Re: Thoughts on Ansible? [was: Thoughts on Anible?]

[Osamu Aoki]

Hi,


On Sat, Aug 19, 2017 at 02:32:45PM -0400, RavenLX wrote:
> ...
> 
> > I think use of ansible or any similar tool is not prerequisite of
> > "development".  It's a configuration management system.   It's a nice
> > and interesting tool I am thinking to learn but I don't use it yet.
> 
> The biggest thing with me is my memory. In that if it takes too much to do a
> task (ie. Ansible taking several lines to make a directory vs. Python taking
> one line), then I tend to forget how things are done and it gets confusing.
> 
> > You need C, Shell, Perl, Python, git, ... skills first for
> > development.
> 
> Great suggestions! Thank you. I do have some C programming skills but not
> that great. I can get around the CLI pretty well, I'm proficient with Perl,
> am learning Pythong. I had used git but I don't really have much to share
> (right now) and so I don't have an account anymore there. Most of my stuff
> is for my work, which doesn't really share stuff (though I could share my
> code if I wish). I'm not thinking of going into a side-hobby of programming.
> At work I manage a web server, and pretty much am more comfy with Perl and
> Bash and now Python. But I also want to be sure to keep up with the times,
> so to speak. So I wondered if other admins recommended Ansible as a "must
> have skill" or just optional.

You are now talking different things.  "developer" --> "admins"

If you are managing multiple servers as profession, you need to make such
process recorded and reproducible.  That's what configuration management
system is for and it is becoming one of the very basic tool to know.

Good luck.

Osamu

Re: Thoughts on Anible?

[Osamu Aoki]

Hi,



On Fri, Aug 18, 2017 at 07:01:03PM -0400, RavenLX wrote:
> Please forgive my goofy questions. I am not really that well versed in CM
> (Configuration Management) and this post is really going to show it.
> 
> Because this will more than likely be tltr (too long to read), I'll try to
> make it as fun as possible. Forgive if my humor is a bit strange.
> 
> I noticed "Anible" isn't in the debian repos. Seems to be a Red Hat
> scripting engine for creating automated installs.

Just a fact correction.

I think you are mistyping.  Ansible is in the repo of Debian:

$ apt show ansible
Package: ansible
Version: 2.2.1.0-2
Priority: optional
Section: admin
Maintainer: Harlan Lieberman-Berg 
Installed-Size: 12.8 MB
Depends: python-crypto (>= 2.6), python-jinja2, python-paramiko, 
python-pkg-resources, python-yaml, python:any (<< 2.8), python:any (>= 
2.7.5-5~), python-httplib2, python-netaddr
Recommends: python-kerberos, python-selinux, python-winrm (>= 0.1.1), 
python-xmltodict
Suggests: cowsay, sshpass
Homepage: https://www.ansible.com
Tag: admin::automation, admin::configuring, admin::file-distribution,
 admin::package-management, implemented-in::python,
 interface::commandline, role::program, use::configuring,
 works-with::software:running
Download-Size: 1,675 kB
APT-Sources: http://deb.debian.org/debian stretch/main amd64 Packages
Description: Configuration management, deployment, and task execution system
 Ansible is a radically simple model-driven configuration management,
 multi-node deployment, and remote task execution system. Ansible works
 over SSH and does not require any software or daemons to be installed
 on remote nodes. Extension modules can be written in any language and
 are transferred to managed machines automatically.

...
> I heard about "Ansible" and how it is supposed to make it easier to write
> installation scripts.

Correct spelling ;-)

I think use of ansible or any similar tool is not prerequisite of
"development".  It's a configuration management system.   It's a nice and
interesting tool I am thinking to learn but I don't use it yet.

You need C, Shell, Perl, Python, git, ... skills first for development.

Osamu

Re: how to make bootable live wheezy USB that doesn't use isohybrid

[Osamu Aoki]

Hi,

On Tue, Jun 14, 2016 at 08:50:28AM -0400, Jeffrey Mark Siskind wrote:
> I'd like to make a live wheezy USB dongle. I followed the instructions on

Why wheezy ... it's old.

>https://www.debian.org/CD/faq/#write-usb
> 
> to make a live USB from
> 
>
> http://cdimage.debian.org/mirror/cdimage/archive/7.11.0-live/amd64/iso-hybrid/debian-live-7.11.0-amd64-standard.iso

There are many ways to make a bootable HDD image if this can be booted.

By using aufs or newer overlayfs, you can choose to make OS persistent
while your running system writing on RAM(tmpfs).

 https://wiki.debian.org/DebianEeePC/HowTo/InstallOnSDcardOrUsbStick
 I have not tested but this may give you hint.

I also use kvm to install system on USB connected HDD from my working
Debian system.

many ways ...

Re: Extundelete

[Osamu Aoki]

Hi,

On Tue, Feb 16, 2016 at 08:49:33AM -0500, Susan Cragin wrote:
> Fooling around, lost "susan" folder. Want folder back. May have
> recursively deleted everything in it. Want txt org and a couple other
> extensions.  (I've got days-old backup but want to try recovering
> files.) So, extundelete or something else?  or Which Linux-on-a-stick
> has the best recovery tools?

0. Boot from CD/USK-key/another partition
1. Use dd to take disk image of the filesystem
2. Move image to anothe machine
3. Install testdisk to this machine
4. Recover data

Re: removing TexLive Docs packages

[Osamu Aoki]

Hi,

On Wed, Dec 23, 2015 at 10:27:25AM -0500, The Wanderer wrote:
> On 2015-12-23 at 09:55, Udyant Wig wrote:
> 
> > On 12/23/2015 12:15 PM, Sivaram Neelakantan wrote:
> > 
> >> I installed TexLive recently and during one of the upgrades, the 
> >> entire texdoc package got installed too.  that is humoungous and
> >> I'd like to remove the doc packages alone and also stop future
> >> installs of the same type whenever I Upgrade TexLive.
> > 
> >   An appreciable problem.

I used to worry this when I were on P5 system etc.
Now disk is cheap, so I do not care much. But ...

> >   # To prevent these packages from getting installed later:
> >   # aptitude hold '~ntexlive~ndoc'
> 
> Does this latter actually work?
> 
> I would expect it to use the same hold mechanism as 'apt-mark hold' or
> 'echo "packagename hold" | dpkg --set-selections', and in my experience
> holding a package as not installed does not work; it doesn't even get
> recorded in the package-selection state, and certainly doesn't get
> respected later when another package Depends or Recommends that package.
> 
> If that's been changed, or if there's a mechanism which avoids it and
> does work, I'd be very interested.

I agree.  I tested "hold" on an uninstalled package with aptitude,
nothing happened.

So your option is to use /etc/apt/preferences as described in
apt_preferences(5) where new glob(7) syntax for package is described.

This is not tested but it should be like:

   Package: /^texlive-.*-doc$/
   Pin: release a=*
   Pin-Priority: 10

This glob feature is available from apt (0.8.14)  Apr 2011.

Osamu

Re: removing TexLive Docs packages

[Osamu Aoki]

Hi,

This fixes error in my previous post.

On Wed, Dec 23, 2015 at 10:27:25AM -0500, The Wanderer wrote:
> On 2015-12-23 at 09:55, Udyant Wig wrote:
> >   # To prevent these packages from getting installed later:
...
> If that's been changed, or if there's a mechanism which avoids it and
> does work, I'd be very interested.

>From root account:
 # cat >/etc/apt/preferences.d/block-texlive << EOF
Package: /^texlive-.*-doc$/
Pin: release a=*
Pin-Priority: -1
EOF

This worked.  (Priority 10 was not small enough.)

Osamu

Re: discuss debian 's attitude to ppa

[Osamu Aoki]

Hi,

On Sat, May 23, 2015 at 01:58:16PM +0800, mudongliang wrote:
 Yesterday I installed the LMDE(Linux Mint based Debian Jessie),which is a
 rolling release issued by Linux Mint!

LMDE is based on Debian but not exactly a Debian.  It's Linux Mint.

 The installation is of no question! But when I want to install openyoudao
 from ppa , it returns me not support ppa!
 
 So I search it ! Although I don't find some real proof , but I can see one
 thing ! Maybe debian will not support ppa ,

PPA created by Debian and based on Debian will be supported but that
still has not been created.  Random PPA style archive is not supported
although some binaries may work.

 because in the talk of Neil McGovern, he says some weakness of ppa! And from
 the development of Debian ,
 ppa's development seems to be not able to ensure its stablity!

I do not know what you are talking about, but I thought Neil is pushing
to create PPA asa paert of service.

Osamu


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20150523124316.GA19222@goofy.local

Re: Wifi Problem

[Osamu Aoki]

Hi,

On Thu, Mar 12, 2015 at 09:35:13PM +0400, Gajadur Dwijesh wrote:
 Hello..i have Debian wheezy 7.8 KDE installed on my laptop..i am not able
 to get wifi connection..can someone help please?
 Thank you

Debian does not install non-free things in its default installation.

Unfortunately, many wifi devices has binary blobs which makes them
non-free.

But we still provide support them via non-free archive.

Please read: (I used Google)
 https://wiki.debian.org/WiFi/HowToUse
 http://linuxconfig.org/install-intel-wireless-firmware-on-debian-7-wheezy
 
http://linuxforcynics.com/how-to/install-wifi-firmware-and-wpa-support-in-debian
 https://wiki.debian.org/Firmware
 
Osamu


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20150314141023.GA20619@goofy.local

Re: fakeroot to build a package?

[Osamu Aoki]

On Thu, Feb 19, 2015 at 08:53:14AM +0200, Eugene Zhukov wrote:
 Hi Osamu,
 
 On Wed, Feb 18, 2015 at 5:35 PM, Osamu Aoki os...@debian.org wrote:
  On Tue, Feb 17, 2015 at 02:10:57PM +0200, Eugene Zhukov wrote:
  Hello,
 
  I came across this building tutorial [1]. It advertises using
  fakeroot debian/rules binary
  command to build a package. Needless to say it doesn't work for all 
  packages.
  I find this tutorial confusing. I know of two other IMHO better pages
  on the topic [2] and [3].
  Did I miss something or the first wiki needs fixing.?
 
  [1] https://wiki.debian.org/BuildingTutorial
  [2] https://wiki.debian.org/IntroDebianPackaging
  [3] https://www.debian.org/doc/manuals/maint-guide/
 
  These are good ones :-)  (I wrote recent portion of maint-guide)
 
 I updated [1] to use debuild where possible for consistency and
 clarity with other tutorials.

As I browsed, its use of dpatch seems to be very outdated in the day of
deb file format 3.0 (patch).
 
  But multi-arch description in maint-guide is weak due to the limitation
  of dh_make command.
 
  You may also wish to install debmake package and read its documentation.
  This is rather new tool and may have some rough edges.
 
  Regards,
 
  Osamu
 
  Debmake documentaion is available at:
https://people.debian.org/~osamu/maint-guide.html
 
 This debmake link is very useful, thanks!

I will update its example section with autotool/cmake examples :-)

Osamu


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20150221123741.GA3437@goofy.local

Re: fakeroot to build a package?

[Osamu Aoki]

On Tue, Feb 17, 2015 at 02:10:57PM +0200, Eugene Zhukov wrote:
 Hello,
 
 I came across this building tutorial [1]. It advertises using
 fakeroot debian/rules binary
 command to build a package. Needless to say it doesn't work for all packages.
 I find this tutorial confusing. I know of two other IMHO better pages
 on the topic [2] and [3].
 Did I miss something or the first wiki needs fixing.?
 
 [1] https://wiki.debian.org/BuildingTutorial
 [2] https://wiki.debian.org/IntroDebianPackaging
 [3] https://www.debian.org/doc/manuals/maint-guide/

These are good ones :-)  (I wrote recent portion of maint-guide)

But multi-arch description in maint-guide is weak due to the limitation
of dh_make command.

You may also wish to install debmake package and read its documentation.
This is rather new tool and may have some rough edges.

Regards,

Osamu

Debmake documentaion is available at:
  https://people.debian.org/~osamu/maint-guide.html

(I may remove this later once it is on debian.org site...)


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20150218153526.GA28623@goofy.local

Re: apt-get source download has files not in git repository

[Osamu Aoki]

HI,

On Thu, Nov 20, 2014 at 10:05:08PM -1000, Joel Roth wrote:
 On Fri, Nov 21, 2014 at 08:03:58AM +0200, Andrei POPESCU wrote:
  On Jo, 20 nov 14, 12:29:32, Joel Roth wrote:
...
 Yes, they are. I'm curious at which step they get generated.

Your original post should have dbus in the command:

 $ apt-get source dbus

All files in .pc directories are generated when the source package is
unpacked.  Please read the dpkg-source manpage.  You are unpacking the
most common source format 3.0 (quilt).

 I found the files listed above also need to be present to 
 build the package. 

If you unapply patches, you can safely remove files in .pc.  These files
are used to track patching state of the source tree.  So the source tree
with all the patches unapplied does not have these files under .pc.

Osamu


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20141122150647.GA1892@goofy.local

Re: ifupdown -- it ain't so bad

[Osamu Aoki]

Hi,

On Thu, Sep 04, 2014 at 09:00:01PM -0400, Podrigal, Aron wrote:
 On Thu, Sep 4, 2014 at 2:17 PM, Brian a...@cityscape.co.uk wrote:
 
  On Thu 04 Sep 2014 at 14:06:28 -0400, Podrigal, Aron wrote:
 
   I had experienced some difficulties using ifupdown in the past, I than
  came
   across this while searching the web
  
  http://pureperl.blogspot.com/2013/01/the-debian-ifupdown-package-and.html
   my question is, is that still true? or there was updates to the package
   since than?
  
   also what would be a good alternative for service networking restart ?
 
  Is what still true?
 
 What this article claims
 http://pureperl.blogspot.com/2013/01/the-debian-ifupdown-package-and.html
 That ifupdown is not reliable. Basicly I see  errors when bringing up my
 vlan interfaces


Facts:

ifupdown is mere important package.  So you can have system without it
if your required packages does not depend on it.

This is just a blog post by someone.  He does not seem to be active
promoting any package of his.  Besides he puts down the coding style of
ifupdown which I thought is a very elegant one.  If he does not like
ifupdown, I guess other tool may be used.

ifupdown has its share of problem but now it can handle IPv6 too.

It is designed to be very flexible.  Most of problem can be handled
easily by its customization.  Besides, it does not prohbit you to put
init-rc script with ip commend etc.

Osamu


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20140907135432.GA2589@goofy.local

Re: Does LXDE really require lightdm?

[Osamu Aoki]

Hi,

On Sat, Jun 21, 2014 at 08:31:50PM -0400, Steve Litt wrote:
 Hi all,
 
 I installed LXDE on a no-X, no-desktop virgin network Wheezy 64bit
 install with non-free software allowed, and on the next boot it went
 into lightdm. The only thing I could find that installed and required
 lightdm was LXDE. I uninstalled LXDE, installed Xfce, installed
 whatever bestows startx, and bang, X from the CLI command line, no *dm
 needed.

