date:20081211

Bug#428733: Coming back on this locking problem or samba with executables called from logon scripts

2008-12-11 Thread Arno van Amersfoort


Hello Christian,


Christian Perrier wrote:

retitle 428733 Samba keeps locks on executable files when launched from logon 
scripts by Windows clients
tags 428733 moreinfo
thanks

Hello Arno,

While doing a mass bug triage of bugs in the samba package, I went on
the issue you reported in http://bugs.debian.org/428733, where
executable files launched from Windows during logon scripts were kept
locked by samba.

I went on your very very long smb.conf file and first piped it through
testparm so that only parameters that differ from the default settings
are kept. See attached file.

To narrow down this problem, I would first recommend you to drop the
following parameters unless you have a very good reason for having
them:

[global]
smb ports = 139
  

I've already disabled this a while back.

name resolve order = lmhosts hosts bcast
  

I disabled this one too now.

deadtime = 10
  

I also already this a while back.

socket options = IPTOS_LOWDELAY TCP_NODELAY SO_SNDBUF=8192 
SO_RCVBUF=8192
  

Same thing.

enhanced browsing = No
  

Same thing.

strict locking = No
  
I have disabled this now. Now that I think of it this could potentiolly 
fix the problem.

dos filetime resolution = Yes
  

I disabled this too.

From discussions with Samba developers, tweaking those is generally
not recommendedwhatever you may find on so-called well-informed
sites (the socket options, specifically)

I'd also like to know whether you still experience that problem and
are in position to do further testing. At the moment you reported the
problem, i was mostly ignored, unfortunately, after you provided a
very big level 10 debug log. 
  
Now that I've changed some of the settings you suggested I'll recheck 
whether it fixes the problem. I'll report back ASAP.


Thanks for your efforts.

cheers,

Arno



  


--
Ing. A.C.J. van Amersfoort (Arno)
Electronics & ICT Engineer

Leiden Institute of Physics (LION), Electronics Department (ELD)
Huygens Laboratory (Room 1007), Leiden University
Postal Address: P.O. Box 9504, 2300 RA Leiden
Visit Address : Niels Bohrweg 2, 2333 CA Leiden
The Netherlands

Phone: +31-(0)71-527.1894
Fax  : +31-(0)71-527.5819
E-mail   : a.c.j.van.amersfo...@eld.physics.leidenuniv.nl
Homepage : http://rocky.eld.leidenuniv.nl




--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#508489: installation-reports: [hppa] invalid module format when loading xfs.ko

2008-12-11 Thread Christian Perrier

Quoting Stillwell, Bryan (bryan.stillw...@hp.com):

> During the install I wanted to create an XFS partition on /dev/sdb, but
> that option wasn't available.  After the install I tried doing this by
> hand, but I couldn't even load the xfs module:
> 
> # modprobe xfs
> FATAL: Error inserting xfs
> (/lib/modules/2.6.26-1-parisc/kernel/fs/xfs/xfs.ko): Invalid module
> format
> 
> In dmesg I see:
> [17182155.252000] module xfs relocation of symbol xfs_btree_read_bufs is
> out of range (0x3ffefffe in 17 bits)
> 
> This looks pretty similar to bug #350482:
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=350482


Could you test with the daily builds from 
http://www.debian.org/devel/debian-installer?



signature.asc
Description: Digital signature

Bug#500556: [Python-modules-team] Bug#500556: ipython: just want to thank for the progress ; -)

2008-12-11 Thread Stephan Peijnik

On Thu, 2008-12-11 at 22:26 -0500, Yaroslav Halchenko wrote:
> I was going to ask why they are not yet in unstable, but then saw in NEW
> queue
> foolscap  0.3.2.dfsg-1
>   source all  unstable
>   1 day   Debian Python Modules Team  #499699
> 
> so soon soon soon ;)
> 
> can we hope that as soon as foolscap appears in unstable we get 0.9
> ipython uploaded along? ;-)

ipython 0.9.1 has actually already been prepared for uploading to
experimental (due to the lenny freeze, see [0]) as soon as foolscap
enters unstable. 

As noted in the bug report (see [1]) you can download both foolscap and
ipython packaged from my personal repository at [2].
I have been using the combination of both packages for more than a month
now and haven't found a single problem, so you should be safe to install
them.

