Stretch to Buster with sysvinit

[Roger Lynn]

Hi,

I have three Stretch AMD64 systems with sysvinit - a desktop and laptop 
running KDE and a headless server. Is there any information available 
anywhere to tell me what will happen when I attempt to upgrade them to 
Buster? The release notes don't mention it and other sources I can find just 
talk about switching from systemd to sysvinit, which doesn't look easy with 
a desktop environment.


The KDE systems have systemd-shim installed, which is not present in Buster. 
Is this going to cause problems? Will the server be okay? Should I just stay 
with Stretch until Bullseye is released or consider moving to Devuan after 
more than 20 years with Debian?


Thanks,

Roger

Re: Missing KDE "Switch user" command in Stretch

[Roger Lynn]

On 29/10/17 23:50, rhkra...@gmail.com wrote:
> On Sunday, October 29, 2017 12:45:34 PM Roger Lynn wrote:
>> So does nobody here use KDE? 
> 
> Just to not leave that questions unanswered, I use KDE, and I suspect many 
> others do, but:
> 
>(1) My everyday machine is still using Wheezy, and
>(2) I haven't ever had occasion to use switch user from the desktop (I do 
> use su at the command line)--but I can see where it might be useful to have 
> something like an anonymous "account" to surf the web (I mean, one different 
> than your everyday userid, email address, etc.)

I presume it's mainly useful when you have several users of a computer. I'm
struggling to imagine why you would need it if you were the only user.

Roger

Re: Missing KDE "Switch user" command in Stretch

[Roger Lynn]

On 29/10/17 20:00, Doug wrote:
> 
> On 10/29/2017 02:44 PM, Eike Lantzsch wrote:
>> KDE on Stretch here.
>> But sorry can't help. My KDE-leave dialog still has "switch user" on Stretch.
>> I used testing until shortly before the release of Stretch. I changed
>> "testing" to "stretch" in the souces.list to stay with Stable for a while.
>> That way I didn't do a real update.
>> But I remember that I had to revert to another theme at some time during my
>> use of testing, because SDDM lost some functionality on the way. Maybe you
>> like to check that out too.

> KDE 5 (Plasma) on PCLinuxOS: Application Menu > Power / Session > Switch 
> User

Either PCLinuxOS or that version of KDE have a different menu structure,
because that's quite different to what I have.

I think I will risk upgrading my wife's laptop and see what happens...

Roger

Re: Missing KDE "Switch user" command in Stretch

[Roger Lynn]

On 29/10/17 19:50, Eike Lantzsch wrote:
> KDE on Stretch here.
> But sorry can't help. My KDE-leave dialog still has "switch user" on Stretch. 
> I used testing until shortly before the release of Stretch. I changed 
> "testing" to "stretch" in the souces.list to stay with Stable for a while. 
> That way I didn't do a real update.
> But I remember that I had to revert to another theme at some time during my 
> use of testing, because SDDM lost some functionality on the way. Maybe you 
> like to check that out too.
> 
> here ->
> Look and Feel: Breeze
> Desktop theme: Air
> Splash screen: Oxygen
> 
> Login Screen (SDDM): Breeze

Thank you for confirming that the command is supposed to be there and
therefore there must be something wrong with my configuration. Unfortunately
changing the theme doesn't seem to make any difference.

Roger

Re: Why does resolv.conf keep changing?

[Roger Lynn]

On 26/10/17 13:40, Stefan Monnier wrote:
>> Given that multiple packages potentially touch/change resolv.conf (at
>> least resolvconf and the various DHCP clients),
> 
> That is not true: when resolvconf is installed, *no package* should
> (modulo bugs) ever change /etc/resolv.conf.
> 
> Instead all changes go through resolvconf, which only modifies
> /run/resolvconf/resolv.conf (and those changes usually get reflected
> into /etc/resolv.conf by making it a symlink to that file, but that's
> not mandatory).
> 
> Stefan "who thinks resolvconf should always be installed"

I think it would be preferable for every package which wants to write to the
resolv.conf file to be required to use the resolvconf package to do so
(which is what I, and I think some other people, thought the situation was
until we read this thread). Uninstalling the resolvconf package would then
mean nothing could write to the resolv.conf file.