I think you should learn to use aptitude to look-into Debian's
resources.  Here are the answer by running aptitude.

 1) Am I correct that Debian's LXDE package installs lightdm?

It depends on what yopu mean by LXDE package.  

If you mean task-lxde-desktop, yes it is depends.

If you mean lxde, practically yes since it is recommends.

 2) Does that come from the LXDE project, or is it a Debian thing?

  Homepage: https://launchpad.net/lightdm
  http://www.freedesktop.org/wiki/Software/LightDM/

 3) Is there a way to turn off LXDE's install of lightdm?

If you chose lxde, you install without recommends.  That is easy with
aptitude and apt-gey can do that via command line.  Read the manual
pages of them.

 The whole reason I'm switching from Xubuntu to Debian is to get away
 from both Plymouth and *dm. Fortunately, I find LXDE desireable, but no
 way do I find it necessary.

You can go less with bare Openbox window manager :-)

Osamu


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20140623145237.GA14440@goofy

Re: sudo in X-environment -- polkit solution

[Osamu Aoki]

Hi,

I am wondering why you even need to use wrapper explicitly?

On Sat, May 03, 2014 at 12:43:17PM +0200, Ralf Mardorf wrote:
 On Sat, 2014-05-03 at 06:29 -0400, Tom H wrote:
  Thanks. As I said earlier, I consider this a bug. What's the point of
  using gksu/gksudo if you have do use a wrapper that you could use
  around su/sudo?

If I type system-config-printer to my user shell, I get GUI running
with root privilege :-)
 
 You still will get a GUI to enter the password. 

Yes, I enter my user password.

 When running su/sudo by
 a launcher, you at least need to run it in a terminal emulation.

No.  GUI program or console program without fancy wrapper can start with
the proper privilege, here.

  Maybe pkexec is the solution?

Yes.

Did you add yourself as a member of sudo group?
 https://www.debian.org/doc/manuals/debian-reference/ch04.en.html#_policykit
 4.6.2. PolicyKit

PolicyKit is an operating system component for controlling system-wide
privileges in Unix-like operating systems.

Newer GUI applications are not designed to run as privileged processes.
They talk to privileged processes via PolicyKit to perform
administrative operations.

PolicyKit limits such operations to user accounts belonging to the sudo
group on the Debian system.

I think this is helping me :-)  

(The above text may be obsoleted soon by logind.)

Regards,

Osamu


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20140504150847.GA20950@goofy

Re: Question about wheezy-backports

[Osamu Aoki]

Hi,

On Sat, May 03, 2014 at 11:28:22PM -0600, Paul E Condon wrote:
 I have added to my sources.list the following line:
 
 deb http://ftp.us.debian.org/debian/ wheezy-backports main
 
 But according to http://backports.debian.org/Instructions/, that won't
 make the backport appear in interactive aptitude, 

I do not see this?

It says: All backports are deactivated by default.

So aptitude in full screen interactive mode does not show it as the default
choice in the package list mode but you can select it by going into 1
level from the package list mode.  There you can select the backported
packages.

 or be automatically 'upgraded' to the backported version.  

True.

 To actually install the
 backport, I also have to add the phrase -t wheezy-backports to the
 install command. 

If you are using command line mode.  True.  The same goes to apt-get.

 And I have to know that there is a backport for the
 particular deb in which I have an interest. I suppose I could try
 reading the debian-backports-announce mailinglist, but I am so seldom
 bothered with having the latest version that I can't believe I will do
 that. Is there also a simple list of backported debs that I can browse
 on the web to know what is available, or a wiki pointer? 

See Andrei's post for the right solution.  If you are much lazier, just
bump apt preferences to make them automatically installed. See:

 $ man apt_preferences

Regards,

Osamu


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20140504153254.GB20950@goofy

Re: dist-upgrade installs new package, lists it as autoremove candidate

[Osamu Aoki]

Hi,

On Sat, Apr 26, 2014 at 03:29:49PM +0200, Carlo wrote:
 I would tell you a my tip:
 
 - if you want go from stable to testing/unstable before you must
 change your channel in repository's file located in
 /etc/apt/sources.list.
 
 - after this edit, these are the right sequence of command by terminal:
 
 # apt-get update
 # apt-get upgrade
 # apt-get install apt dpkg aptitude
 # aptitude safe-upgrade
???
 # aptitude full-upgrade
???
 # update-grub

Where did you get this idea of using aptitude, Carlo?  Is this in
current release/installation guide?  Curious.  safe-upgrade may be
harmless.  But full-upgrade had some odd behavior cases.

(OP may have been hit by multiarch related issue.  Your suggestion of
apt-get install ... or similar may help avoid some of them.)

I thought we are not recommending to use the command line use of
aptitude for the release-to-release upgrade.  (I love aptitude in
console terminal view.)

Regards,

Osamu


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20140426151838.GD29369@goofy

Re: Installing several OSes from a single flash drive

[Osamu Aoki]

On Sat, Jan 11, 2014 at 12:30:50AM +, d...@661.org wrote:
 
 I've been tinkering with the idea of installing different operating
...

 This tinkering exposed a problem with Debian's hd-install boot
 images. As is pointed out in Miron's article, an hd-install image
 will check the boot media for a Debian ISO from which to draw
 packages.  There is a flaw in the implementation in that the
 installer will mount EVERY ISO it finds to the same mount point.  i

Sounds like some code in debian-installer.

 http://packages.qa.debian.org/d/debian-installer.html
 http://qa.debian.org/developer.php?login=debian-b...@lists.debian.org
 https://wiki.debian.org/DebianInstaller/CheckOut
 https://wiki.debian.org/DebianInstaller/Git
 http://d-i.alioth.debian.org/doc/internals/

If you fix issues, file a bug report to it or discuss with people there
at debian-b...@lists.debian.org

Good luck,

Osamu


-- 
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/20140125083710.GB2540@goofy

Re: POP3 softwares.

[Osamu Aoki]

Hi,

On Sat, Jan 18, 2014 at 09:46:15PM -0800, Peter Easthope wrote:
 Problem
 Given a remote POP3 server which provides SSL.  Also a MUA on a
 local host, which retrieves by POP3 when activated by the user and
 lacks SSL.  Try to move messages from the server to the MUA.
 
 Candidate Solution
 Let getmail, in the POP3-over-SSL instance, retrieve messages from

Good choice for getmail :-)

 the server and queue in an mbox file.  

OK.

 Run qpopper to allow the MUA
 to retrieve from the mbox on demand.

??? Why bother with qpopper.  It is a local mbox.

 Questions
 1. Rather than getmail and qpopper, can one software do this task?
If so, which is recommended?

You are using getmail + qpopper + some local MUA

Just make things simple by dropping qpopper.

 2. The documentation for qpopper suggests that it can only acquire
messages from a spool.  If the answer to 1 is no, is there
some way to pipe messages directly from the retrieving POP3 to
the delivering POP3?

By the way, mbox spool is not always good for you if you use noatime
filesystem.  getmail supports maildirs etc which may be better if your
MUA supports it.

I actually use mailfilter to split mail into many different mail box.
See:
  
http://www.debian.org/doc/manuals/debian-reference/ch06.en.html#_the_remote_mail_retrieval_and_forward_utility
  
http://www.debian.org/doc/manuals/debian-reference/ch06.en.html#_mail_delivery_agent_mda_with_filter

Osamu (getmail maintainer)


-- 
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/20140125091829.GC2540@goofy

Re: Confused about dist-upgrade

[Osamu Aoki]

Hi,

On Fri, Jan 24, 2014 at 10:25:01AM +, Brian wrote:
 On Fri 24 Jan 2014 at 11:54:12 +0200, Andrei POPESCU wrote:
 
  The recommended way is specific to each release and thoroughly 
  documented in the Release Notes.
 
 Which is where I got what I wrote from:

I bet it was only for the recent ones.
 
   If the system being upgraded provides critical services for your
   users or the network[2], you can reduce the downtime if you do a
   minimal system upgrade, as described in Section 4.4.5, “Minimal
   system upgrade”, followed by a kernel upgrade and reboot, and then
   upgrade the packages associated with your critical services.
   Upgrade these packages prior to doing the full upgrade described
   in Section 4.4.6, “Upgrading the system”. This way you can ensure
   that these critical services are running and available through the
   full upgrade process, and their downtime is reduced.
 
 Alternatively:
 
   In some cases, doing the full upgrade (as described below)
   directly might remove large numbers of packages that you will want
   to keep. We therefore recommend a two-part upgrade process: first
   a minimal upgrade to overcome these conflicts, then a full upgrade
   as described in Section 4.4.6, “Upgrading the system”
 
 Sections 4.1.3. and 4.4.5. of
 
   http://www.debian.org/releases/stable/i386/release-notes/ch-upgrading.html

Have you heard glibc or perl transitions.  Those complication may happen again. 
Also there are more in the Release Notes.  namely,

http://www.debian.org/releases/stable/i386/release-notes/ch-information.en.html

I think we are doing our best avoid complications but the canonical
place for the recommended way is the Release Notes.  Nothing else.

So as a baseline, process you described is fine.  But Andrei's point is
very important.

Osamu


-- 
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/20140125115146.GD2540@goofy

Re: Debianly Correct place to add ~/bin to $PATH ? a new problem

[Osamu Aoki]

Hi,

On Tue, Dec 10, 2013 at 04:05:55PM -0700, Bob Proulx wrote:
 Paul E Condon wrote:
  Andrei POPESCU wrote:
   Most display managers don't source ~/.profile (or /etc/profile). I tried 
   to discuss this with developers, but there was hardly any interest so I 
   settled for sourcing them in ~/.xsessionrc instead (which is sourced by 
   the X startup scripts):

Though I have different idea what it should be... inconsistency is the
major problem.
 
 It has always annoyed me that logging in with the graphical login
 managers *by default* do not source the .profile (or .bash_profile)
 files.  And worse when some distros have tried to set this up they
 have sometimes made a mess of things.  (For example SuSE did a hard
 source of . ~/.profile /dev/null 21 with all errors redirected to
 avoid noise in the case of bash versus not-bash syntax.)  Red Hat
 actually has done the best job of this.  I wish Debian would do it
 similarly.

Well, .profile sourcing broke some system.  Changing things around these
thing randomly is real headache ... there shoulkd be some standard.

   ,[ ~/.xsessionrc ]
   | . /etc/profile
   | . ~/.profile
   `

For me, I gave up using those.  These changes with upstream quite often
and unreliable.  I just use .bashrc now to set my shell.  Any higher
level environment variable settings are done via PAM.  /etc/environment

 Lots of files are possible.  This is topic drive.
...
  The bottom line is that I don't have an .xsessionrc , and therefore
  its not a good place for me to put profile files sourcing. 
  
  Suggestions?
 
 If you want it then simply create it.  Andrei told you everything you
 need to know.  Specifically that his file had exactly two lines in it
 and it worked for him to load the two files that he wanted loaded.

Yes.

 I as well gave an alternative solution which accomplishes the task in
 a different way.  My preferred method starts up the shell as a login
 shell and therefore the shell will load up all files that the the
 shell loads up at login time.  The end effect will be very similar.
 The choice of either of those or yet another method is yours to make.

Yes.  That is one way.  Important thing is to have simple and direct
method to do what you wish to get.  What worked at one point breaks,
that is the time to trethink your approach.

Osamu



-- 
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/20140125120551.GE2540@goofy

Re: Re (2): POP3 softwares.

[Osamu Aoki]

Hi,

On Sat, Jan 25, 2014 at 05:42:29AM -0800, pe...@easthope.ca wrote:
 Date: Sat, 25 Jan 2014 18:18:29 +0900
 From: Osamu Aoki osamu_aoki_h...@nifty.com
  ??? Why bother with qpopper.  It is a local mbox.
 
 Ideally the one mbox should work, but I'm not sure that the MUA and 
 getmail will each respect access by the other.  qpopper will respect  
 getmail in access to the mbox and respond to the MUA with POP3. This 
 at the cost of a second mbox and delays up to the retrieval interval 
 of qpopper.

I see.  If qpopper can do it via locking why not make your local MUA to
do the same.

maildrop package comes with lockmail for such need.  (I suspect procmail
comes with similar util.)  You just have to use such thing.

Also, use of maildir format instead of mbox avoids locking issue.

Osamu


-- 
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/20140125160844.GB23831@goofy

Re: nano vs VI

[Osamu Aoki]

On Thu, Jan 02, 2014 at 11:17:24AM +, Curt wrote:
 On 2014-01-02, Paul Cartwright pbcartwri...@gmail.com wrote:
  On 01/01/2014 08:01 PM, John Hasler wrote:
  man select-editor
  man select-editor
  No manual entry for select-editor
 
 I think you need to install 'sensible-utils', which provides
 
 /usr/bin/sensible-browser
 /usr/bin/sensible-editor
 /usr/bin/sensible-pager
 
 I guess it's the sensible thing to do (I haven't done it!)
 
 Oops; he said 'select-editor'.  Mmm, maybe he meant what I 'read' 'cause I
 can't find select-editor anywhere.

I never heard about it :-) but I have it here.

$ select-editor

Select an editor.  To change later, run 'select-editor'.
  1. /bin/ed
  2. /usr/bin/emacs23
  3. /usr/bin/mcedit
  4. /usr/bin/vim.basic
  5. /usr/bin/vim.gtk
  6. /usr/bin/vim.nox
  7. /usr/bin/vim.tiny

Choose 1-7 []: 6
$ dpkg -S select-editor
sensible-utils: /usr/bin/select-editor
sensible-utils: /usr/share/man/man1/select-editor.1.gz

As I see its changelog, it is only in testing/sid:

sensible-utils (0.0.9) unstable; urgency=low

  * Fix bashism in select-editor
Patch by Thorsten Glaser
Closes: #711321

 -- Anibal Monsalve Salazar ani...@debian.org  Thu, 06 Jun 2013 22:19:39 +1000

sensible-utils (0.0.8) unstable; urgency=low

  * Merge 0.0.7ubuntu1
- add select-editor (Closes: #482774)
- modify sensible-editor to pick the editor from select-editor
  * Removes reference to non-existent BROWSER documentation in environ(7)
Patch by Matt Kraai
Closes: #567250
  * Commented-out Vcs-* headers
Closes: #661752
  * Standards Version is 3.9.4

 -- Anibal Monsalve Salazar ani...@debian.org  Thu, 06 Jun 2013 12:27:15 +1000

Osamu


-- 
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/20140105083749.GA6693@goofy

Re: Reporting missing package during install

[Osamu Aoki]

On Sun, Dec 08, 2013 at 09:09:53PM -0500, Neal Murphy wrote:
 On Sunday, December 08, 2013 07:27:41 PM Andrei POPESCU wrote:
  On Du, 08 dec 13, 19:14:49, Neal Murphy wrote:
   For me, I usually set up 'sudo su'
  
  sudo has the '-s' and '-i' switches, why mix with 'su'?
  