-- Stephan

[0]
http://svn.debian.org/wsvn/python-modules/packages/ipython/trunk/?op=log
[1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=500556#20
[2] http://archive.sp-its.at/

signature.asc
Description: This is a digitally signed message part

Bug#502562: libpam-poldi: new upstream version 0.4

2008-12-11 Thread Lionel Elie Mamane

On Fri, Oct 17, 2008 at 11:17:41PM +0200, Joachim Breitner wrote:
> Am Freitag, den 17.10.2008, 23:06 +0200 schrieb Lionel Elie Mamane:

>> Please package the new upstream version;

> My main issue is that scdaemon is a meant to be run as a user
> daemon, but is now used as part of system authentication. So it
> needs at least a way to be configured system wise (WRT smartcard
> driver), and the security implications have to be considered.

> Moritz Schulte said he’d do a bugfix release shortly after that, but
> hasn’t done so yet. CC’ing him: Moritz, what’s the state of this?

Is there any news on that front?

-- 
Lionel



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#508532: aptitude: Would like to be able to check orphaned status before installing

2008-12-11 Thread Daniel Dickinson

Package: aptitude
Version: 0.4.11.10-1lenny1.1+b1
Severity: wishlist


It'd be nice if you could see that a package was orphaned before installing it.

-- Package-specific info:
aptitude 0.4.11.10 compiled at Oct 27 2008 21:47:55
Compiler: g++ 4.3.2
Compiled against:
  apt version 4.6.0
  NCurses version 5.6
  libsigc++ version: 2.0.18
  Ept support enabled.

Current library versions:
  NCurses version: ncurses 5.6.20080830
  cwidget version: 0.5.12
  Apt version: 4.6.0
linux-gate.so.1 =>  (0xb7f6c000)
libapt-pkg-libc6.7-6.so.4.6 => /usr/lib/libapt-pkg-libc6.7-6.so.4.6 
(0xb7e96000)
libncursesw.so.5 => /lib/libncursesw.so.5 (0xb7e58000)
libsigc-2.0.so.0 => /usr/lib/libsigc-2.0.so.0 (0xb7e51000)
libcwidget.so.3 => /usr/lib/libcwidget.so.3 (0xb7d8d000)
libept.so.0 => /usr/lib/libept.so.0 (0xb7ccc000)
libxapian.so.15 => /usr/lib/libxapian.so.15 (0xb7b76000)
libz.so.1 => /usr/lib/libz.so.1 (0xb7b61000)
libpthread.so.0 => /lib/i686/cmov/libpthread.so.0 (0xb7b48000)
libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0xb7a59000)
libm.so.6 => /lib/i686/cmov/libm.so.6 (0xb7a33000)
libgcc_s.so.1 => /lib/libgcc_s.so.1 (0xb7a26000)
libc.so.6 => /lib/i686/cmov/libc.so.6 (0xb78cb000)
libutil.so.1 => /lib/i686/cmov/libutil.so.1 (0xb78c7000)
libdl.so.2 => /lib/i686/cmov/libdl.so.2 (0xb78c2000)
/lib/ld-linux.so.2 (0xb7f6d000)
Terminal: screen
$DISPLAY is set.
`which aptitude`: /usr/bin/aptitude
aptitude version information:

aptitude linkage:

-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: i386 (i686)

Kernel: Linux 2.6.26-1-686 (SMP w/1 CPU core)
Locale: LANG=en_CA.UTF-8, LC_CTYPE=en_CA.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages aptitude depends on:
ii  apt [libapt-pkg-libc6. 0.7.18Advanced front-end for dpkg
ii  libc6  2.7-16GNU C Library: Shared libraries
ii  libcwidget30.5.12-3  high-level terminal interface libr
ii  libept00.5.22High-level library for managing De
ii  libgcc11:4.3.2-1 GCC support library
ii  libncursesw5   5.6+20080830-2shared libraries for terminal hand
ii  libsigc++-2.0-0c2a 2.0.18-2  type-safe Signal Framework for C++
ii  libstdc++6 4.3.2-1   The GNU Standard C++ Library v3
ii  libxapian151.0.7-4   Search engine library
ii  zlib1g 1:1.2.3.3.dfsg-12 compression library - runtime

Versions of packages aptitude recommends:
ii  aptitude-doc-en [apt 0.4.11.10-1lenny1.1 English manual for aptitude, a ter
ii  libparse-debianchang 1.1.1-2 parse Debian changelogs and output

Versions of packages aptitude suggests:
pn  debtags(no description available)
ii  tasksel   2.77   Tool for selecting tasks for insta

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#508531: emacspeak: [INTL:it] Italian translation of the debconf templates

2008-12-11 Thread Vincenzo Campanella

Package: emacspeak
Version: 28.0.5
Severity: wishlist
Tags: l10n patch

Enclosed please find the Italian translation of the Debconf template.