I've been in the fortunate position that I've never needed to install the
resolvconf package and nothing has ever tried to modify my resolv.conf file.
I've obviously led a sheltered life.

Roger

Re: Missing KDE "Switch user" command in Stretch

[Roger Lynn]

On 19/10/17 00:43, Roger Lynn wrote:
> Hi,
> 
> When I upgraded to Stretch I lost the "Switch user" command from the Leave
> section of the KDE launcher menu, from the KDE lock screen and from the SDDM
> login screen. This should give the choice of activating an existing session
> or starting a new one. Without it, it is difficult to tell if I am already
> logged in, so I am in danger of creating duplicate sessions.
> 
> The login screen used to have a menu with "Cancel Session", "Remote Login"
> and "Console Login" commands too. These have also disappeared.

So does nobody here use KDE? Or do those of you that use KDE not have
"Switch user" commands and not mind? Or does does it all work perfectly for
you? Should I switch to a different display manager and screen locker, or do
I need a completely different desktop environment?

Roger

Re: Why does resolv.conf keep changing?

[Roger Lynn]

On 24/10/17 08:20, Roberto C. Sánchez wrote:
> So, it happened again at 03:02:26 that resolv.conf was changed.  I
> looked in the logs and found nothing that appeared to be a close
> correlation.  There were several DHCPREQUEST/DHCPACK exchanges before
> and after, but my network has a near constant stream of such exchanges
> with no more than 3 or 4 minutes between exchanges.  It seems unlikely
> to me that the DHCPREQUEST/DHCPACK exchange could be the cuplrit.
> 
> I am fairly out of ideas at this point.  Anyone?

Did you remove the resolvconf package, which is responsible for updating
resolv.conf in dynamic networking environments?

Roger

Re: Can imagemagick really be safely purged or removed?

[Roger Lynn]

On 23/10/17 11:00, Jörg-Volker Peetz wrote:
> Since I use a pure 64-bit system, I overlooked that only one of
> imagemagick-6.q16 or imagemagick-6.q16:i386 can be installed.
> I think it's better to re-install imagemagick-6.q16.
> And then show the output when trying to purge imagemagick, please.
> I think there must be involved a i386 version of some cups package.

Celejar answered the question in the first post. It's very simple:
imagemagick can not be removed because cups-filters depends on it.

Roger

Re: Can imagemagick really be safely purged or removed?

[Roger Lynn]

On 20/10/17 05:00, Celejar wrote:
> The description of the imagemagick package (8:6.9.7.4+dfsg-11+deb9u1)
> on my Stable system includes the statements:
> 
> "This is a dummy package.  You can safely purge or remove it."
> 
> But trying to remove it rips out cups, because:
> 
> ~$ aptitude why imagemagick
> i   cups Depends cups-filters (>= 1.0.24-3~)
> i A cups-filters Depends imagemagick (>= 6.4~)
> 
> What gives?

imagemagick can only be removed when nothing depends on it.
imagemagick-6.q16 has been marked as replacing imagemagick, so most packages
which depend on imagemagick will automatically use imagemagick-6.q16.
However some packages, including cups-filters, depend on particular versions
of imagemagick and the automatic replacement doesn't appear to work in this
case.

You will only be able to remove imagemagick when cups-filters' dependencies
are updated or imagemagick-6.q16 finds a better way of declaring that it
provides imagemagick.

The :386 packages are irrelevant and only need installing in very specific
circumstances.

Roger

Re: attach a big picture

[Roger Lynn]

On 21/10/17 07:10, arne wrote:
> On Sat, 21 Oct 2017 01:36:39 -0400
> Felix Miata  wrote:
>> If you install and run Apache on Debian you can simply provide a URI
>> to an image file on your own Debian PC.
> 
> I am sory but I do not like Apache on my system behind a VPN.
> I am running nginx, and I like to keep it that way. Do you think an
> extra apache server won't hurt in this set-up?

