >
> > There is no /dev/ttyzf file under udev. I'm not sure this device
> > must be present on the system thus the bug is against bootlog.
>
> Sure, bootlogd is not ready for all production systems and therefor
> turned off by default. You just proved again why that is a good idea.
> I'm changing the title and setting the severity to wishlist.

Agree. Of cause it doesn't renders initscripts package unusable. I filed
a bug report thinking of bootlogd instead of initscripts. My bad. I'll be
more careful next time.

Thanks.

-- Max

Revision history for this message

Debian Bug Importer (debzilla) wrote on 2005-12-31:

#15

Message-ID: <email address hidden>
Date: Wed, 10 Mar 2004 15:59:36 +0100
From: Miquel van Smoorenburg <email address hidden>
To: <email address hidden>
Subject: retitle

retitle 237056 bootlogd: should integrate better with udev ?
thanks

Revision history for this message

Debian Bug Importer (debzilla) wrote on 2005-12-31:

#16

Message-ID: <email address hidden>
Date: Thu, 15 Dec 2005 13:26:36 +0100
From: Thomas Hood <email address hidden>
To: <email address hidden>
Subject: Still a problem?

Submitter: Do you still have this problem with the latest udev?

I use udev and bootlogd works fine for me.
--
Thomas Hood

Revision history for this message

Debian Bug Importer (debzilla) wrote on 2005-12-31:

#17

Message-ID: <email address hidden>
Date: Sat, 17 Dec 2005 21:58:10 +0200
From: Max Kutny <email address hidden>
To: Thomas Hood <email address hidden>, <email address hidden>
Subject: Re: Bug#237056: Still a problem?

On Thu, Dec 15, 2005 at 01:26:36PM +0100, Thomas Hood wrote:
> Submitter: Do you still have this problem with the latest udev?
>
> I use udev and bootlogd works fine for me.

It's fine for me too, thanks!

-- Max

Revision history for this message

Debian Bug Importer (debzilla) wrote on 2005-12-31:

#18

Message-ID: <email address hidden>
Date: Sat, 17 Dec 2005 21:39:46 +0100
From: Thomas Hood <email address hidden>
To: <email address hidden>
Subject: Re: Bug#237056: Still a problem?

Max Kutny wrote:
> It's fine for me too, thanks!

Good to hear!

--
Thomas Hood

Revision history for this message

Debian Bug Importer (debzilla) wrote on 2005-12-31:

#19

Message-Id: <email address hidden>
Date: Sat, 24 Dec 2005 15:33:25 -0500
From: Nathanael Nerode <email address hidden>
To: <email address hidden>
Subject: I'm getting this bug.

But only on some machines.
Any ideas? :-P

Revision history for this message

Debian Bug Importer (debzilla) wrote on 2005-12-31:

#20

Message-ID: <email address hidden>
Date: Mon, 26 Dec 2005 19:44:30 +0100
From: Thomas Hood <email address hidden>
To: Nathanael Nerode <email address hidden>, <email address hidden>,
Debian Bug Tracking System <email address hidden>
Subject: Re: [Pkg-sysvinit-devel] Bug#237056: I'm getting this bug.

reopen 237056
stop

Nathanael Nerode wrote:
> But only on some machines.
> Any ideas? :-P

Not without some clues.

--
Thomas

Revision history for this message

Debian Bug Importer (debzilla) wrote on 2005-12-31:

#21

Download full text (3.4 KiB)

Message-ID: <email address hidden>
Date: Sat, 31 Dec 2005 05:47:21 -0500
From: Nathanael Nerode <email address hidden>
To: <email address hidden>, <email address hidden>,
<email address hidden>, <email address hidden>, <email address hidden>
Subject: udev breaks bootlogd

reassign 237056 udev
severity 237056 grave
retitle 237056 udev initscript breaks bootlogd by leaving /dev/pts unmounted
found 237056 0.076-6
thanks

I've found the source of bug 237056. I think it's udev's responsibility,
but you can discuss that amongst yourselves.

bootlogd calls a routine in glibc which tries to get a UNIX98 PTY. If that
fails, it looks for a BSD (legacy) PTY. If that fails (perhaps because
they're not compiled into the kernel, because nothing much uses them), it
gives up and dies.

The problem lies in the rcS.d init scripts. On my system we have:
S01glibc.sh
S02mountvirtfs
S04udev
S05bootlogd
...
S36mountvirtfs
...

/dev/pts is mounted in mountvirtfs.
S05bootlogd runs as early as possible, but after mountvirtfs, because it
wants /dev/pts (and also /proc).

However, S04udev *unmounts* /dev/pts. And it's not remounted until S36.

Now, bootlogd runs after S04udev, presumably to guarantee that the console
device is available. But this means that /dev/pts isn't available.

As it says in S04udev:
# we need to unmount /dev/pts/ and remount it later over the tmpfs

Except it isn't remounted later, until S36. This is really bad because
it could easily break other programs launched in the initscripts in between,
as well. This is why I assigned this bug to udev at grave severity: it
breaks arbitrary other packages.

The bug might be solved in one of the following ways:
* have udev actually remount /dev/pts (and /dev/shm) at the end of its script
  The natural solution. Creates code duplication, though. Unless udev
  invokes /etc/mountvirtfs directly, in which case it has to watch out
  and make sure that 'set -e' is disabled for the invocation.
* drop yet another invocation of mountvirtfs, between udev and bootlogd
  Requires moving udev's script up to S03 to make room. Ugh.
* split mountvirtfs into multiple scripts, one for /proc and /sys which
  runs before udev, and one for /dev/pts and /dev/shm which runs after.
* move bootlogd before udev
  Requires more static devices, *and* might cause other problems (because
  /dev would vanish under the running bootlogd and reappear); but has the
  advantage of logging udev events. Probably the best long term solution,
  but it's unlikely that it will work right now (anyone want to test it?)

As an aside, none of these solutions are robust to parallelized initscripts
-- but then neither are udev or bootlogd, period. Essentially
nothing can safely run at the same time as udev. Essentially all
initscripts should run after bootlogd for bootlogd to make sense -- but only
if bootlogd is present. And these aren't because the later scripts in any
sense 'require' them.

These two scripts fit very poorly into the parallelized initscripts
concept, and probably deserve special treatment for that reason; worth
thinking about. Perhaps every other initscript should be required to
depend on both ude...

Message-ID: <20051231104721.GA8622@twcny.rr.com>
Date: Sat, 31 Dec 2005 05:47:21 -0500
From: Nathanael Nerode <neroden@twcny.rr.com>
To: jdthood@yahoo.co.uk, 23706@bugs.debian.org,
	pkg-sysinit-devel@lists.alioth.debian.org, md@linux.it, control@bugs.debian.org
Subject: udev breaks bootlogd

reassign 237056 udev
severity 237056 grave
retitle 237056 udev initscript breaks bootlogd by leaving /dev/pts unmounted
found 237056 0.076-6
thanks

I've found the source of bug 237056.  I think it's udev's responsibility,
but you can discuss that amongst yourselves.

bootlogd calls a routine in glibc which tries to get a UNIX98 PTY.  If that
fails, it looks for a BSD (legacy) PTY.  If that fails (perhaps because
they're not compiled into the kernel, because nothing much uses them), it
gives up and dies.

The problem lies in the rcS.d init scripts.  On my system we have:
S01glibc.sh
S02mountvirtfs
S04udev
S05bootlogd
...
S36mountvirtfs
...