Best regards
vince


# ITALIAN TRANSLATION OF EMACSPEAK-28.0.5'S.PO-DEBCONF FILE
# COPYRIGHT (C) 2008 THE EMACSPEAK'S COPYRIGHT HOLDER
# This file is distributed under the same license as the emacspeak package.
#
# Vincenzo Campanella , 2008.
msgid ""
msgstr ""
"Project-Id-Version: emacspeak-28.0.5\n"
"Report-Msgid-Bugs-To: emacsp...@packages.debian.org\n"
"POT-Creation-Date: 2008-12-02 18:19+0100\n"
"PO-Revision-Date: 2008-12-03 08:09+0100\n"
"Last-Translator: Vincenzo Campanella \n"
"Language-Team: Italian \n"
"MIME-Version: 1.0\n"
"Content-Type: text/plain; charset=UTF-8\n"
"Content-Transfer-Encoding: 8bit\n"
"X-Generator: KBabel 1.11.4\n"

#. Type: select
#. Description
#: ../templates:3001
msgid "Default speech server:"
msgstr "Server vocale predefinito:"

#. Type: select
#. Description
#: ../templates:3001
msgid ""
"The /etc/emacspeak.conf file will be configured so that the command '/usr/"
"bin/emacspeak' will start Emacs with emacspeak support using this server."
msgstr ""
"Il file «/etc/emacspeak.conf» verrà configurato in modo che il comando "
"«/usr/bin/emacspeak» avvierà Emacs con il supporto a emacspeak usando "
"questo server."

#. Type: select
#. Description
#: ../templates:3001
msgid ""
"You may change the selection later by running 'dpkg-reconfigure emacspeak' "
"as root, or temporarily override the selection by setting the environment "
"variable DTK_PROGRAM."
msgstr ""
"È possibile modificare successivamente questa selezione eseguendo "
"«dpkg-reconfigure emacspeak» come root, o sovrascrivere temporaneamente "
"questa selezione impostando la variabile di ambiente DTK_PROGRAM."

#. Type: string
#. Description
#: ../templates:4001
msgid "Hardware port of the speech generation device:"
msgstr "Porta hardware del dispositivo per la generazione vocale:"

#. Type: string
#. Description
#: ../templates:4001
msgid ""
"If a hardware device is used to generate speech, please enter the Unix "
"device file associated with it, such as' /dev/ttyS0' or '/dev/ttyUSB0'"
msgstr ""
"Se viene usato un dispositivo hardware per la generazione vocale, inserire "
"il file di device Unix associato a esso, come per esempio «/dev/ttyS0» o "
"«/dev/ttyUSB0»"

#. Type: string
#. Description
#. Translators, please do not translate "none"
#: ../templates:4001
msgid "If you use a software method to generate speech, please enter 'none'."
msgstr "Se si usa un metodo software per la generazione vocale, inserire «none»."

#. Type: error
#. Description
#: ../templates:5001
msgid "${port} is not a character special device"
msgstr "${port} non è un device speciale a caratteri"

#. Type: string
#. Description
#: ../templates:6001
msgid "Users of speech server:"
msgstr "Utenti del server vocale:"

#. Type: string
#. Description
#: ../templates:6001
msgid ""
"Users must be members of group ${group} to access the speech server "
"connected to ${port}. Please review the space-separated list of current "
"members of that group, and add or remove usernames if needed."
msgstr ""
"Gli utenti devono essere membri del gruppo ${group} per poter accedere "
"al server vocale connesso alla porta ${port}. Controllare l'elenco, che è "
"separato da spaziature, dei membri attuali di questo gruppo ed eventualmente "
"aggiungere o rimuovere nomi di utenti."