You don't need Apache if you already have nginx installed. (nginx is what I
would use if I needed a web server.)

Roger

Missing KDE "Switch user" command in Stretch

[Roger Lynn]

Hi,

When I upgraded to Stretch I lost the "Switch user" command from the Leave
section of the KDE launcher menu, from the KDE lock screen and from the SDDM
login screen. This should give the choice of activating an existing session
or starting a new one. Without it, it is difficult to tell if I am already
logged in, so I am in danger of creating duplicate sessions.

The login screen used to have a menu with "Cancel Session", "Remote Login"
and "Console Login" commands too. These have also disappeared.

The only workaround I can find is to manually check every virtual console to
see if I have a session running there. This is not a suitable solution for
other users. I think my son is currently logged in three times and is
probably trying to edit the same file in each KDE session.

What additional information do I need to supply? How can I get the commands
back? Have I configured something wrongly? Is this a bug in KDE? If so, it
would appear to be a serious regression since the version in Jessie.

Thanks,

Roger

Re: Client daemon for sorting e-mail via IMAP

[Roger Lynn]

On 23/01/13 10:20, Jochen Spieker wrote:
 Chris Davies:
 Alois Mahdal alois.mahdal.1-ndm...@zxcvb.cz wrote:
 
 Since I'm using various clients, filtering using rules in MUA is not
 practical.  I would prefer to have all this logic in a single place,
 (namely my personal VPS box), where for example a script would exist
 just for purpose of regularly checking new mail and moving the
 new messages to given folders based on pre-defined rules.
 
 Exim4 with dovecot (IMAP) and sieve works well for me.
 
 (Almost the) same here. Postfix+Dovecot with Sieve. The only thing I
 miss is the ability to run external programs like crm114 for spam
 filtering. It is possible to do that with Sieve as well, but I do not
 know of any nice way to train the filter.

And here - Exim with Dovecot and Sieve. Many mail clients include an
extension to manage Sieve filters. I run a global Spamasssin from Exim,
which means that definite spam can be rejected at SMTP time and maybe spam
can be first greylisted and then filtered into a Junk folder. I use the
Dovecot Antispam plugin to train Spamassin. It can be configured to work
with a variety of spam filters. Seems to work well.

Roger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/8o7dt9-ih9@silverstone.rilynn.me.uk

Re: Samba usershare errors

[Roger Lynn]

On 24/01/13 21:50, Igor Cicimov wrote:
 On 25/01/2013 3:51 AM, Roger Lynn ro...@rilynn.me.uk
 mailto:ro...@rilynn.me.uk wrote:
 My smb.conf file looks like this. Several similar share definitions have
 been omitted for brevity.

 [global]
 workgroup = FUNDAMENTALS
 server string = %h server
 interfaces = 127.0.0.0/8 http://127.0.0.0/8, bond0
 bind interfaces only = Yes
 obey pam restrictions = Yes
 pam password change = Yes
 unix password sync = Yes
 syslog = 0
 log file = /var/log/samba/log.%m
 max log size = 1000
 load printers = No
 os level = 65
 preferred master = Yes
 domain master = Yes
 dns proxy = No
 wins support = Yes
 panic action = /usr/share/samba/panic-action %d
 idmap config * : backend = tdb
 invalid users = root
 [Service]
 comment = Service files
 path = /srv/smb/service
 read only = No
 create mask = 0775
 force create mode = 0664
 directory mask = 0770
 force directory mode = 0770

 I am also getting lots of cups errors like this:
 [2013/01/24 15:29:32.978276,  0] printing/print_cups.c:110(cups_connect)
   Unable to connect to CUPS server localhost:631 - Connection refused
 [2013/01/24 15:29:32.978505,  0]
 printing/print_cups.c:487(cups_async_callback)
   failed to retrieve printer list: NT_STATUS_UNSUCCESSFUL

 I presume this is because cups is not installed. Is there any way to stop
 Samba from continuously trying access it?

 Remove printer section from config.

I haven't got a printer section. The only print related command I know of is
load printers = No.