  Kind regards,
  Andrei
 
 'sudo su' rolls off the fingers easier.

'sudo sh' is as easy on finger (no shift) and do not feel as bad.  

Osamu


-- 
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/20131212131450.GB6615@goofy

Re: Installing TrueType Fonts (TTF)

[Osamu Aoki]

Hi,

On Tue, Dec 10, 2013 at 07:53:20PM +0600, Muntasim-Ul-Haque wrote:
 Hi,
 I was a GNOME user and recently I switched to XFCE in Debian to see
 how things works here after reading that XFCE would be the default
 DE in the upcoming Debian Jessie. BTW, TrueType Fonts (TTF) would
 open in GNOME with a dialog box, and it was just one-click away. But
 here in XFCE, I see nothing that can open the TTFs. That's why I
 cannot install them easily here. So, how can I do that?
 With thanks,
 Muntasim-Ul-Haque

#1: install from package nmanagers:
  package name starting with: fonts- (all Debian supplied fonts)

#2: install non-Debian fonts.
  Put font files in ~/.fonts/ directory.

Osamu


-- 
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/20131212133922.GC6615@goofy

Re: Shutdown computer after a specific command has been executed

[Osamu Aoki]

Hi,

On Mon, Dec 09, 2013 at 02:38:45PM +0100, Ralf Mardorf wrote:

 One way would be to use a script that runs e.g. apt-get and then the
 shutdown command.

 #!/bin/sh
 apt-get update
 apt-get upgrade
 shutdown -h now # or poweroff or halt

   If you want it shut down regardless of the outcome of apt, then this
   should do it:
  
   sudo apt-get upgrade; sudo shutdown -h now
   
   Wrong, if the upgrade should take to long, then you need to type the
   password after the upgrade. Better run
   
   $ sudo -i
   # apt-get update ; apt-get upgrade ; shutdown -h now

I think Ralf is right on everything he mentioned.

But I want one line solution :-)

 sudo sh -c apt-get update  apt-get upgrade; shutdown -h now

Osamu


-- 
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/20131212134904.GD6615@goofy

Re: Toshiba Satellite U940

[Osamu Aoki]

On Wed, Nov 27, 2013 at 10:19:32AM +0100, ha wrote:
 Anybody tried installing Debian on Toshiba Satellite U940?
 Any know issues?
 I cannot find much information on the net.
 I can see that some Toshiba laptops are not very Deban-frendly
 (https://wiki.debian.org/InstallingDebianOn/Toshiba), so I wonder...
 
 Thanks to anybody who cares.

I do not use Toshiba Satellite U940 but Toshiba Satellite T571.

System installed without problem.

So did my old Toshiba (Sony, Dell, Panasonic, ...).  I usually have
mature systems.  I did have problem with Sony but the problem was solved
with newer kernel etc., on later release Debian.

It is not too much about vender but about CPU version and peripheral
chip version.  Since you looks like having very new CPU with the sexy
slim note PC, you may need newer kernel which is available only in
testing/unstable.

I guess booting with installation media into rescue mode should give you
idea for compatibility.

Osamu


-- 
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/20131207161747.GC26370@goofy

Re: Update of kde-window-manager wants to remove kde-standard

[Osamu Aoki]

Hi,


On Sun, Nov 10, 2013 at 03:04:02PM +0100, Rainer Dorsch wrote:
 Hi,
 
 I think that will get eventually resolved, but I am just curious how I can 
 find 
 the root cause of this upgrade issue.

Hmmm.. root cause seems to be aptitude package dependency resolver.
This is so typical ...  I like aptitude but I do not like this
happening.
 

  plasma-dataengines-workspace : Depends: libkworkspace4abi2 (= 4:4.11.2-3) 
 but 
 4:4.11.3-1 is to be installed.
 
 there is certainly plasma-dataengines-workspace Version: 4:4.11.3-1, which 
 depends on libkworkspace4abi2 (= 4:4.11.3-1)
 
 so why is aptitude complaining about this?

There may be some packages which want 4:4.11.2-3 version.

This is the new updated text from my Debian Refrence:

  The package dependency resolver of the aptitude command tends to
  suggests mass package removals when packages in unstable is temporary
  inconsistent. This situation is a bit frightening. Usually, apt-get
  dist-upgrade should resolve this situation. This situation seems to be
  caused mostly by the version skew among packages depended or recommended
  by a meta-package such as gnome-core.

http://www.debian.org/doc/manuals/debian-reference/ch02.en.html#_literal_apt_get_literal_literal_apt_cache_literal_vs_literal_aptitude_literal

Good luck,

Osamu


-- 
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/20131110151157.GA23598@goofy.localdomain

Re: Jessie Minimum Kernel Requirement

[Osamu Aoki]

On Wed, Oct 16, 2013 at 08:21:35AM -0400, Tom H wrote:
 On Wed, Oct 16, 2013 at 8:00 AM, Darac Marjal mailingl...@darac.org.uk 
 wrote:
  On Wed, Oct 16, 2013 at 03:54:06PM +0400, recovery...@gmail.com wrote:
  On Wed, 16 Oct 2013 11:46:40 +0200
  Florian Lindner mailingli...@xgm.de wrote:
 
  What is the minimum kernel version for the upcoming Jessie? Can I rely
  on that this does no change after the freeze? (IIRC at Nov 5th)
 
  Considering one should be always able to do apt-get dist-upgrade from
  wheezy to jessie (and that means using wheezy's kernel with jessie's
  userland, at least temporary), it should be safe to assume that minimum
  kernel requirement for jessie's userland is 2.6.32.
 
  This isn't really a safe assumption. There have been transitions in the
  past (such as udev) where the dist-upgrade should be performed as:
* Update sources.list
* Install new kernel and new udev
* Reboot
* Proceed with dist-upgrade
 
  After the reboot, you're then running on the previous release's userland
  with the current release's kernel. That situation SHOULD be more stable
  than the other way around as kernels /rarely/ remove functionality. But
  if you run new userland on an old kernel, and it tries to call
  functionality that's not there, you can run into trouble.
 
  Since wheezy's kernel version is fixed for its lifetime, it's highly
  unlikely that such requirement will change in the future.
 
 Furthermore, Wheezy's kernel is 3.2 not 2.6.32.

Furthermore, current unstable only has 3.10 (Dropped 3,2).

Soon, testing/jessie will only have 3.10.


-- 
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/20131016140128.GD7716@goofy.localdomain

Re: apt-get vs. aptitude

[Osamu Aoki]

Hi,

There used to be a very similar discussion on the Debian Devel mailing
list...

On Sun, Oct 13, 2013 at 11:46:41AM +0400, Dmitrii Kashin wrote:
 Tom H tomh0...@gmail.com writes:
 
  Oh I have fond memories of aptitude breaking my system. Once it
  suggested me to remove most of my system, including apt, I thought it
  was going to upgrade it so I confirmed it. I had to reinstall apt from
  the debian packages website.
 
  In this new installation I gave it another try but when it started
  suggesting very weird plans (like remove all gnome packages) I happily
  went back to apt and never looked back.
...
  If aptitude's such a destructive package, why is it still in the 
  repositories?
 
 I think that aptitude works quite well for the easiest cases. And it is
 the only instrument I know which allow to see dependency chains. It was
 dselect some time ago which could do it too as I know, but now it seems
 to be dead. BTW, it provides with good capabilities for searching
 through packages.

Also aptitude is a nice browser to see available package interactively.

dpkg (1.17.1) was uploaded on 28 Jul 2013.  So calling dselect dead is
not correct.  (I agree the days for the dselect popularity is gone but it
is not dead.)

 There're quite good beginnings in this project as you can see. So it's
 popular and is in the repository.

Aptitude is still a good interactive tool.

But neither the Release Notes or Debian Reference mention aptutide as
the primary tool for the system upgrade any more.  There is a reason
behind this change.

Also, in 2.2.1. apt-get / apt-cache vs. aptitude(*) of Debian Reference,
I have note to address this issue as:

| Although the aptitude command comes with rich features such as its
| enhanced package resolver, this complexity has caused (or may still
| causes) some regressions such as Bug #411123, Bug #514930, and Bug
| #570377. In case of doubt, please use the apt-get and apt-cache commands
| over the aptitude command.

(*) 
http://www.debian.org/doc/manuals/debian-reference/ch02.en.html#_literal_apt_get_literal_literal_apt_cache_literal_vs_literal_aptitude_literal

Recent Release Notes only uses apt-get and apt-cache in it.

Osamu


-- 
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/20131013221812.GA5454@goofy.localdomain

Re: Hibernate/resume fails after adding SSD to debian 7

[Osamu Aoki]

Hi,

On Sun, Sep 29, 2013 at 12:31:01AM -0400, Len Berman wrote:
 I moved my SSD from SATA 1 - SATA 5 and reran update-initramfs.  The
 resume now works.  Don't understand why this should be the case since all
 disks are referenced by their UUIDs.

All tools might not support UUID on your system.  mount may be
isupported but resume scripts, at least, was not supporting UUID before.

Oh, one more thing.

I had problem with my SSD this week.  System root file system became
read-only and crashed badly  I turned out to be the firmware version
issue of Crucial/Micron RealSSD m4/C400/P400 causing problem after 5184
hours of usage. So, you may also wish to check:

$ sudo smartctl -a /dev/sda

or similar.  My case prints:

This drive may hang after 5184 hours of power-on time:\n
http://www.tomshardware.com/news/Crucial-m4-Firmware-BSOD,14544.html\n;
See the following web pages for firmware updates:\n
http://www.crucial.com/support/firmware.aspx\n;
http://www.micron.com/products/solid-state-storage/client-ssd#software;,

I see many other drives has firmware bug issues.
/var/lib/smartmontools/drivedb/drivedb.h

Firmware update was iso image so I could do it without having any
windows for my drive. 

Osamu


-- 
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/20130929143008.GD5889@goofy.localdomain

Re: Should I build a nmu for stable or a backport for wheezy-backports?

[Osamu Aoki]

Hi,

On Tue, Aug 06, 2013 at 08:17:03PM +0100, Rui Miguel P. Bernardo wrote:
 Hello list,
 
 let's say there is a bug in a stable package and that bug breaks the
 program functionality. Later the fix was uploaded to unstable/testing
 but never got in time for stable. For reference I'm talking about
 http://bugs.debian.org/679657.
 
 I tried 2 ways to solve this:
 
 a) I've downloaded the stable version of the package, applied the
 patch that fixed the problem and built a wheezy-backports package;

What you described  is the way we make stable updates.  I have done this
kind of things.
  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=714759

As you can see, it takes a bit more than usual efforts.

Is this something all stable user needs to be exposed?

 b) I've downloaded the maintainers git repository (unstable), revert
 some commits and build a wheezy-backports;

Usually, backport is simply recompiled version of testing package on
stable platform (with only dpkg/debheler updated to backport).
 
 Backports exists for recent packages from unstable/testing that were
 adapted and rebuilt for stable. What I did in a) is not that: I have
 rebuilt a stable package and applied a patch.

If you are doing it only for you, do it anyway.

 If a) is not a backport is it a nmu then? Should I build a) as a
 stable nmu and try to search for a sponsor to upload it to stable? Can
 this be done?

This is not A or B question.  2 different criteria.

 Or, to have a valid backport of the package, I MUST make b), which is
 to backport the testing/unstable package?

testing.  Please read backport docs.

 What I'd like is to have the stable version of the package fixed in
 debian stable, where it is not working, not to have an upgraded
 package from backports.

Please read Debian policy on stable update.  You also need to cordinate
with the maintainer.  You asking here indicate you have lots to learn.

 I hope this email is not to confusing as my doubts :) I'd like to have
 my doubts cleared because there is at least one more package
 (avelsieve) I'd like to upload, via nmu or backports, depending on the
 answers to my doubts.

Good luck.

Osamu


-- 
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/20130807122306.GA6078@goofy.localdomain

Re: Starting Google Chrome - resend

[Osamu Aoki]

Hi,

On Fri, Jul 05, 2013 at 12:14:20PM -0400, Ethan Rosenberg, PhD wrote:
 I'm not sure that this ever got to the list, so I am resending.
 
 Dear list -
 
 I apologize for having to revisit this, but I can't get it to work,
 which I cannot understand.  I followed the instructions I used
 previously, and no luck.
 
 Download from Google.
 
 cd /home/ethan/Downloads
 
 dpkg -i google-chrome-stable_current_i386.deb
 
 works w/ errors.

Without listing what errors you experienced, it is impossible to guess.

 apt-get -f install
 
 No luck.  No executable found.

Please post exact message.

 Created the directory /opt.
 
 Tried the above again.

Did error message requested you to do this?

 No luck.

Basically, there should be no error etc.  Please read the manual and
install required packages in advance.

Osamu


-- 
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/20130706131401.GA29062@goofy.localdomain

Re: Debian/Linux tutorials.

[Osamu Aoki]

Hi,

On Thu, Jun 06, 2013 at 08:11:45AM -0500, Richard Owlett wrote:
 Andrei POPESCU wrote:
 On Jo, 06 iun 13, 14:11:21, atar wrote:
 Hi there!!
 
 I wanted to know please where can I enrich my knowledge about Linux
 at general and especially about Debian regarding variety of topics
 such as the system boot process, what are tty's? and where are
 located the scripts that the system runs during the boot and
 shutdown process, how to manage groups policies and how manage the
 whole system at all. where can I find some useful
 resources/documentation/references that speak in clear fashion,
 because the man pages often uses some concepts which aren't familiar
 to the beginner user, so I look for some resources that are friendly
 with the beginner user.
 
 http://www.debian.org/doc/
 
 Speaking as a beginner myself, those are written by and for experts.

I do not know which one are you talking ... there is too much and many
are highly technical documents.

I hamble work is:
  http://www.debian.org/doc/manuals/debian-reference/

It is technical end-user oriented

I think you maybe looking for 
 The Debian Administrator's Handbook
 Author: Raphaël Hertzog, Roland Mas
 Publisher: Freexian
 URL: http://debian-handbook.info
Which is hidden deep in the link for Debian book.

Also for non-technical / user oriented guide, Ubuntu may give you such
introduction.  Although we try to make it easy, driving Debian requires
you to be technical to some extent 

 I found out, during an engineering co-op assignment 40 years ago,
 just how difficult it is to write technical documentation aimed at
 someone with a different background. And it was only a ten page
 production line test procedure.