#. Type: string
#. Description
#: ../templates:6001
msgid ""
"If you later add users to the system, you can either reconfigure the "
"emacspeak package afterwards, or enroll the user in ${group} with 'adduser "
"${group} '."
msgstr ""
"Se si desiderano aggiungere successivamente degli utenti, sarà possibile o "
"riconfigurare il pacchetto emacspeak, o aggiungere l'utente al gruppo ${group} "
"eseguendo «adduser ${group} »."

#. Type: string
#. Description
#: ../templates:6001
msgid ""
"Group membership is checked at login time, so new members must log out and "
"log in again before using the speech server."
msgstr ""
"L'appartenenza al gruppo viene verificata al momento dell'accesso al sistema, "
"per cui i nuovi membri dovranno necessariamente uscire e rientrare prima di "
"poter utilizzare il server vocale."

#. Type: error
#. Description
#: ../templates:7001
msgid "Invalid username ${user}"
msgstr "Nome utente ${user} non valido"

#. Type: error
#. Description
#: ../templates:7001
msgid "There is no user named ${user}, so no such user could be added to ${group}."
msgstr "Non esiste alcun utente con il nome ${user}, per cui l'utente non ha potuto essere aggiunto a ${group}."

#. Type: error
#. Description
#: ../templates:8001
msgid "${port} non-writable by unprivileged users"
msgstr "La porta ${port} non è scrivibile da utenti senza privilegi"

#. Type: error
#. Description
#: ../templates:8001
msgid ""
"Since the speech device is connected to ${port}, unprivileged users must "
"have read/write access to that device."
msgstr ""
"Poiché il device voc

Bug#508530: bugzilla: [INTL:it] Italian translation of the debconf templates

2008-12-11 Thread Vincenzo Campanella

Package: bugzilla
Version: 3.2.0.0-rc2.2
Severity: wishlist
Tags: l10n patch

Enclosed please find the Italian translation of the Debconf template.

Best regards
vince

# ITALIAN TRANSLATION OF BUGZILLA-3.2.0.0-RC2.2'S.PO-DEBCONF FILE
# COPYRIGHT (C) 2008 THE BUGZILLA'S COPYRIGHT HOLDER
# This file is distributed under the same license as the bugzilla package.
#
# Vincenzo Campanella , 2008.
msgid ""
msgstr ""
"Project-Id-Version: bugzilla-3.2.0.0-RC2.2\n"
"Report-Msgid-Bugs-To: bugzi...@packages.debian.org\n"
"POT-Creation-Date: 2008-12-02 18:17+0100\n"
"PO-Revision-Date: 2008-12-03 07:47+0100\n"
"Last-Translator: Vincenzo Campanella \n"
"Language-Team: Italian \n"
"MIME-Version: 1.0\n"
"Content-Type: text/plain; charset=UTF-8\n"
"Content-Transfer-Encoding: 8bit\n"
"X-Generator: KBabel 1.11.4\n"

#. Type: password
#. Description
#: ../bugzilla3.templates:2001
msgid "Password confirmation:"
msgstr "Conferma della password:"

#. Type: string
#. Description
#: ../bugzilla3.templates:3001
msgid "Email address of Bugzilla administrator:"
msgstr "Indirizzo di posta elettronica dell'amministratore di Bugzilla:"

#. Type: string
#. Description
#: ../bugzilla3.templates:3001
msgid ""
"Please enter the email address of Bugzilla administrator; all mail for the "
"administrator will be sent to this address. This email address is also used "
"as the administrator login for Bugzilla."
msgstr ""
"Inserire l'indirizzo di posta elettronica dell'amministratore di Bugzilla; "
"tutti i messaggi destinati all'amministratore verranno inviati a questo "
"indirizzo. Questo indirizzo verrÃ  utilizzato anche per l'autenticazione "
"dell'amministratore di Bugzilla."

#. Type: string
#. Description
#: ../bugzilla3.templates:3001
msgid ""
"A valid address must contain exactly one '@', and at least one '.' after the "
"@. You'll be able to change this setting through Bugzilla's web interface."
msgstr ""
"Un indirizzo di posta elettronica valido deve contenere esattamente un Â...@Â» e "
"almeno un Â«.Â» dopo Â...@Â». Questa impostazione potrÃ  essere successivamente "
"modificata attraverso l'interfaccia web di Bugzilla."