Thanks,

Roger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/0a8dt9-ej9@silverstone.rilynn.me.uk

Samba usershare errors

[Roger Lynn]

Hi,

I am running the Debian package of Samba 2:3.6.6-4 on an up to date Wheezy
server. I am getting a lot of errors logged similar to this:

log.sophie-pc:[2013/01/24 08:38:38.848419,  0]
param/loadparm.c:9114(process_usershare_file)
log.sophie-pc:  process_usershare_file: stat of
/var/lib/samba/usershares/servic failed. Permission denied
log.sophie-pc:[2013/01/24 08:38:38.849233,  0]
param/loadparm.c:9114(process_usershare_file)
log.sophie-pc:  process_usershare_file: stat of
/var/lib/samba/usershares/servic failed. No such file or directory
log.sophie-pc:[2013/01/24 08:38:38.849679,  0]
param/loadparm.c:9114(process_usershare_file)
log.sophie-pc:  process_usershare_file: stat of
/var/lib/samba/usershares/servic failed. No such file or directory

As far as I know usershares are disabled. The clients are running a variety
of recent versions of Windows. It most often seems to happen with PDF files,
but there are others too.

My smb.conf file looks like this. Several similar share definitions have
been omitted for brevity.

[global]
workgroup = FUNDAMENTALS
server string = %h server
interfaces = 127.0.0.0/8, bond0
bind interfaces only = Yes
obey pam restrictions = Yes
pam password change = Yes
unix password sync = Yes
syslog = 0
log file = /var/log/samba/log.%m
max log size = 1000
load printers = No
os level = 65
preferred master = Yes
domain master = Yes
dns proxy = No
wins support = Yes
panic action = /usr/share/samba/panic-action %d
idmap config * : backend = tdb
invalid users = root
[Service]
comment = Service files
path = /srv/smb/service
read only = No
create mask = 0775
force create mode = 0664
directory mask = 0770
force directory mode = 0770
[Sophie]
comment = Home Directories
path = /home/sophie/share
invalid users = root, manfred
read only = No
create mask = 0775
force create mode = 0444
directory mask = 0775
force directory mode = 0555

What might be causing the above errors?

I am also getting lots of cups errors like this:
[2013/01/24 15:29:32.978276,  0] printing/print_cups.c:110(cups_connect)
  Unable to connect to CUPS server localhost:631 - Connection refused
[2013/01/24 15:29:32.978505,  0] printing/print_cups.c:487(cups_async_callback)
  failed to retrieve printer list: NT_STATUS_UNSUCCESSFUL

I presume this is because cups is not installed. Is there any way to stop
Samba from continuously trying access it?

Thank you,

Roger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51015e3c.7050...@rilynn.me.uk

Re: sshd_config match keyword syntax

[Roger Lynn]

On 30/08/12 16:20, Brian wrote:
 On Thu 30 Aug 2012 at 14:37:34 +0100, Roger Lynn wrote:
 I want to force everyone except members of a particular group to run sftp
 when they ssh into a server. So at the end of /etc/ssh/sshd_config I have:
 
 Match Group !sshers
   ForceCommand /usr/lib/openssh/sftp-server
 
 However I can't get the group negation to work. If I remove the '!' it works
 as expected, in that members of sshers are forced to run sftp. With the '!'
 the condition is never met, no one is forced to run sftp and the whole
 stanza appears to do nothing.
 
 The documentation on the Match keyword is not very helpful, but it appears
 that the above should be allowed. What am I doing wrong? Is it a bug?
 
 Two questions. I'll go for the first one. First read the PATTERNS section
 of ssh_config(5). Then think about
 
Match Group *
 
 and
 
Match Group *,!sshers

Thank you to Camaleón and to you, this line works.

I did read the PATTERNS section, but it didn't help very much. Although it's
not explicitly stated I infer that the comma operates as a logical 'and'.
What is not clear is exactly what the Group criterion matches or how the
negation operator works.

If Group sshers matches all members of the 'sshers' group then I would
assume that Group !sshers would match all users who are not in that group.
However that seems not to be the case. It appears that it actually means
something like don't match members of the sshers group, which means you
need the counter-intuitive *, in front of it to get the expected effect.
This is not stated anywhere.