We do not have that much resource ... to be honest.  You are welcomed to
contribute :-)

Osamu


--
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/20130617154216.GA11149@goofy.localdomain

Re: Installing Debian on an SSD

[Osamu Aoki]

On Sun, Jan 13, 2013 at 05:05:12AM +0800, Hormatzhan Yiltiz wrote:
 On Wed, Jan 9, 2013 at 3:48 AM, Bob Proulx b...@proulx.com wrote:
 
http://wiki.debian.org/SSDoptimization
 
 
 This page, as you pointed out, is more than a guideline. Can someone please
 polish this up a little bit? That would be a wonderful and appreciative.

To my surprise, ... you are right.  Fixed.

If you can help, please make sensible editting.

Osamu


-- 
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/20130121163641.GB8926@goofy.localdomain

Re: admin password?

[Osamu Aoki]

Hi,

On Sun, Dec 30, 2012 at 06:50:20AM +0200, Dionyssis Goulimis wrote:
 Hi,
 
 I downloaded the last edition of Debian yesterday but I can't remember
 my Admin password, what is the best way to recover it?

If you did not lock system with grub password, you can always login as
root without password.

Google with grub init=/bin/bash mount you get many howtos describing: 

Boot with 

 init=/bin/bash

or 

 init=/bin/sh 

for GRUB.  Then at root prompt

 mount -o remount,rw /

to reset with passwd.

Cheers,

Osamu


-- 
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/20121231040341.GA24153@goofy.localdomain

Re: Help netbooting a diskless, [headless] system [solved/tips]

[Osamu Aoki]

Hi,

On Sat, Dec 29, 2012 at 12:33:15PM -0800, Ross Boylan wrote:
 I got things working and made some  discoveries along the way.  Maybe I
 should put this material on a wiki; any suggestions where?

Create account at wiki.debian.org and think about putting it there under
appropriate location.  You are doing good job ... I haven't read it but
it is non-trivial to work under consumer grade network router.

When you put it, please describe network configuration with schematic
picture etc.  

Thanks,

Osamu


-- 
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/20121230004651.GA6260@goofy.localdomain

Re: something about cups-pdf

[Osamu Aoki]

Hi,

On Fri, Dec 28, 2012 at 12:54:31AM +0800, lina wrote:
 :/var/spool# ls -lrt cups-pdf/
 total 8
 drwxrwxrwt 2 root lpadmin 4096 Dec 27  2011 ANONYMOUS
 drwxr-x--x 2 root lpadmin 4096 May 12  2012 SPOOL

Same here:

root@*:/var/spool/cups-pdf# ls -l
total 8
drwxrwxrwt 2 root lpadmin 4096 Aug 22 12:59 ANONYMOUS
drwxr-x--x 2 root lpadmin 4096 Dec 21 22:36 SPOOL

 Is it normal? What does the  ANONYMOUS for? drwxrwxrwt ?

Normal.

According to /usr/share/doc/cups-pdf/README.Debian (The place you
should be checking first.), I see:

| This Debian package of CUPS-PDF is configured as follow:
| 
|  * Log files are created by group lpadmin to match CUPS permissions.
|  * Anonymous Samba output to /var/spool/cups-pdf/ANONYMOUS/ directory.
|  * Standard user output to ${HOME}/PDF/ directory.

drwxrwxrwt is explained in man ls.  Basically, it is a directory
anyone can read/modify/write/...

You see more detailed explained by me here:
 
http://www.debian.org/doc/manuals/debian-reference/ch01.en.html#_filesystem_permissions

Osamu


-- 
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/20121228121817.GA12575@goofy.localdomain

Re: Tools to retrieve images from dead hard drive and/or deleted partitions

[Osamu Aoki]

Hi,

On Thu, Dec 27, 2012 at 10:17:56PM +0100, berenger.mo...@neutralite.org wrote:
 Ok, first, sorry to ask two questions at a time: dead drives and
 deleted partitions are different things, I know that.
 ...
 (2), I've more hopes, IF the user stopped the destructive process
 before the disk was fully erased, but I'll need the good tools.

Look for Debian package testdisk

  http://packages.debian.org/sid/testdisk

When my PC HDD became dead completely, I used this testdisk to recover
photos on SD card in camera after erase.  Of corse, any files
overwritten by newer images were lost.  But, it was a very easy job and
I got most of the recent photos.  It guides you through nicely. (I worked
on disk image first...)

After everything went fine, I tried to run unrecommended mode --- work
directly on erased SD card.  Wow, it worked for me.

PhotoRec mentioned by  another comes as a part of testdisk suites of
tools.

I still recommends you to make disk image first.  If buying big HDD for
disk image is not your option for some reason, you may try this
dangerous direct recovery trick using some live CD and testdisk package.
It is better than doing nothing...

In any case, you should practive recovery on small test case first ...

Osamu


-- 
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/20121228123553.GB12575@goofy.localdomain

Re: Tools to retrieve images from dead hard drive and/or deleted partitions

[Osamu Aoki]

Hi,

On Fri, Dec 28, 2012 at 01:19:21PM +0100, Ralf Mardorf wrote:
 On Fri, 2012-12-28 at 12:50 +0100, berenger.mo...@neutralite.org wrote:
   I have heard good things of PhotoRec:
  
   http://en.wikipedia.org/wiki/PhotoRec
   http://www.cgsecurity.org/wiki/PhotoRec
  
   But I have no personal experience of it.  The advice is always to 
   copy the
   drive that needs rescuing to another drive, and work on the copy, not 
   the
   orginal.  Again, I have not tried it.
  
   Lisi
  
  Thanks for the hint, I'll try it this afternoon on (2)
 
 Mount the drive read only, than there is no need to backup. There are
 all kinds of commands to recover data, usually they don't recover the
 data on the corrupted drive, but save the files to another place, so
 it's no problem to mount a drive read only.

Yah, if you can mount, then it is good idea to mount as read-only first
to recover and secure data by copying them to another disk. 

But that may not get you as much data as we wish.  testdisk/photorec
tools can work on corrupted disk without using normal mounting.  You
certainly need place to write recovered data.  Making disk image and
working on it makes things easy. 

Osamu


-- 
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/20121228124520.GA13645@goofy.localdomain

Re: Tools to retrieve images from dead hard drive and/or deleted partitions

[Osamu Aoki]

Hi,

On Fri, Dec 28, 2012 at 01:23:47PM +0100, Ralf Mardorf wrote:
...
 PS: Even if you make a backup first, by all means mount the drive or
 partition read only, before you access it.

Hmmm... Maybe I was too short on explanation.

The best practice is NOT TO MOUNT AT ALL.

If your desktop system auto mount disk, just run like:

$ sudo mount
... check output to see if device is automounted:
$ sudo umount /dev/sdc1 (or what ever disk partition mounted)

(It is better not to face this situation but this tends to happen
unintentionally.)

Use dd to make disk image.  For that you do not need to mount
drive.  From root with /dev/sdc being bad drive and /dev/sdd being good
bigger drive, do this first.

root # dd if=/dev/sdc of=/dev/sdd

This is the best way to get a safe copy to work with the testdisk
command on the copy.  This way, recovered data is safe too.

See more on this
 
http://www.debian.org/doc/manuals/debian-reference/ch10.en.html#_making_the_disk_image_file


Osamu

PS: I am not recommending to mount nor work on the original data disk.



-- 
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/20121228182608.GB11977@goofy.localdomain

Re: Implementing a personalized repository

[Osamu Aoki]

On Fri, Dec 28, 2012 at 10:50:27AM -0600, Richard Owlett wrote:
 I routinely work without a viable network connection of any kind. My
 response was to purchase a eight DVD set of squeeze. Using the DVD
 directory structure as an example I have copied the contents of
 /pool to a 64GB flash drive. I have also copied the directory
 structure of /dists to the flash drive.
 
 I've lost links to:
a. creating the packages and packages.gz files.

http://www.debian.org/doc/manuals/debian-reference/ch02.en.html#_small_public_package_archive

b. how to point apt/synaptic/... to this repository.

  file:///path/to/repo instead of http://ftp.us.debian

Osamu
PS: I never done this for large archive such as what you indicated.  


-- 
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/20121228183037.GC11977@goofy.localdomain

Re: Implementing a personalized repository

[Osamu Aoki]

On Fri, Dec 28, 2012 at 03:44:18PM -0600, Richard Owlett wrote:
 Osamu Aoki wrote:
 On Fri, Dec 28, 2012 at 10:50:27AM -0600, Richard Owlett wrote:
 I routinely work without a viable network connection of any kind. My
 response was to purchase a eight DVD set of squeeze. Using the DVD
 directory structure as an example I have copied the contents of
 /pool to a 64GB flash drive. I have also copied the directory
 structure of /dists to the flash drive.
 
 I've lost links to:
 a. creating the packages and packages.gz files.
 
 http://www.debian.org/doc/manuals/debian-reference/ch02.en.html#_small_public_package_archive
 
 I'm not sure that section answers the exact question I'm attempting
 to ask. *BUT* I presume the overall document is aimed at typical
 average sane users. In past three score and ten those adjectives
 have ever been used to describe me ;/

The postupload hook script initiated by dupload(1) creates updated
archive files for each upload.

So it is a bit convoluted method and it is not a direct shell command.

 HOWEVER, after browsing the table of contents at
 http://www.debian.org/doc/manuals/debian-reference/index.en.html I
 suspect the posting of a half-dozen questions has been
 short-circuited. Thank you.
 
 
 b. how to point apt/synaptic/... to this repository.
 
file:///path/to/repo instead of http://ftp.us.debian
 
 Osamu
 PS: I never done this for large archive such as what you indicated.
 
 Chuckle I read a blog somewhere which commented on small
 archives vs fully standard compliant archives. Although I'll hold it
 in my hand, I'm aiming at the later.

FYI: I updated wiki page:

 http://wiki.debian.org/HowToSetupADebianRepository

Osamu


-- 
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/20121229074555.GA6098@goofy.localdomain

Re: ssh fails - SOLVED (was: Re: git pull fails with OpenSSL version mismatch error)

[Osamu Aoki]

On Mon, Dec 24, 2012 at 10:48:00AM +0200, Andrei POPESCU wrote:
 On Du, 23 dec 12, 20:15:19, Joel Roth wrote:
  
  Ah, I didn't even think to try an apt-get upgrade. 
  I usually just upgrade apps individually as I need to...
  an attitude based on (possibly) outdated fears of
  getting stuck in between upgrades of C libraries
  or other large-scale brokenness.
 
 IMVHO your method is less safer than 'apt-get upgrade' or
 'aptitude safe-upgrade'.

I can not agree with you more.  But there seem to exist some myth on
this feeling of safe upgrade using partial upgrade.

With symbols file in the package, we as Debian seem to care library
compatibility issues when transitioning package from unstable to
testing.  But this manual method does not check such thing.

If anyone have suggestion to improve debian-reference to prevent people
to take such strategy, let me know.  Also, concrete case example of why
such method is less safer than 'apt-get upgrade' or 'aptitude
safe-upgrade', let us know.

Osamu


-- 
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/20121224144647.GA8023@goofy.localdomain

Re: Proofreading of Debian documentation - was [Re: Preseeding - keyboard-configuration issue]

[Osamu Aoki]

On Sun, Dec 16, 2012 at 06:53:26AM -0600, Richard Owlett wrote:
 Andrei POPESCU wrote:
 But the underlying problem is lack of manpower. How do you[1] expect to
 
 solve it without getting involved?
 
 [1] and I mean you personally, the Debian Project is already trying to
 attract more contributors. See for example the General Resolution to
 welcome non-packaging contributors
 http://www.debian.org/vote/2010/vote_002
...
 In response to Andrei and Chris:
... 
 I'm making a list of problems using Debian GNU/Linux Installation
 Guide. If/when I find the problem is the guide, not myself, I will
 submit specific bug reports. But in itself that will not prevent
 other symptoms of the same underlying problem.

Also, please consider reading 
 debian-b...@lists.debian.org (working on boot itself)
 debian-...@lists.debian.org  (release note etc.)
mailing list.  We do not have SIG but have mailing list.
 
 The response I had hoped for was on the lines of  there's a SIG for
 documentation which you can contact at 
 
 If  one does not exist I've some ideas, but I hate reinventing the
 wheel if it already exists.

Then please do not make separate list.  Use usertag to track it by
filing solution with patch :-)

Osamu


-- 
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/20121219121527.GB5762@goofy.localdomain

Re: partition table not sticky

[Osamu Aoki]

Hi,

On Fri, Dec 07, 2012 at 01:06:17PM +0530, J. B wrote:
...
 I have repeatedly do the same but no luck. no luck with cfdisk. Changes are 
 not stored !!!

This part I do not know.

 only gparted is successful but it makes my 320 GB HDD to 280 GB. 40GB lost !!!

This is most likely because of gparted is creating GUID partition table
(GPT) instaed of good old MBR.
 http://en.wikipedia.org/wiki/GUID_Partition_Table
 http://en.wikipedia.org/wiki/Master_boot_record
 
 What is the problem with fdisk and cfdisk ???

I bet you used gparted first to create GPT and now trying to use these.
This is just a wild guess.  (Some fdisk can not handle GPT as I
remember)

If you wish to make HDD use MBR instead of GPT, you can reset it by
something like (assuming your disk is /dev/hdx):
$ sudo dd if=/dev/zero of=/dev/hdx bs=4k
$ sudo fdisk /dev/hdx

This should work.

You can also set partition table as msdos via parted.
$ sudo parted /dev/hdx mklabel msdos

See:
  
http://www.debian.org/doc/manuals/debian-reference/ch09.en.html#_disk_partition_configuration
  
Osamu


-- 
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/20121207122332.GA6027@goofy.localdomain

Re: Locale issue

[Osamu Aoki]

On Fri, Dec 07, 2012 at 01:17:20AM -0800, ML mail wrote:
 Hello,
 
 Since now already a few weeks I noticed that my locale settings on various 
 Debian 6.0 servers got by some mysterious way broken... For example any 
 tool/command using PERL will issue the following warning:
 
 perl: warning: Setting locale failed.
 perl: warning: Please check that your locale settings:
     LANGUAGE = (unset),
     LC_ALL = (unset),
     LC_CTYPE = UTF-8,
     LANG = en_US.UTF-8
     are supported and installed on your system.
 perl: warning: Falling back to the standard locale (C).

Debian 6.0 = squeeze

This is typical if you use  perl under LANG = en_US.UTF-8.
This was fixed for wheezy ay perl

| perl (5.12.2-2) experimental; urgency=low
| ...
| + Squelch useless locale warnings during package maintainer scripts.
|   (Closes: #508764)
| + Improve LC_NUMERIC documentation. (Closes: #379329)
| + Fix sprintf not to ignore LC_NUMERIC with constants. (Closes: #601549)
| ...
| 
|  -- Niko Tyni nt...@debian.org  Tue, 02 Nov 2010 11:34:32 +0200


--
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/20121207123405.GB6027@goofy.localdomain

Re: Dual boot with OS X and Wheezy.