#. Type: string
#. Description
#: ../bugzilla3.templates:4001
msgid "Real name of Bugzilla administrator:"
msgstr "Vero nome dell'amministratore di Bugzilla:"

#. Type: password
#. Description
#: ../bugzilla3.templates:5001
msgid "Password for the Bugzilla administrator account:"
msgstr "Password dell'amministratore di Bugzilla:"

#. Type: boolean
#. Description
#: ../bugzilla3.templates:6001
msgid "Did you customize the Status/Resolutions values?"
msgstr "I valori Â«Status/ResolutionÂ» sono stati personalizzati?"

#. Type: boolean
#. Description
#: ../bugzilla3.templates:6001
msgid ""
"If you customized values in the Status/Resolution field, you must edit /usr/"
"share/bugzilla3/lib/checksetup.pl before running it."
msgstr ""
"Se i valori nel campo Â«Status/ResolutionÂ» sono stati personalizzati, Ã¨ "
"necessario modificare Â«/usr/share/bugzilla3/lib/checksetup.pl prima di "
"eseguire il programma."

Bug#461447:

2008-12-11 Thread Dmitrijs Ledkovs

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I have packaged 0.5 version as a NMU against Sid.
I have written an email to the package maintainer about it.

.dsc .chages and diff.gz are attached to the launchpad bug here:
https://bugs.launchpad.net/debian/+source/cgmail/+bug/184700

The package was build against Sid using pbuilder.
It was tested on my machine which is running Ubuntu Intrepid.

The source package has also been uploaded to my PPA and it is available here
https://launchpad.net/~dmitrij.ledkov/+archive


With best regards



Dmitrijs Ledkovs


-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)

iEYEARECAAYFAklCCY8ACgkQh/npNECdyEao6gCdF+sMm/K7KZMtrLaWDP3dAnIJ
FGoAnieYTH+hlDGo6FKk68UjKgp+sL0c
=crm4
-END PGP SIGNATURE-



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#508529: jailtool: Does not copy hidden files

2008-12-11 Thread Daniel Dickinson

Package: jailtool
Version: 1.1-4
Severity: normal


Subject says it all

-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: i386 (i686)

Kernel: Linux 2.6.26-1-686 (SMP w/1 CPU core)
Locale: LANG=en_CA.UTF-8, LC_CTYPE=en_CA.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages jailtool depends on:
ii  perl  5.10.0-18  Larry Wall's Practical Extraction 
ii  rsync 3.0.3-2fast remote file copy program (lik

jailtool recommends no packages.

jailtool suggests no packages.

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#508505: liblocale-maketext-lexicon-perl: please install xgettext.pl in /usr/bin

2008-12-11 Thread Gunnar Wolf

gregor herrmann dijo [Thu, Dec 11, 2008 at 11:13:16PM +0100]:
> On Thu, 11 Dec 2008 23:34:21 +0200, Damyan Ivanov wrote:
> 
> > > It would be nice to install xgettext.pl (currently provided in
> > > examples/) in /usr/bin.  It's only a simple wrapper script, but it has
> > > several features the `xgettext' (from gettext) does not provide, e.g.
> > > support for Template Toolkit sources.
> > The problem with installing that in /usr/bin is that
> > (1) .pl extension is to be avoided, and if we obey that rule,
> > (2) the package would have to conflict with gettext, on which depend 
> > ~50 other packages.
> 
> Maybe we could use some dpkg-divert stuff?

Well, diverting a widely used binary against a completely unknown
binary that has no need to have that same name is a no-go as
well... But it could be shipped as /usr/bin/xgettext_tt or something
like that? It does not need to have the same name!

> > How about installing xgettext.pl as /usr/bin/xgettext-lexicon? That 
> > could work, provided someone writes a man page for it :)
> 
> That, too.
> 
> Are they both compatible?

Oh, already suggested :)

-- 
Gunnar Wolf - gw...@gwolf.org - (+52-55)5623-0154 / 1451-2244
PGP key 1024D/8BB527AF 2001-10-23
Fingerprint: 0C79 D2D1 2C4E 9CE4 5973  F800 D80E F35A 8BB5 27AF



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#239111: following etch->lenny grub upgrade instructions locks up a system with XFS root

2008-12-11 Thread Robert McQueen

severity 239111 grave
found 239111 0.97-47
thanks

Sorry to drag this ancient bug up to RC status, but I just got *totally*
screwed by it. I've just upgraded an etch system to lenny, and following
the instructions in grub's NEWS.Debian file, I ran 'grub-install (hd0)'
after verifying my device.map in order to update the bootloader, under
the threat that my system would not be able to boot a >= 2.6.23 kernel.