I find the logic strange and difficult to follow but the documentation is
just awful.

Thanks again,

Roger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/f1lfh9-ecd@silverstone.rilynn.me.uk

sshd_config match keyword syntax

[Roger Lynn]

Hi,

I want to force everyone except members of a particular group to run sftp
when they ssh into a server. So at the end of /etc/ssh/sshd_config I have:

Match Group !sshers
  ForceCommand /usr/lib/openssh/sftp-server

However I can't get the group negation to work. If I remove the '!' it works
as expected, in that members of sshers are forced to run sftp. With the '!'
the condition is never met, no one is forced to run sftp and the whole
stanza appears to do nothing.

The documentation on the Match keyword is not very helpful, but it appears
that the above should be allowed. What am I doing wrong? Is it a bug?

I am running openssh-server 1:6.0p1-2 on a new Wheezy installation.

Thanks,

Roger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/503f6c9e.4090...@rilynn.me.uk

Re: PC speaker stopped working after upgrade to Squeeze

[Roger Lynn]

On 05/10/11 01:30, Roger Lynn wrote:
 On 04/10/11 20:10, Curt wrote:
 On 2011-10-02, Roger Lynn ro...@rilynn.me.uk wrote:
 When I upgraded my Lenny AMD64 system to Squeeze earlier this year, the
 PC speaker (ie motherboard buzzer) stopped working. To get it to work I
 have to remove and re-add the pcspkr kernel module using modprobe every
 time I reboot. The only relevant bits I can find in dmesg are:
 
 Same thing happened to me. I never figured it out. 
 
 Looks like this bug, though, now that I'm googling again:
 
 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=604197
 
 The workaround is to add the line blacklist pcspkr to
 /etc/modprobe.d/blacklist.conf and the line pcspkr to /etc/modules,
 which has for effect to load the module later (like when you reload it,
 I guess).  Then it works first time every time.
 
 Thank you, that looks like exactly the same problem, and the workaround
 seems logical. I'll try it and see what happens next time I reboot,
 although I don't do that very often.

I forgot to report back that this did have the desired effect. I just
need to remember to take it out again next time I upgrade.

Roger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/bqhqm8-dh9@silverstone.rilynn.me.uk

Re: PC speaker stopped working after upgrade to Squeeze

[Roger Lynn]

On 04/10/11 14:00, Camaleón wrote:
 On Mon, 03 Oct 2011 23:24:35 +0100, Roger Lynn wrote:
 On 03/10/11 14:50, Camaleón wrote:
 On Sun, 02 Oct 2011 22:04:09 +0100, Roger Lynn wrote:

 It means there is a PC Speaker device detected by the system, which is
 fine.
 
 But it only works the second time. Could something be disabling it at
 boot, which is then overridden when I reload the module?
 
 Then you mean pc speaker _works_ but you need to reload (i.e., 
 modprobe -r pcspkr  modprobe pcspkr) the module to get it working? 
 Anyway, you can check if the module is loaded when you boot by issuing 
 lsmod | grep pcspkr.

Yes. pcspkr is loaded at boot, but it doesn't work until I remove and
reload it.

 Is this be a kernel or udev bug, or just a weird hardware / software
 configuration combination?
 
 Maybe a mix of them. I still have not clear why it works in some
 computers and doesn't in others.
 
 It used to work on this one until Squeeze.
 
 I'm also starting to think this could be something related to alsa or the 
 sound card settings. What's your sound card chipset?

The motherboard manual says, Realtek ALC889A codec. lspci says,
00:14.2 Audio device: ATI Technologies Inc SBx00 Azalia (Intel HDA)
and 01:00.1 Audio device: ATI Technologies Inc HD48x0 audio.