[Osamu Aoki]

On Sat, Dec 01, 2012 at 12:49:20PM -0500, Ed Jabbour wrote:
 Trying to dual boot OSX and Wheezy on a late 2008 Macbook running 
 Mountain Lion.  Following steps found on the web, I first created an 

Quite unclear here what exactly are you doing ... normal bootcamp or
EFI ...

 empty partition using OSX's Disk Utility - 80 GB for OSX and 80GB free 
 for Wheezy.  However, when I reached the partitioning step in the 
 Wheezy install, it id'd the free space as 113GB, which is the 
 remaining space on the entire disk.  In other words, it seems that 
 Wheezy is not recognizing the OSX partitioning structure.  I'm afraid 
 that if I simply partition that 113GB space, I will have messed up the 
 partition structure for OSX.

This is odd.  You need to print exact data shown by partition utility.
Oh, did you use parted? or something else...

 I hope I'm making sense here.  Any pointers, steps, hints, greatly 
 appreciated.

http://wiki.debian.org/InstallingDebianOn/Apple
http://wiki.debian.org/MacBookPro
http://wiki.debian.org/MacBook
 


-- 
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/20121203123832.GA15520@goofy.localdomain

Re: Kobo desktop (native)

[Osamu Aoki]

On Thu, Nov 29, 2012 at 12:02:18PM -0300, Beco wrote:
 (1) Native
 # dpkg -i kobo-desktop.deb
 (Reading database ... 308087 files and directories currently installed.)
 Preparing to replace kobo-desktop 2.1.3-1 (using kobo-desktop.deb) ...
 Unpacking replacement kobo-desktop ...
 dpkg: dependency problems prevent configuration of kobo-desktop:
  kobo-desktop depends on libusb-0.1-4.
  kobo-desktop depends on libpng3.
  kobo-desktop depends on libjpeg62.
  kobo-desktop depends on libzip1.
  kobo-desktop depends on libssl0.9.8.
 
 dpkg: error processing kobo-desktop (--install):
  dependency problems - leaving unconfigured
 Errors were encountered while processing:
  kobo-desktop

Hmmm why bother with non-official :-)

Just install calibre.

I use it just under Debian only (No windows.)

Here are some info source (Some are in Japanese) I use.

My advise is backup ! and reset it often so it does not accumulate junks
to slow down it further.  It is slow.

(FYI: I am enjoying Soseki Natsume and G.H. Wells now !
Late 19th and early 20th centuty is quite interesting.)

usbnet:
http://d.hatena.ne.jp/mjt/20120721/p1

telnet:
http://www.ekesete.net/log/?p=5197
English summary
http://www.chauveau-central.net/pub/KoboTouch/

PRIMARY INFO English
http://ccoffing.net/tech/embedded/kobo/
http://wiki.mobileread.com/wiki/Kobo_Touch_Hacking

EXT SD card
http://www.richud.com/wiki/Kobo_eReader_touch


J Summary!
http://www49.atwiki.jp/kobotouch/ Japanese unofficial wiki
http://zapanet.info/kobo/


Do not have this but here are useful URLs
http://itpro.nikkeibp.co.jp/article/Watcher/20120806/414542/
http://en.wikipedia.org/wiki/Kobo_Touch




http://d.hatena.ne.jp/mjt/20120721/p1
http://lin.blog.so-net.ne.jp/2012-07-26
http://d.hatena.ne.jp/Kenji_s/20120826/kobo_setup
http://d.hatena.ne.jp/Kenji_s/20120904/kobo_sshd
http://d.hatena.ne.jp/Kenji_s/20120905/kobo_usbnet

http://tkamada.blogspot.jp/2012/08/kobo-touchgoogle-analytics.html
http://blog.antenna.co.jp/CSSPage/2012/08/koboepub3.html




-- 
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/20121130140658.GD5965@goofy.localdomain

Re: vim -- adduser vs useradd/usermod

[Osamu Aoki]

Hi,

On Fri, Nov 23, 2012 at 07:56:24AM -0600, Nate Bargmann wrote:
 * On 2012 23 Nov 06:14 -0600, Andrei POPESCU wrote:
  On Vi, 16 nov 12, 16:33:17, Nate Bargmann wrote:
   * On 2012 16 Nov 14:02 -0600, james gray wrote:
i am using vim to add the one and only name in a usr account to the 
groups
file for printing.
   
   I would use, as super user:
   
   # usermod -Ga lp username
  
  On Debian the 'adduser' tool should be used.

Maybe this is a bit strong statement.  I would just say the use of
adduser on Debian system makes life easy on Debian.
 
 Unfortunately, the name alone doesn't lend itself to knowing that it
 will modify a user account although the man page says:
 
Add an existing user to an existing group
   If called with two non-option arguments, adduser will add  an
 existing user to an existing group.
 
 The page is silent as to whether existing group membership is preserved
 as with the command example I offered.  To know it would be required for
 each adminstrator to test this command due to this documentation bug.

I think there is a reason why adduser is not called changeuser (or
updateuser) (just joking though). I am very certain this only ADD
things.  If it REMOVES something, this is a serious documentation bug
not to mention it.  But it just works as expected...  I think.

 In this case, 'usermod' provides the adminstrator *exactly* the options

Although adduser for this is the popular utility to use here and the one I
probably use, usermod seems to be as legitimate utility and it comes as a part
of shadow-utils i.e. the passwd package.

Both should work.  Otherwise, file a bug report :-)

 needed to modify a user account.  A single or multiple groups may be
 easily added, or the group list may be passed in such a way as to easily
 remove an account from several groups at once while preserving or adding
 membership in those passed.

If adduser should do such things, there should be a lot of unhappy people
filing bug report ...

 I don't see that capability in 'adduser'
 although there is the complementary 'deluser' which removes a sepcific
 user from a specific group.  To remove an account from multiple groups
 would require multiple invocations.
 
 Power or simplicity?  The choice is yours.

Power -- certainly.  If you read /usr/share/doc/adduser/examples/README,
there are good power tools described.

For not so much involved like me, adduser is a friendlier interactive
frontend to useradd. At least, this is my impression and that is why I
used it for Debian Reference.

I google and find interesting old post here.

 Subject: Re: adduser, useradd: are there any differences
 From: Peter J. Holzer hjp-deb...@hjp.at
 Date: Sun, 12 Sep 2004 20:55:39 +0200
 On 2004-09-12 18:12:02 +0200, Roland Wegmann wrote:
...
 useradd and groupadd are the posix or SUS-standardized tools to create
 users and groups. You can expect them to work (almost) identically on all 
 Linux
 distributions as well as on HP-UX, Solaris, AIX, ...
 
 adduser and groupadd are distribution specific wrappers which make it
 simpler to create users which conform to some (distribution-specific or
 local) policy. The options vary wildly between different distributions
 and other UNIXes may not have these commands at all.

(As I checked web, FEDORA/CENTOS/GENTOO, adduser seems to be the symlink
to useradd.)

Osamu


-- 
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/20121124131224.GB22175@goofy.localdomain

Re: Thunar, USB-sticks and big files

[Osamu Aoki]

On Sun, Nov 18, 2012 at 07:43:54PM +0100, Andreas Rönnquist wrote:
  Try 'sync' after the write is supposedly complete; see what happens
  when the system actually tries to write to the device. Open a shell
  and 'tail -f /var/log/messages' to see if anything is griping about
  the device during writes.
  
  Hmmm. You *are* waiting for the cached data to be flushed to the
  drive before unplugging it, right?
  
 
 hmm, I now I have tried several different USB-sticks, and I still have
 the problem (It perhaps isn't quite as serious as I believed though)...
 
 If having copied a big file, unmounting through Thunar pretty much
 always results in the following error message:
 
  Failed unmounting 'Devicename'
  Did not receive a reply. Possible causes include: the remote
  application did not send a reply, the message bus security policy
  blocked the reply, the reply timeout expired, or the network
  connection was broken.
 
 - after a short while (way before the actual writing is done) - But, if
 I keep waiting for some time _after_ this, the data is still written to
 the device as it should. However, it doesn't give any clue whatsoever
 to when it actually is finished writing the files to the device, so it
 is easy to (by mistake) physically take the USB-stick before the data is
 actually written.
 
 However, I do have one USB-stick, that actually lights up a LED during
 copying, making it pretty easy to spot when it is safe to remove it and
 not, but I kind of wish that that sort of feature wouldn't be necessary
 to safely use an USB-stick...
 
 If I don't copy any files, or just copy small files, I get that
 standard It's now safe to remove the device when unmounting - This
 is what I expect in the scenario with bigger files described above
 too after copying is done, but no...

I do not know how slow is your USB stick but running sync in terminal
should write disk cache to USB stick for sure.

Osamu


--
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/20121121115024.GB11848@goofy.localdomain

Re: Can;t fix doc-base installation problem

[Osamu Aoki]

On Tue, Nov 20, 2012 at 10:49:03AM -0800, Gary Roach wrote:
 Hi everyone:
 
 I've had to restore one of my systems from cold iron. Aptitude seems

I guerss you are using unstable.

 to be hanging up when trying to install / purge / update doc-base.
 I've tried Aptitude, dpkg and apt-get including the -f switch with
 no  joy. Below is a typical example:
 
root@mycomputer:/home/gary# aptitude purge doc-base
The following packages will be REMOVED:
   doc-base{p}

Did you check our BTS?
 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=648936
This looks like one you hit.

0 packages upgraded, 0 newly installed, 1 to remove and 0 not upgraded.
Need to get 0 B of archives. After unpacking 504 kB will be freed.
Do you want to continue? [Y/n/?]
(Reading database ... 297009 files and directories currently installed.)
Removing doc-base ...


Now calling 
 /var/lib/dpkg/info/doc-base.prerm

cannot find install-docs on path

This is the problem.

If I were you, I will try following as work around.  But before doing
this, check all discussion on the above mentioned BTS.  There may be
better ways.

You should see in /var/lib/dpkg/info/doc-base.prerm:
case $1 in
remove)
remove_all_docs
;;
upgrade)
;;
failed-upgrade)
# Prepare to upgrade from version $2 of this package TO THIS VERSION.
# This is only used if the old version's prerm couldn't handle it,
# and returned non-zero.  (Fix old prerm bugs here.)
;;
esac

Edit this to be:

case $1 in
remove)
true
;;
upgrade)
;;
failed-upgrade)
# Prepare to upgrade from version $2 of this package TO THIS VERSION.
# This is only used if the old version's prerm couldn't handle it,
# and returned non-zero.  (Fix old prerm bugs here.)
;;
esac

and try removing this again.  This is just work around.  If there are
other issues, please work around issues.

I think by installing newer version of doc-base, this should fix this.
...


-- 
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/20121121143746.GA23161@goofy.localdomain

Re: Thunar, USB-sticks and big files

[Osamu Aoki]

Hi,

On Sun, Nov 18, 2012 at 11:58:29AM +0100, Andreas Rönnquist wrote:
 Hi guys!
 
 I have a problem copying big files to an USB-stick using thunar in
...

 I do this to copy ripped DVD's to USB memory for viewing in a TV with
...
 Nothing extremely serious, I can still copy the files using terminal as
 I said, but it would be nice to have it working using only Thunar too.

You see this list talking the same problem :-)
You are hitting 4GB limit of FAT.

Reformat USB stick with
 * ext2/3/4 if moving around Linux.
 * Darwin UFS, I think, for Mac
 * NTFS for Windows ...

Osamu



--
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/20121118122815.GA24927@goofy.localdomain

Re: Squeeze on a EEE PC 901, trackpad problems

[Osamu Aoki]

Hi,

On Fri, Nov 16, 2012 at 09:05:04PM -0600, zxcvbob wrote:
 I just installed Debian with the Xfce desktop on a EEE PC model 901.
 I've got things working pretty well, but I can't enable tap-to-click on
 the trackpad.  I go to Settings-Mouse in the Xfce menu, and there is no
 tab or checkbox for tapping.  I'm guessing I need to enable something in X?

For these questions, use search engine with
   eeepc site:wiki.debian.org :-)

This leads you to:

 http://wiki.debian.org/DebianEeePC
 http://wiki.debian.org/DebianEeePC/TipsAndTricks

 (I just barely know what I'm talking about) I've run Ubuntu Linux
 before, but that was preconfigured.  I have never tried to set
 everything up myself.

For basic debian operations:
 http://www.debian.org/doc/manuals/debian-reference/

Good luck.

Osamu


-- 
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/20121117103251.GA9030@goofy.localdomain

Re: jigdo + approx + netinstall -- need for a coherent system

[Osamu Aoki]

Hi,

On Fri, Nov 16, 2012 at 03:38:04PM +, Russell L. Harris wrote:
 Have I missed something?

maybe ... many iso images are there for people without good network
access.

... first think about what you really wish to do.

Let me assume your objective is to install the current testing system to
several machines on your LAN and keep them updated with minimal network
load, ...

 (1) jigdo-lite provides a foolproof and relatively easy method for
 the individual to keep on hand locally an up-to-date copy of one or
 more Debian releases, in the form of ISO images for CD or DVD.  In
 particular, jigdo-lite:
...

Unless you are testing all ISO images, you do not need to have them all.
All you need is just the 1st CD image (not nessasary the latest one.
netinst image is good enough.) if you have decent network access.

 (2) However, there is no simple and obvious approach to automation of
 a jigdo update of set of images such as the images of the weekly Debian
 testing (Wheezy) update.
...

You do not need to update it unless your image is broken badly with some
bug.

 (3) The netinst technique provides a simple and nearly-universal
 method for the installation of Debian.
...
Yep.

 (4) approx provides a simple method for the individual to install
 Debian in one or more machines in the local LAN, while eliminating the
 need to burn a stack of CDs or DVDs and shuffle the stack of disks
 during each installation.
...
Alternatively, you can run any http proxy server.  I used to use squid.
By setting it properly, your bandwidth is minimized.

Your first attempt to access repository for your jigdo image should be
cased and should be used later for installation if it has not been
changed and required. (This happens, I think, if you download large
CD/DVD image and decide to install via network.)

 (5) Moreover, approx provides the potential of a near-zero-bandwidth
 installation:
...
Yep.

 (6) With the advent of inexpensive flash USB sticks and solid-state
 drives, both CD and DVD today are obsolescent; indeed, it is common to
 see laptops which have no provision for an internal optical drive.

Creating USB image from CD image is documented somewhere like:
  http://wiki.debian.org/BootUsb
  http://www.debian.org/CD/faq/#write-usb

 (7) approx-import allows .deb package files to be added to
 the approx cache.

If you accessed repository via approx for whatever need, those should be
cached already.  You usually specify proxy server with environment
variable for this.  Read each manual how to do this for each
application.  The $http_proxy environment variable can be used to
override proxy server setting in the /etc/apt/apt.conf file for apt.

http://www.debian.org/doc/manuals/debian-reference/ch02.en.html#_proxy_server_for_apt

There is no need to manually add them after download.