Unfortunately, because of this grub-install hanging while my XFS root
filesystem is frozen, my entire (remote, headless) system is now
unresponsive on all of the ssh logins I had open to it, as well as on
its serial console, because any login activity ends up trying to write
to disk. (I could sysrq or powercycle it, but I'm not actually certain
whether the bootloader will be intact currently. Thankfully its the dom0
of a Xen system, and the mission-critical domUs are working still, so I
can risk a reboot over the weekend.)

Previously this bug has just been a minor annoyance because under d-i,
you can work around this freeze by going to the shell (which is in
memory in a different filesystem) and killing grub-install, unfreezing
the filesystem, and doing it manually. However, this is the first time
I've been enticed into using grub-install on a live system, and now I've
realised quite what a screwup this actually is. It's moved from a mild
annoyance for deviants like me who insist on choosing grub from the d-i
menu when using an XFS root, to a way to lock-up and potentially even
brick (I can tell you tomorrow evening) a running system.

As I noted, er, four and a half years ago[1], and by many previous
reporters, what the script currently does will *always* fail where /boot
and / are the same filesystem:
xfs_freeze -f
grub --bullshit... # tries to write to the filesystem and log file too,
# will never make any progress because / is frozen
xfs_freeze -u
On a live system, not only does this just hang the grub-install process,
but it effectively stops you from being able to log in to fix the
situation too.

1: http://lists.debian.org/debian-boot/2004/06/msg02301.html

Previous reporters have provided patches[2] to change the behaviour to this:
xfs_freeze -f
xfs_freeze -u
grub --bullshit...
The failures of this approach (see below) are that grub might not
succeed in upgrading/installing, or that it might install an older
version of grub. It will, however, never freeze your system.

2:
http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=152;filename=xfs_freeze.diff;att=1;bug=239111

Earlier reporters on the bug in 2006 commented that the above doesn't
work reliably. This is due to #306966, which is present in etch's
kernel, and means xfs_freeze -f can return before the data is actually
synced. I've spoken to the XFS developers about this on IRC, and
apparently it was fixed upstream around 2.6.19 or 20. So actually,
applying that patch to re-order things like that will make this freezing
bug go away on lenny kernel system once and for all, and d-i won't need
to force LILO on XFS users any more. :)

However it /will/ re-introduce the race condition so that grub-install
fails when you're running under etch's kernel, but that's /much/ less of
a problem than locking up or bricking the system, because you can work
around it by just installing grub manually. So, definitely for lenny we
must apply a patch which fixes the script to do freeze / unfreeze / grub.

The only remaining issue is then: how old a grub can really not read the
2.6.26 kernel that comes with lenny?

If etch's grub /can/ read lenny's kernel, and it's only /older/ grubs
that have problems, then this race bug will affect a pretty small number
of people. We can just edit NEWS.Debian then to say "If grub-install
fails and you have XFS /boot and etch's kernel, reboot and try it
again". Any people with >3 year old Debian systems with XFS root and
grub are probably able to figure out how to reinstall grub manually.

If etch's grub /can't/ read lenny's kernel, then I guess we need to
provide a smoother migration path for people who are doing as I did and
running grub-install, so we can apply the patch with the sleep
workaround[3] and include a message:
"If you got a file not found error and use an XFS root filesystem,
please try re-running grub-install with the --sync-sleep=60 option to
work around a bug in the 2.6.18 kernel in etch."

3:
http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=179;filename=xfs_freeze.diff;att=1;bug=239111

Regards,
Rob

P.S. If we were being "enterprise", we'd probably push an etch kernel
patch which fixed the race condition with xfs_freeze, so people could
reliably upgrade their kernel and grub at the same time before rebooting
into lenny. But, eh, screw it.

P.P.S. The XFS developers would like to take this opportunity to remind
viewers that reading/writing the block device underneath a filesystem is
the true bug here, and that the grub shell should just be made to use
FIBMAP to find the stage 1.5

1 2 3 >

1 - 100 of 247 matches

Mail list logo