Thanks,

Roger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/5adsl8-o3f@silverstone.rilynn.me.uk

Re: [OT] British vs. American English

[Roger Lynn]

On 04/10/11 10:20, Richard wrote:
 Perhaps go to the link provided and read through and you will find a table of 
 al the HVDC links, there are
 plenty in all continents.
 Especially for grid sharing between countries.
 AC 3 phase links are unsuitable for this purpose, If the governments of the 
 EU can not agree on measures to
 halt the problems with the Euro, do you really expect them to be able to 
 agree on the exact phase of a AC
 link between countries.
 You CAN NOT just connect adjacent suppliers AC system together unless the 
 phase is matched.
 http://en.wikipedia.org/wiki/List_of_HVDC_projects

That is true, but it hasn't stopped most of Europe's electricity
networks from being synchronised, and not just between members of the
EU. See
http://en.wikipedia.org/wiki/Synchronous_grid_of_Continental_Europe

The synchronous grid of Continental Europe is the largest synchronous
electrical grid (by connected power) in the world. It is interconnected
a single phase-locked 50 Hz mains frequency electricity grid that
supplies over 400 million customers in 24 countries, including most of
the European Union.

There are many undersea HVDC connections around Europe and some
proposals for long distance HVDC lines.

Roger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/19hsl8-fgf@silverstone.rilynn.me.uk

Re: PC speaker stopped working after upgrade to Squeeze

[Roger Lynn]

On 04/10/11 20:10, Curt wrote:
 On 2011-10-02, Roger Lynn ro...@rilynn.me.uk wrote:
 Hi,

 When I upgraded my Lenny AMD64 system to Squeeze earlier this year, the
 PC speaker (ie motherboard buzzer) stopped working. To get it to work I
 have to remove and re-add the pcspkr kernel module using modprobe every
 time I reboot. The only relevant bits I can find in dmesg are:
 
 Same thing happened to me. I never figured it out. 
 
 Looks like this bug, though, now that I'm googling again:
 
 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=604197
 
 The workaround is to add the line blacklist pcspkr to
 /etc/modprobe.d/blacklist.conf and the line pcspkr to /etc/modules,
 which has for effect to load the module later (like when you reload it,
 I guess).  Then it works first time every time.

Thank you, that looks like exactly the same problem, and the workaround
seems logical. I'll try it and see what happens next time I reboot,
although I don't do that very often.

Thanks again,

Roger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/3dgsl8-qdf@silverstone.rilynn.me.uk

Re: PC speaker stopped working after upgrade to Squeeze

[Roger Lynn]

On 03/10/11 14:50, Camaleón wrote:
 On Sun, 02 Oct 2011 22:04:09 +0100, Roger Lynn wrote:
 When I upgraded my Lenny AMD64 system to Squeeze earlier this year, the
 PC speaker (ie motherboard buzzer) stopped working. 
 
 Are you using GNOME?

No. I actually use KDE, but as this is a problem on console before KDE
(or KDM) have been run, I don't think that should affect it. The KDE
notifications are set to use the system bell and that works if the
console bell is working.

 IIRC, beeper started to use canberra and associated libraries to play 
 motherboard's beep sound and so it outputs to PC speakers instead to the 
 boards one. Have you checked if the PC speaker output volume is at high 
 level?

There is a beep entry in AlsaMixer, but it doesn't seem to do anything.
There are no packages with canberra in the name installed.

 To get it to work I have to remove and re-add the pcspkr kernel module
 using modprobe every time I reboot. 
 
 (...)
 
 Is that relevant?
 
 Relevant for what? :-?

I quoted some text from dmesg about some sort of PCBeep device that I've
never heard of. I wondered if it might have anything to do with my lack
of a working bell.

 It means there is a PC Speaker device detected by the system, which is 
 fine.

But it only works the second time. Could something be disabling it at
boot, which is then overridden when I reload the module?

 My kernel is currently linux-image-2.6.32-5-amd64 2.6.32-37 and the
 motherboard is a Gigabyte GA-MA790FXT-UD5P.
 
 Why do I need to reload the pcspkr module? 
 
 If you are in GNOME, most sure because of the above explanation.
 
 Is there any way I can get it to work from boot? 
 