 %%%
 
 In view of the foregoing considerations, it would be nice to see
 development of a procedure whereby jigdo-lite, approx, and
 netinst are transformed into a compatible and coherent system in
 which the logistics are simplified.  
 
 (1) Perhaps nothing more than a script or two would be necessary.

none
 
 (2) It should be possible with a single command to update the weekly
 testing set of ISO images.

yes but why do this.  

 (3) There perhaps should be four sets of .jigdo and .template
 files:
 
 - a set for CD images 
 - a set for DVD images
 - a set for a single netinst image 
 - a set for a single tutti (that is, all together) image

???  Why you need these all ???

 (4) The tutti image, loop-mounted, could be the source for
 all other images.

??? yes but why make it so complicated.  The debs are all available in apt
proxy.  

 (5) With the tutti approach, one could obtain the weekly update of
 Debian testing by jigdo download of only a single file.
 Moreover, download of the single tutti file could be
 accomplished automatically by a script invoked by cron.

???
 
 (6) It should be possible to import files directly into the approx
 cache, irrespective of source: 
???

If you are half-offline, think about apt-zip ... 

Osamu


-- 
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/20121117110359.GB9030@goofy.localdomain

Re: Password problem confirmed as repeatable - was [Re: Newbie password problem(s)]

[Osamu Aoki]

Hi,

On Sat, Nov 17, 2012 at 06:31:50AM -0600, Richard Owlett wrote:
...
 I've been a computer user since before Mr. Torvalds was born. I
 just never had reason to have contact with *nix. Even when working
 for DEC I was much more into analog than digital.

I see.  I guess you were a VMS or some mainframe user...

As for apt-zip:
 http://alioth.debian.org/projects/apt-zip
 http://packages.qa.debian.org/a/apt-zip.html
 http://bugs.debian.org/cgi-bin/pkgreport.cgi?src=apt-zip

It seems somewhat stalled.  You may wish to takeover project to update
it to be compatible with current apt repository.  The last update was
2008.

 By the way, we have some basics for use as:
http://www.debian.org/doc/manuals/debian-reference/
 
 Thanks for reminding me. I haven't used it as, being on a dial-up
 connection, online manuals are inconvenient. Is it available as a
 PDF or as a single HTML file?

For installation, you should start with Install Manual:
  http://www.debian.org/releases/testing/installmanual
  http://www.debian.org/releases/stable/installmanual
  PDFs are available. 
  If you use 64bit PC, you want amd64 version.
  It is good idea to read this if you really wish to tweak.

It is available as Debian Package.
  http://packages.qa.debian.org/d/debian-reference.html
  apt-get install debian-reference-en
  This is more about Unix shell focused.
  Once you install it, you can see it via browser.

(I know I used to have PDF.  Building PDF for non-English via XML-XeTeX
is still not so easy.  That will be my next project.  Maybe I should
enable just for English and French first which I know is easy.)

Another good source is THE DEBIAN ADMINISTRATOR'S HANDBOOK
  http://packages.qa.debian.org/d/debian-handbook.html
  apt-get install debian-handbook
  General guide.

 Is there a similar volume of preseeding, especially for those of us
 who insist on installing without the availability of an internet
 connection.
 [I'm in rural SW MO on the end of a 56k dial-up connection ;]

I see.  You got no speed limit for automobile but speed limit for
Internet.  56k should be practically impossible for tracking 
testing without apt-zip.

If you have access to fast network somewhere, download DVD-1 at least.
That should give you decent system. (There are 2 more DVDs to be
complete.)
  http://www.debian.org/devel/debian-installer/
  http://www.debian.org/devel/debian-installer/errata

All you need is few file images.  You do not need network to install DVD
or CD.  (I checked how big archive is, it seems DLBL image is only
one... so guess)

If you do not install large desktop application and keep it just with
consoles, tracking testing may be possible with direct 56k connection.

Regards,

Osamu


-- 
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/20121117154638.GA18311@goofy.localdomain

Re: Password problem confirmed as repeatable - was [Re: Newbie password problem(s)]

[Osamu Aoki]

Hi,I thought about different approach ...

Since 

On Sun, Nov 18, 2012 at 12:46:38AM +0900, Osamu Aoki wrote:
 Hi,
 
 On Sat, Nov 17, 2012 at 06:31:50AM -0600, Richard Owlett wrote:
 ...
  I've been a computer user since before Mr. Torvalds was born. I
  just never had reason to have contact with *nix. Even when working
  for DEC I was much more into analog than digital.
 
 I see.  I guess you were a VMS or some mainframe user...
 
 As for apt-zip:
  http://alioth.debian.org/projects/apt-zip
  http://packages.qa.debian.org/a/apt-zip.html
  http://bugs.debian.org/cgi-bin/pkgreport.cgi?src=apt-zip
 
 It seems somewhat stalled.  You may wish to takeover project to update
 it to be compatible with current apt repository.  The last update was
 2008.

Usually, I recommend not to try to make full archive copy to reduce our
server load.  But if you are careful, this may be a simpler solution
than fixing apt-zip.  You may also avoid some parts of archive.  These
debian package maybe useful to be as efficient as possible to make
partial mirror.

 debmirror  http://packages.qa.debian.org/d/debmirror.html
   (This is made by debhelper developer and current)
 apt-mirror http://apt-mirror.sourceforge.net/
http://packages.qa.debian.org/a/apt-mirror.html
 apt-p2phttp://www.camrdale.org/apt-p2p/
http://packages.qa.debian.org/a/apt-mirror.html

Osamu



-- 
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/20121117161801.GA19855@goofy.localdomain

Re: debmirror failing to download sources (*.orig.*)

[Osamu Aoki]

Hi,

On Sat, Nov 17, 2012 at 08:20:57PM +1100, Zenaan Harkness wrote:
 Gday, here's my debmirror cmd:
 
 /usr/bin/debmirror --source --i18n --exclude='/Translation-.*\.bz2$'
 --include/Translation-(en).*\.bz2$' --exclude-deb-section=games
 --exclude-deb-section=debug --exclude-deb-section=news --verbose
 --progress --arch=amd64,i386
 --section=main,main/debian-installer,contrib,non-free
 --dist=wheezy,sid --host=ftp.iinet.net.au --method=ftp
 --root=debian/debian /public/debian/debian
 
 (above should be a single line)
 
 Here is the output (somewhat abbreviated since I deleted
 /public/debian/debian/.temp):
 Mirroring to /public/debian/debian from
 ftp://anonym...@ftp.iinet.net.au/debian/debian/
 Arches: amd64,i386
 Dists: wheezy,sid
 Sections: main,main/debian-installer,contrib,non-free
 Including source.
 Pdiff mode: use
 Will clean up after mirroring.
 Attempting to get lock ...
 Updating remote trace files (using rsync) ...
 iiNet Rsync Server

I do not know why but you asked for ftp method but rsync is used.
 
 - Only have one rsync connection Per IP
 - No more then two rsync updates of a package per day
 
 Excessive connections may result in your IP being firewalled for 24 hours.

They do not like using rsync this way.

Please check:
 /etc/debmirror.conf
 ~/.debmirror.conf


-- 
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/20121117162815.GA20634@goofy.localdomain

Re: Password problem confirmed as repeatable - was [Re: Newbie password problem(s)]

[Osamu Aoki]

Hi,

On Fri, Nov 16, 2012 at 01:53:47PM -0600, Richard Owlett wrote:
 Terho Uotila wrote:
 Unless you want to try very minimal system you probably want shadow
 passwords.
 
 I do not understand.
 Going by text displayed during the installation  I had the
 impression that shadow passwords were for those overly paranoid
 about supposed incremental security advantages. My machines will
 effectively be the ultimate in single user systems. I will be the
 only one with physical access. They will have only occasional
 restricted access to the internet over a dial up connection. It is
 very unlikely that there will ever be a LAN among my machines. I do
 not see that Debian even claims that shadow passwords *could* be of
 any benefit to me ;/

1. Use of shadow is almost standard these days.

2. If your are attacked while you are using user account to access
   internet, the use of shadow password gives you one thin but extra
   layer of protection to protect your root account after your user
   account is hacked.

3. Disabling shadow does not give you much speed or space.  Configuring
   system to odd configuration is usually not good idea since it is
   tested less.

 On Thu, 15 Nov 2012 11:04:21 -0600
 Richard Owlett wrote:
 
 2. Did not enable shadow passwords
 (...)
 3. Attempted su, neither root nor user password accepted
 
 You may have hit a bug.
 
 Obviously there is a bug.
 The question is it me or Debian?

In this context, more likely Debian or its upstream.
 
 I noticed a problem with disabling shadow
 passwords some time back and I don't know if anyone has fixed it, as
 (ahem) I haven't gotten around to reporting it yet...
 
 For you I have such a deal.
 I've submitted a bug [693...@bugs.debian.org] on which you may
 comment.
 This will be especially useful if the problem has been observed in
 something later than version 6.0.5.
 
 
 I recall (maybe wrongly) figuring out that when shadow passwords were
 not enabled, installer still saved root password into shadow before
 disabling it, or something on those lines.
 
 I don't know. Though ~3 score and 10, this is my first foray into
 *nix.

Welcome and if you are newbie, stay within safe area :-)
Do not try to do something complicated before getting system understood.

 I could fix my installation at the time by going into single user mode
 and changing root password, after which it worked correctly.
 
 I don't understand that.

I did not know this but passwd package ships command called
shadowconfig.  Use shadowconfig on.  

See man shadowconfig and  /usr/share/doc/passwd/README.Debian for more.
(If you installed from CD, you may alternatively reinstall system ...)

By the way, we have some basics for use as:
  http://www.debian.org/doc/manuals/debian-reference/

Good luck,

Osamu 


-- 
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/20121117073208.GA25050@goofy.localdomain

Re: Number of Debian packages available.

[Osamu Aoki]

On Tue, Oct 16, 2012 at 11:30:41AM +0100, Lisi wrote:
 Hello, all!
 I have searched Wikipedia and the Debian wiki.  I have Googled.  I am clearly 
 using the wrong search terms, although I tried rewording in sundry different 
 ways.
 
 Approximately, in round terms, how may packages are available in Debian 
 (Squeeze?)
 1. in main
 2. in main, contrib and non-free

See Debian Reference on Debian site which should be displaying such
number scanned last night :-) I just updated.

  
http://www.debian.org/doc/manuals/debian-reference/ch02.en.html#listofdebianarchivearea

 area number of packages
 main38064
 contrib   231
 non-free  508

This is for upcoming wheezy :)

 I have an idea of roughly 20,000 in my head, but cannot remember why I think 
 it and it may be vastly out.  Nor into which of my two categories the figure 
 falls, if by any miracle it is correct.

Off by half

Osamu


-- 
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/20121019151512.GA21775@goofy.localdomain

Re: Number of Debian packages available.

[Osamu Aoki]

On Tue, Oct 16, 2012 at 02:36:43PM +0300, Andrei POPESCU wrote:
 On Ma, 16 oct 12, 14:00:10, Andrei POPESCU wrote:
  
  $ aptitude search ~Astable | wc -l
  43004
  $ aptitude search ~Astable~scontrib | wc -l
  271
  $ aptitude search ~Astable~snon-free | wc -l
  583
 
 Oups, 'stable' will also match 'unstable', so the correct search (also 
 excluding non-Debian sources) is:
 
 $ aptitude search '~A^stable~ODebian' | wc -l
 28875
 $ aptitude search '~A^stable~ODebian~scontrib' | wc -l
 189
 $ aptitude search '~A^stable~ODebian~snon-free' | wc -l
 408
 
 My numbers are still a bit bigger than the count from the Packages file, 
 but don't know why (it doesn't seem to be due to virtual packages).

I think your number is about right for squeeze.

Debian FAQ package number is not scripted.  So needs to file bug report
to package.

Osamu


-- 
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/20121019152347.GB21775@goofy.localdomain

Re: [LAU] OnDemand-performance was(OT: klang)

[Osamu Aoki]

On Fri, Aug 10, 2012 at 04:18:23PM +0200, Ralf Mardorf wrote:
 On Fri, 2012-08-10 at 00:22 +1200, Chris Bannister wrote:
  On Mon, Aug 06, 2012 at 10:34:03AM +0200, Ralf Mardorf wrote:
   every email address in the mail is shown as is, so at least respect
   the privacy of others and remove email addresses, or much better,
  
  If you post to a mailing list, there *is* no privacy.
 
 There are different levels of privacy, at least mailman archives will
 hide the email addresses and some e.g.
 http://lists.jackaudio.org/listinfo.cgi/jack-devel-jackaudio.org only
 allows to read the archive, when you're subscribed.

But you are on Debian mailing list.  It is all public...

Mailing list is meant to be public.

Osamu


-- 
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/20120811121400.GB14240@goofy.localdomain

Re: Not getting mailing list response in my inbox

[Osamu Aoki]

On Sat, Jun 23, 2012 at 03:52:10PM +, Camaleón wrote:
 On Sun, 24 Jun 2012 00:20:46 +0900, Osamu Aoki wrote:
 
  On Sat, Jun 23, 2012 at 03:32:27PM +0530, Harshad Joshi wrote:
  Hello to admins and mailing list managers...
  
  I am not able to receive responses posted to my topics i mailed to the
  list..Can someone help to sort out this issue? Is it gmail that is
  preventing mails from getting in? #help
  
  This is a famous Gmail feature.
 
 (...)
 
 I think you mean this:
 
 ***
 Not receiving email from groups
 http://support.google.com/mail/bin/answer.py?hl=entopic=1668979hlrm=esctx=topicanswer=6588
 ***
 
 What Gmail's webmail filters is _your own_ replies but not the other user 
 replies to your posts; that would be too much insane... even for 
 Google ;-P

True.

What is the chance of these replies to be in Google's spam box?

I do not know about this rule.  At least once they put one message to
spam box, the whole thread seems to ge there.  ... That was my vague
feeling ...

Osamu


--
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/20120624125817.GB24712@goofy.localdomain

Re: Not getting mailing list response in my inbox

[Osamu Aoki]

Hi,

On Sat, Jun 23, 2012 at 03:32:27PM +0530, Harshad Joshi wrote:
 Hello to admins and mailing list managers...
 
 I am not able to receive responses posted to my topics i mailed to the
 list..Can someone help to sort out this issue? Is it gmail that is
 preventing mails from getting in? #help

This is a famous Gmail feature.

You are sending mail from Gmail (Web interface I suppose.)

Then Gmail do not put looped mail into your inbox.

There are some archive mail here explaining where on Google help page
this is documented.

Osamu


-- 
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/20120623152046.GA24894@goofy.localdomain

Re: Random display freezes [was: Less responsive input...blind-typing]

[Osamu Aoki]

Hi,

Where is your /tmp mounted.

Please post mount output.

On Fri, Jun 01, 2012 at 09:11:42AM -0700, John Magolske wrote:
 * John Magolske listm...@b79.net [120525 17:13]:
  For a while now I've been experiencing a strange behavior where there
  are frequent and regular freezes when typing or issuing commands
  in the shell. All of a sudden I'll find myself blind-typing for 3
  seconds or so before a backlog of characters burst onto the screen.
 
 Ok, so I opened up htop to look up  kill off non-essential daemons,
 adjusted vm.dirty_ratio and vm.dirty_background_ratio [1], tried
 changing and tuning the IO scheduler [2] (switched from cfq to
 deadline) ... all to no avail, still these infernal 3-second freezes.

This makes things too complicated to resolve.  Keep as much in default
otherwise others will not know how things are on your end.
 


-- 
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/20120602133952.GB4571@localhost

Re: Random display freezes [was: Less responsive input...blind-typing]

[Osamu Aoki]

Hi,

On Sat, Jun 02, 2012 at 12:45:12PM -0700, John Magolske wrote:
 Hi,
 
 Thanks for the reply.
 
 * Osamu Aoki os...@debian.org [120602 09:53]:
  Where is your /tmp mounted.
 
 Under the main filesystem. Are you thinking of suggesting I mount
 /tmp as a ramdisk?

Since you are on sid, I was wondering if this is some regression of
mounting /tmp as tmpfs (not-exactly RAMDISK but practically the same.
tmpfs may be swapped out to harddisk, though.)
 
  Please post mount output.
 
  ~ % mount
 rootfs on / type rootfs (rw)
 sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
 proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
 udev on /dev type devtmpfs 
 (rw,relatime,size=1498260k,nr_inodes=217054,mode=755)
 devpts on /dev/pts type devpts 
 (rw,nosuid,noexec,relatime,gid=5,mode=620,ptmxmode=000)
 tmpfs on /run type tmpfs (rw,nosuid,noexec,relatime,size=300952k,mode=755)
 /dev/disk/by-uuid/---- on / type ext3 
 (rw,relatime,errors=remount-ro,commit=5,barrier=1,data=ordered)
 proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
 sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
 tmpfs on /run/shm type tmpfs (rw,nosuid,nodev,relatime,size=601900k)
 fusectl on /sys/fs/fuse/connections type fusectl (rw,relatime)
 debugfs on /sys/kernel/debug type debugfs (rw,relatime)
 binfmt_misc on /proc/sys/fs/binfmt_misc type binfmt_misc 
 (rw,nosuid,nodev,noexec,relatime)

Hmmm OK.  Not much different as here.

udev on /dev type devtmpfs (rw,relatime,size=1974800k,nr_inodes=493700,mode=755)
devpts on /dev/pts type devpts 
(rw,nosuid,noexec,relatime,gid=5,mode=620,ptmxmode=000)
tmpfs on /run type tmpfs (rw,nosuid,noexec,relatime,size=396680k,mode=755)
/dev/mapper/Y-X on / type ext4 
(rw,noatime,errors=remount-ro,user_xattr,acl,commit=600,barrier=1,data=ordered,discard)
tmpfs on /run/lock type tmpfs (rw,nosuid,nodev,noexec,relatime,size=5120k)
proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
tmpfs on /run/shm type tmpfs (rw,nosuid,nodev,relatime,size=793360k)
/dev/sda1 on /boot type ext4 
(rw,noatime,user_xattr,acl,commit=600,barrier=1,data=ordered,discard)
none on /sys/kernel/security type securityfs (rw,relatime)
fusectl on /sys/fs/fuse/connections type fusectl (rw,relatime)
binfmt_misc on /proc/sys/fs/binfmt_misc type binfmt_misc 
(rw,nosuid,nodev,noexec,relatime)

You said ...
 This is up-to-date Sid on an X200s ThinkPad, running Screen in a
 framebuffer console.

Anyway, try disabling tmpfs first by editting /etc/default/rcS nd restart.  You
seem to be on low memory system.

Also check screen bug pages.  Thare are may bug reported.
  http://bugs.debian.org/cgi-bin/pkgreport.cgi?src=screen

Check this problem is screen related or not.

Osamu


-- 
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/20120603034916.GC25941@localhost

Re: chroot or virtual machine

[Osamu Aoki]

Hi,

On Thu, May 17, 2012 at 11:33:21AM +0100, Roger Leigh wrote:
 On Thu, May 17, 2012 at 09:15:26AM +0100, Chris Davies wrote:
  Roger Leigh rle...@codelibre.net wrote:
   http://people.debian.org/~rleigh/schroot.pdf

This is schroot(8) in PDF :-)

...
 Arguably, we should probably document the setup procedure.  While
 we describe all the configuration options, we don't detail how to
 set up the actual chroot--it's assumed it already exists, though
 we should probably include examples of how to create them.

I think you have enough information already.  They are just a bit
convoluted, though.

In schroot(8), you have sbuild(8) as SEE ALSO.

Then, in sbuild(8), you have sbuild-setup(7) and sbuild-createchroot(8)
as SEE ALSO.

There are enough information in sbuild-setup(7) and
sbuild-createchroot(8) needed for making chroot.

I wish a bit more direct and obvious connection from  schroot(8) to
the content of sbuild-setup(7). 

In some way, sbuild-setup(7) could be renamed as schroot-setup(7) and
moved into schroot package, etc. There may be additional reorganization
needed along with this change.  Then everything is quite tidy and nice.

Regards,

Osamu


-- 
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/20120518150832.GA10441@localhost

Re: Installing Wheezy on btrfs only (multi-device)

[Osamu Aoki]

Hi,

On Mon, May 07, 2012 at 03:11:36AM +0200, Steven Post wrote:
...
  I think with skill and knowledge you presented, if you are successful
  doing this with help of shell etc., you should present specific
  procedure needed to do this to d-i BTS as wishlist bug.  That should get
  it supported smoothly for upcoming release.
 
 Good idea.

If you have specific issues with grun, grub maintainer/BTS may be
another option.

Anyway, with your skill, you my get better reply from
debian-b...@lists.debian.org .

Good luck.

Osamu


-- 
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/20120507135137.GA5328@localhost

Re: Installing Wheezy on btrfs only (multi-device)

[Osamu Aoki]

Hi,

On Sun, May 06, 2012 at 02:11:58PM +0200, Steven Post wrote:
 On Sun, 2012-05-06 at 13:51 +0900, Osamu Aoki wrote:
  Hi,
  
  On Sat, May 05, 2012 at 01:24:59AM +0200, Steven Post wrote:
   Hello list,
   
   I'm testing out a new system, the idea is to combine 6 hard drives into
   a single btrfs volume (raid10) (using subvolumes for /, /home, etc). I
   understand that it is now possible to also have /boot on btrfs
   (previously impossible because of GRUB).
  
  I tried to install sysyem with much simpler configuration:
   / :  a single btrfs on a partition  /dev/sda1
  swap: a single swap on a separate partition  /dev/sda2
  
  It did not work due to missing fsck.btrfs in btrfs-tools.
   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=668832
   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=567681
 
 I've had the same problem, but I was aware of it from an earlier attempt
 using virtualbox and a separate ext4 /boot partition. This can be solved
 by changing /etc/fstab (booting from a live cd), look for the line with
 the btrfs system, then change last number on that line to a '0' (the
 'pass' column).

Yah, that's one way.  My bug report cited another work around.

   Unfortunately the Debian installer (daily build for wheezy) is unable to
   create multi-device btrfs volumes.
  
  Yah.. but you probably can do this via shell screen.
 
 I don't think I'm able to do that from the rescue mode shell without
 using an existing installation, without the existing installation I get
 a 'Command not found' error on 'btrfs device scan', I assume this
 includes the whole btrfs command.

It is easiest to do it using an existing installation.

Well... I see btrfs-tools-udeb package.  If you install via expermode,
you may be able to chose to install btrfs-tools-udeb via menu.
Otherwise, install such packages via wget and dpkg to rescue system
provided you have enough memory

   I tried the a normal install and then copy it over to a multi-device
   btrfs system, afterwards adding the remaining disk:
   1) using a single drive to install using ext4 for /boot and btrfs for /
   2) after (successful) installation boot from a live cd (ubuntu 12.04 in
   this case, but it shouldn't really matter)
  
  How successful?  Did it boot?
 
 It boots, yes, but you still need to change /etc/fstab using a live cd
 because of the missing fsck tool.

Yes that is a start.  I usually do this while installing with d-i CD in
advance (or in rescure mode later which is essentially a live cd)

   3) create partitions for btrfs on the other 5 drives using gdisk
   4) create a multi-device btrfs filesystem with those partitions, with a
   subvolume for the root filesystem and for /home
   5) copy all the data from the first disk to the new btrfs volume
   6) chroot into the rootfs of the new volume, after mounting (-o
   bind) /dev, /proc and /sys into the mounted system
   7) change fstab for the new drives and modify /etc/default/grub
   8) run update-grub: this is where things start falling down, I get an
   error about / not being mounted so grub fails.
   
   This is apparently caused by a regression in GRUB [1] (Debian bug
   #538118).
  
  Well this is possibly another problem you are facing.  Basically, you
  should know what file to change imanually when copying files from one
  partition to another.  So this is really an expert trick which d-i
  currently does not support for ordinry uses.  But as I mention in the
  above, most simple installation suffer major breakage for btrfs on
  wheezy even if you manually adjust configuration files.
  
 
 I'll have another look if my other attempts from within the d-i fails.
 
 Another thing I noticed during my tests is that the installer will
 recognize existing btrfs partitions and is able to use them, except when
 dealing with a multi-device file system. Using rescue mode you can trick
 d-i into using them properly by executing a shel in an existing
 installation, then issuing a btrfs device scan. Once the scan is done,

This is my wild guess.  btrfs support of d-i is not so mature yet.

 you can exit the shell and go back to the partitioner. There I could
 select a device (partition) from the btrfs system and use it as /, now
 the mount works with the multi-device partition.
 Perhaps d-i should issue the 'btrfs device scan' command when starting
 the partitioner?

I think with skill and knowledge you presented, if you are successful
doing this with help of shell etc., you should present specific
procedure needed to do this to d-i BTS as wishlist bug.  That should get
it supported smoothly for upcoming release.

 Kind regards,
 Steven

Osamu


-- 
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/20120506133012.GA15043@localhost

Re: Installing Wheezy on btrfs only (multi-device)

[Osamu Aoki]

Hi,

On Sat, May 05, 2012 at 01:24:59AM +0200, Steven Post wrote:
 Hello list,
 
 I'm testing out a new system, the idea is to combine 6 hard drives into
 a single btrfs volume (raid10) (using subvolumes for /, /home, etc). I
 understand that it is now possible to also have /boot on btrfs
 (previously impossible because of GRUB).

I tried to install sysyem with much simpler configuration:
 / :  a single btrfs on a partition  /dev/sda1
swap: a single swap on a separate partition  /dev/sda2

It did not work due to missing fsck.btrfs in btrfs-tools.
 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=668832
 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=567681

 Unfortunately the Debian installer (daily build for wheezy) is unable to
 create multi-device btrfs volumes.

Yah.. but you probably can do this via shell screen.

 I tried the a normal install and then copy it over to a multi-device
 btrfs system, afterwards adding the remaining disk:
 1) using a single drive to install using ext4 for /boot and btrfs for /
 2) after (successful) installation boot from a live cd (ubuntu 12.04 in
 this case, but it shouldn't really matter)

How successful?  Did it boot?

 3) create partitions for btrfs on the other 5 drives using gdisk
 4) create a multi-device btrfs filesystem with those partitions, with a
 subvolume for the root filesystem and for /home
 5) copy all the data from the first disk to the new btrfs volume
 6) chroot into the rootfs of the new volume, after mounting (-o
 bind) /dev, /proc and /sys into the mounted system
 7) change fstab for the new drives and modify /etc/default/grub
 8) run update-grub: this is where things start falling down, I get an
 error about / not being mounted so grub fails.
 
 This is apparently caused by a regression in GRUB [1] (Debian bug
 #538118).

Well this is possibly another problem you are facing.  Basically, you
should know what file to change imanually when copying files from one
partition to another.  So this is really an expert trick which d-i
currently does not support for ordinry uses.  But as I mention in the
above, most simple installation suffer major breakage for btrfs on
wheezy even if you manually adjust configuration files.

 Has anyone attempted such an install before? How should I proceed with
 this?

Good luck.

 Kind regards,
 Steven
 
 [1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=538118



-- 
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/20120506045149.GA6661@localhost

Re: ..when to scrap old junk, was: old machine wheezy vs apt

[Osamu Aoki]

Hi,

On Tue, Apr 17, 2012 at 10:18:16PM +0200, Arnt Karlsen wrote:
 On Tue, 17 Apr 2012 15:10:38 +0100, Dom wrote in message 
 4f8d79de.20...@rpdom.net:
 
  On 16/04/12 21:34, Aleksandar Kostadinov wrote:
   To anybody interested, debian wheezy can't run on 64MB RAM on i386.
   Or actually it can run but package management is unusable so all
   debian benefits are actually lost.

Hmmm I have not tried recently but Debian i386 was for 486 or newer
for good long time.

http://www.debian.org/releases/stable/i386/ch02s01.html.en#id583669

| However, Debian GNU/Linux squeeze will not run on 386 or earlier
| processors. Despite the architecture name i386, support for actual
| 80386 processors (and their clones) was dropped with the Sarge (r3.1)
| release of Debian[2]. (No version of Linux has ever supported the 286 or
| earlier chips in the series.) All i486 and later processors are still
| supported[3].
...
| [2] We have long tried to avoid this, but in the end it was necessary
| due a unfortunate series of issues with the compiler and the kernel,
| starting with an bug in the C++ ABI provided by GCC. You should still be
| able to run Debian GNU/Linux on actual 80386 processors if you compile
| your own kernel and compile all packages from source, but that is beyond
| the scope of this manual.
| [3] Many Debian packages will actually run slightly faster on modern
| computers as a positive side effect of dropping support for these old
| chips. The i486, introduced in 1989, has three opcodes (bswap, cmpxchg,
| and xadd) which the i386, introduced in 1986, did not have. Previously,
| these could not be easily used by most Debian packages; now they can.

This is not RAM size issue but CPU code compatibility issue.

   I tried pulling the .deb files on another machine, then transfer to
   the old machine. But dpkg is killed on 95% reading database (I think
   OOM killer). It fails even if I try to install 1 package.

So install Woddy (r3.0) from old archive ...


-- 
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/20120419132455.GA8136@localhost

Re: apt-get dist-upgrade failed.

[Osamu Aoki]

Hi,

Not all Debian mirrors work right ...

On Wed, Apr 18, 2012 at 11:27:54AM +0530, Amrish Purohit wrote:
...
 500 http://ftp.us.debian.org/debian/ wheezy/main i386 Packages
 N: Unable to locate package libvlccore4

Change mirror site from http://ftp.us.debian.org to and something else
and try again.