 You can get the module to be automatically loaded at booting by adding 
 the corresponding module (pcspkr) into /etc/modules.

pcspkr is already being loaded at boot. The problem is that it doesn't
work until I remove it and load it again. Is adding a module which is
already being loaded to /etc/modules really likely to make any difference?

 Is this be a kernel or udev bug, or just a weird hardware / software
 configuration combination?
 
 Maybe a mix of them. I still have not clear why it works in some 
 computers and doesn't in others.

It used to work on this one until Squeeze.

Thanks for your assistance,

Roger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/3appl8-0i9@silverstone.rilynn.me.uk

PC speaker stopped working after upgrade to Squeeze

[Roger Lynn]

Hi,

When I upgraded my Lenny AMD64 system to Squeeze earlier this year, the
PC speaker (ie motherboard buzzer) stopped working. To get it to work I
have to remove and re-add the pcspkr kernel module using modprobe every
time I reboot. The only relevant bits I can find in dmesg are:

[2.616528] udev[376]: starting version 164
[2.861193] input: PC Speaker as /devices/platform/pcspkr/input/input2
[2.904056] ACPI: WMI: Mapper loaded

during boot, and:

[  885.175967] input: PC Speaker as /devices/platform/pcspkr/input/input8

when re-inserting the module.

There is also:

[4.117058] hda_codec: ALC889A: BIOS auto-probing.
[4.118461] input: HDA Digital PCBeep as
/devices/pci:00/:00:14.2/input/input6
[4.122153] HDA Intel :01:00.1: PCI INT B - GSI 19 (level, low)
- IRQ 19

Is that relevant?

My kernel is currently linux-image-2.6.32-5-amd64 2.6.32-37 and the
motherboard is a Gigabyte GA-MA790FXT-UD5P.

Why do I need to reload the pcspkr module? Is there any way I can get it
to work from boot? Is this be a kernel or udev bug, or just a weird
hardware / software configuration combination?

Thanks for any help,

Roger


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4e88d1c9.7090...@rilynn.me.uk

Samba matchname: host name/address mismatch: ::ffff:127.0.0.1 != localhost

[Roger Lynn]

Hi,

Since upgrading from etch to lenny when it came out, all my Samba logs have
been full of lines like:

[2010/08/13 15:07:11,  0] lib/util_sock.c:matchname(1749)
  matchname: host name/address mismatch: :::127.0.0.1 != localhost
[2010/08/13 15:07:11,  0] lib/util_sock.c:get_peer_name(1870)
  Matchname failed on localhost :::127.0.0.1

As accessing files from Windows clients still works I've been able to work
around it, so I've ignored it, but now I think it's preventing me from
adding printers. There are lots of similar reports around, but I can't find
any solutions. I am using only Debian Lenny packages with security updates
and proposed updates. I am using dnsmasq for DNS. My smb.conf file is below.
I have not knowingly done any customisation or configuration relating to IPv6.

Is this a bug in my configuration, samba or somewhere else, possibly dnsmasq?

Obviously this is IPv6 related. As I can't find any way of disabling IPv6 in
Samba, the simplest solution appears to be to disable IPv6 on the entire
machine. Has anyone got any better suggestions?

Thanks,

Roger


smb.conf:

[global]
workgroup = FOO
server string = %h server
obey pam restrictions = Yes
passdb backend = tdbsam
pam password change = Yes
passwd program = /usr/bin/passwd %u
passwd chat = *Enter\snew\s*\spassword:* %n\n
*Retype\snew\s*\spassword:* %n\n *password\supdated\ssuccessfully* .
unix password sync = Yes
syslog = 0
log file = /var/log/samba/log.%m
max log size = 1000
load printers = No
dns proxy = No
panic action = /usr/share/samba/panic-action %d
invalid users = root
hosts allow = 127., 192.168.1.
hosts deny = ALL

[homes]
comment = Home Directories
path = /home/%S/share
read only = No
create mask = 0755
force create mode = 0444
force directory mode = 0444
browseable = No

[foo]
path = /srv/smb/foo
read only = No
create mask = 0777
force create mode = 0666
directory mask = 0777
force directory mode = 0666


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4c6568b1.5020...@rilynn.me.uk