Good luck,

Osamu


-- 
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/20120418140741.GA9651@localhost

Re: When will Debian 7.0 with Linux Kernel 3.x be Released?

[Osamu Aoki]

Hi,

On Fri, Mar 30, 2012 at 05:29:47PM +0800, Teo En Ming (Zhang Enming) wrote:
 I am not putting pressure on Debian to change.

Sorry... I do not feel pressure from your postings.

Debian practically release new system every 6 hours :-)
See Life with eternal upgrades:
 
http://www.debian.org/doc/manuals/debian-reference/ch02.en.html#_life_with_eternal_upgrades
 

For people like you or me who needs relatively new system for
non-mission critical desktop system as a technically capable user, we
have sid/testing/experimental... repositories.  We just use the word
release in very narrow sense, i.e., only for stable release.

Debian upgrade scripts makes this eternal upgrades possible.

As for Kernel, we have 3.3 packaged as linux-2.6 3.3~rc6-1~experimental.1
 http://packages.qa.debian.org/l/linux-2.6/news/20120305T193428Z.html

If you really wish us to feel pressure for release, send us patches to
fix RC bugs.  (I think you have to learn how we operate first though.)

As being volunteer organization, it used to take years to make stable
release.  Recently, it have been shortend to 2 year cycle.  We need
volunteer to make this keep happning.  (No verbal pressure but patches!)

Osamu


-- 
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/20120410124130.GB5464@localhost

Re: upgrade to Wheezy fails with aptitude

[Osamu Aoki]

Hi,

On Wed, Mar 21, 2012 at 11:01:38AM +0100, Pierre Frenkiel wrote:
 hi,
 I wanted to upgrade from Squeeze to Wheezy, and as I saw in several places
 that aptitude should be preferred to apt-get,  I first tried with it.
 I started with only 1 line in sources.list:
 deb http://ftp.fr.debian.org/debian/ wheezy main contrib non-free
 and tried several times  aptitude safe-upgrade
 Each time, aptitude stayed indefinitely on resolving dependencies and did
 nothing else.
 I then issued: apt-get dist-upgrade, and all worked perfectly, in less
 than 30 minutes.

So your experience is that apt-get is better for such case :-)

The release note 4.4.6. Upgrading the system has:
 
http://www.debian.org/releases/stable/i386/release-notes/ch-upgrading.en.html#upgrading-full
 The upgrade process for other releases recommended the use of aptitude
 for the upgrade. This tool is not recommended for upgrades from lenny to
 squeeze.
 
 My question: is it better to also revert to apt-get for package management,
 or is the problem specific to the upgrade to a new release?
That was the consensus among DDs when releasing squeeze :-)

The rational for this is experience such as yours.

Almost same is not exactly the same between apt-get/aptitude.

http://www.debian.org/doc/manuals/debian-reference/ch02.en.html#_literal_apt_get_literal_literal_apt_cache_literal_vs_literal_aptitude_literal

Maybe, I should put more warning since aptitude dependency behaviour
seems to be changing a bitfor command line.

Osamu


-- 
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/20120325150858.GA701@localhost

Re: [Feedback needed] Setting the right size for /tmp

[Osamu Aoki]

Hi,

I join too late but ...  (I do not use tmpfs for /tmp)

On Tue, Nov 29, 2011 at 04:13:05PM +, Camaleón wrote:
 On Mon, 28 Nov 2011 18:07:43 +, Camaleón wrote:
 
  I'm running an updated wheezy and today faced with this little
  problematic.
 
 (...)
 
  Okay, so /tmp is full. Fine. I know how to solve it but I can foresee
  more situations like this in the future so some questions arise. As the
  current tmpfs default settings for /tmp seem a bit unrealistic (just %
  20 of the RAM?) for even doing common tasks:

This is just the default value for RAMTMP=yes in the /etc/default/rcS file.
You could have much bigger than RAM size as long as you have big enough
swap to support it.  You can do so via /etc/fstab. I once had 10GB tmpfs
with 4GB RAM so I could have DVD image on /tmp on tmpfs.  So 20% of RAM
is not good enough reason to reject tmpfs :-)

  1/ How many room should be set for a /tmp partition? I never had it
  one so I can't make any good estimation.

This depends on what you run.  DVD data may be as big as 5GB.

  2/ Would be better to simply disable tmpfs for /tmp? This is how I've
  been doing all these years.
  
  Any comments are welcome :-)
 
 Jerome, Bob, Dom... thank you all for your input :-)
 
 After carefully reading your suggestions I have decided to disable tpmfs 
 for /tmp and use the old method of having /tmp inside a partition.

Good.  As long as you have lots of RAM, most data written to disk stays
on RAM anyway as cached data if it is very short lived data.  So this
does not slow system.

But that may cause concern for disk wareout if you are using SSD.

 @Jerome, why not a dedicated partition to hold /tmp? Because I would have 
 to decide a fixed partition size and to be sincere, I don't think there 
 is any gain for this specific case, this is a small netbook I use mainly 
 for testing purposes so I don't need to be excesive careful with security 
 or privacy options nor need for speed :-). I prefer to keep things as 
 easy as possible.
 
 @Bob, thanks for pointing out that development mailing list thread. Very 
 interesting. By reading it, I see this is also issue for other users and 
 I have to agree that the defaults are a bit (to say at least) 
 conservative. I'm usually fine with Debian defaults and try to keep them 
 as long as  there is no compelling reason for editing them, e.g., when 
 they choke with common tasks, like making MC to crash for the simple fact 
 of exploring a 75 MiB compressed file :-) 
 
 @Dom, I agree, having /tmp on the same / directory is also the most 
 suitable deal for me.
 
 So, in the end I have set RAMTMP=no option at /etc/default/rcS.

But if you are on laptop with SSD and lots of memory, you may optimize
diskware by slowing down on disk cache flushing from memory.

 
http://www.debian.org/doc/manuals/debian-reference/ch09.en.html#_optimization_of_solid_state_drive

Anyway, even for mormal usage, use of noatime in mount option seems to
be one of the easiest system optimizer.

I stop using huge tmpfs for tmp since it gains nothing for me.

Regards,

Osamu


--
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/20120304091209.GA7714@localhost

Re: Not getting mails in inbox

[Osamu Aoki]

Hi,

On Mon, Jan 30, 2012 at 11:17:12AM +0530, Harshad Joshi wrote:
 I am not getting replies to mails i am sending on this list. i have to
 manually check or refer google to get answers..
 
 can someone look in this matter?

As someone already mentioned, this is Gmail feature.

My solution is to use my connecton ISP's SMTP server to send mail to the
list using mail header with From: os...@debian.org.

My debian.org mail is configured to forward mails to my gmail account.
So I can get a lot of mails.  Gmail filters spam out and allows me to
get as many large mails via POP3.

Osamu


-- 
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/20120201131117.GA6765@localhost

Re: Re(2): POP3 in Debian

[Osamu Aoki]

Hi,

On Thu, Jan 19, 2012 at 08:49:24AM -0800, peasth...@shaw.ca wrote:
 * From: Osamu Aoki os...@debian.org
 * Date: Thu, 19 Jan 2012 21:10:34 +0900
  Both getmail4 and fetchmail can handle this task well ...
 
 According to the Wikipedia, fetchmail can not deliver to mbox.

Well by itself ... yes you are right.

| fetchmail  is  a  mail-retrieval  and forwarding utility; it fetches
| mail from remote mailservers and forwards it to your local (client)
| machine's delivery system.  You can then handle the retrieved mail using
| normal mail user agents such as mutt(1), elm(1) or Mail(1).

Its all about fetching.  You need another program to deliver to ... like
procmail/maildrop/...

  Are you running that qpopper yourself?
 
 Not now.  It was installed years ago before I learned 
 to simplify configuration.
 
  Or connect to remote POP3 site with telnet. 
 
 That explained the problem in a few minutes.  The MUA 
 is putting two blanks between LIST and 1.  The Zimbra 
 POP3 server is accepting one blank and not two.  My 
 log 

You mean your log viewing environment.

 is a variable pitch font and I failed to notice the 
 two blanks.  Should have noticed in the Courier font 
 emessage but failed again.  The ISP must have changed 
 the POP server last weekend.  No public notice and 
 technical support staff were unaware.  My telnet 
 session follows in case anyone is interested.

Good for you.  The real question is who's bus is this.

If this is bug on Debian package which is a bit too tight than what RFC
allows, please file bug report on Debian.  If this is non-RFC complient
bug, maybe complain to ISP or Zimba POP3 server developer.

...
 Now that I understand, the MUA can be changed to send just 
 one blank and interact with the ISP POP3 directly again.

I think your usage of MUA is not a typical one.

MUA is mutt, thunderbird, ... not POP3 server.




Osamu


-- 
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/20120120131035.GA9761@localhost

Re: POP3 in Debian

[Osamu Aoki]

Hi,

Both getmail4 and fetchmail can handle this task well (Though I like
the getmail4 package as being its maintainer.)

On Wed, Jan 18, 2012 at 01:49:22PM -0800, peasth...@shaw.ca wrote:
 Hi,
 
 Since the weekend my MUA POP3 connection has failed to receive 
 messages with this appearing in the log.
 --- POP
 RCV: +OK pop.shaw.ca
 SND: USER peasthope
 RCV: +OK password required for user peasth...@shaw.ca
 SND: PASS 
 RCV: +OK server ready
 SND: UIDL
 RCV: +OK 35 messages
 SND: LIST  1
 RCV: -ERR unable to parse msg
 SND: RETR  1
 RCV: -ERR unable to parse msg
 SND: QUIT
 
 Failure in parsing LIST 1 baffles.  How big a problem can parsing of 6 
 characters be?  To understand further, I want a Debian system to 
 automatically 
 retrieve messages from the POP3 server and put them in /home/peter/mbox.  
 Then qpopper can let the preferred MUA pop them from the mbox.  Which 

Qpopper
The most widely-used POP3 server for UNIX, Linux, etc.
http://www.eudora.com/products/unsupported/qpopper/index.html

Are you running that qpopper yourself?

 software should be configured to retrieve from the ISP POP3 server and 
 store in mbox?  Exim, getmail or another?

I think your message is quite confuting.

Anyway, run your MUA with more verbosity.  mutt have -d option.

Or connect to remote POP3 site with telnet.  

 
 Thanks,  ... Peter E.
 
 
 
 -- 
 Telephone 1 360 639 0202.  Former telephone 1 360 450 2132.  
 bcc: peasthope ... shaw.ca
 http://carnot.yi.org/;
 http://members.shaw.ca/peasthope/index.html#Itinerary;
 
 
 -- 
 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/171057394.56406.45108@cantor.invalid


-- 
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/20120119121034.GA3832@localhost

Re: Stripping down Debian Squeeze

[Osamu Aoki]

HI,

On Fri, Jan 13, 2012 at 04:56:09PM +0530, Bijoy Lobo wrote:
 Hello Everyone,
 
 I am currently Debian Squeeze for a UTM ( Squid3+Snort+OpenVPN+iptables). I
 would like to strip down the OS for better performance of the UTM. It would
 be acting as a gateway to my network so if there is any software within the
 OS which i can uninstall for better performance would certainly help me.

For such system, all desktop application should go.  More for security
and space than performance.

You can remove almost everything but base system as long as you keep
dependency of these programs and required kernel modules for your
hardware.  Oh, keeping aptitude and vim will help your productivity.

This is something you do not get quick short answer... try and learn
yourself using tools such as aptitude.

Osamu


-- 
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/20120114083342.GB15272@localhost

Re: Squeeze64: LD_LIBRARY_PATH / ignored or overwritten under X *SOLVED*

[Osamu Aoki]

On Fri, Jan 13, 2012 at 09:05:48AM +0100, Denny Schierz wrote:
 hi,
 
 Am 12.01.2012 um 23:11 schrieb Andrei Popescu:
 
  On Jo, 12 ian 12, 12:58:42, Denny Schierz wrote:
  
  What could be the problem?
  
  Your display manager doesn't source the system shell configuration 
  files. If you need a solution per user put that stuff in ~/.xsessionrc 
  otherwise move the file under /etc/X11/Xsession.d/
 
 nope, that isn't the problem. The ssh-agent resets the variable.
 Disable ssh-agent in /etc/X11/Xsession.options and it works (again).
 My display-manager is GDM2.

I do not know if you understood the meaning of what Andrei said.

At least nope is a wrong statement.

If you read
| $ cat /etc/X11/Xsession.options 
| # $Id: Xsession.options 189 2005-06-11 00:04:27Z branden $
| #
| # configuration options for /etc/X11/Xsession
| # See Xsession.options(5) for an explanation of the available options.
| allow-failsafe
| allow-user-resources
| allow-user-xsession
| use-ssh-agent
| use-session-dbus

Then you should know this is for /etc/X11/Xsession.

And browsing /etc/X11/Xsession indicate this script runs script files
under /etc/X11/Xsession.d/ with run-parts fashon.  The ssh-agent script
is just one of them.

So if you put a shell sourceable file in /etc/X11/Xsession.d/ after
90gpg-agent
90x11-common_ssh-agent
 but before
99x11-common_start
then that your set-up will not be reset by ssh-agent.

So name it like
  99x11-common_ssh-agent-environment

This is what Andrei ment while assuming you knew roughly how X starting
process works.

Good luck.

Osamu


-- 
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/20120113163855.GA9452@localhost

Re: pure debian testing or a distro based on debian testing

[Osamu Aoki]

On Fri, Jan 13, 2012 at 10:53:35PM +0100, Andrej Kacian wrote:
 On Tue, 10 Jan 2012 15:11:58 +0530
 Amrish Purohit amrish.dis...@gmail.com wrote:
 
 Hi Osamu,
 Thanks for your reply. I will definitely try installing testing over
 stable version. I make my system dual boot so I have a backup. Here is
 the way I install the debian, please correct me if I am wrong.
 I install the core system, then I install gdm3 and gnome-terminal to get
 the GUI. After getting the GUI I install the packages as I need it. So I
 plan to do the same with testing. First I will install core system +
 gdm3 of stable debian, after getting the gui, I will change the
 repositories in /etc/apt/sourcelist and make dist upgrade. I hope this
 will work.
 Please comment if any thing is wrong.
 
 Two points:
 
 - Wouldn't it be better to change the source.list to debian testing straight
 after installation? That way you will not need to download Gnome packages from
 stable, saving bandwidth and time.

That exactly I meant to suggest but Amrish maybe uncomfortable working
without GUI terminal or I may have failed to explain well...  Console is
less problematic.  As long as he stays within X, gdm3 and
gnome-terminal, it is not really bad.  (But X xdm and xterm may be even
better.)
 
 - If you want Gnome, it is probably better to use tasksel[1], so that all the
 Gnome-related goodies are installed and configured. 

Very true.

 Installing just login
 manager and terminal application will only bring in minimal necessary
 dependencies, giving you a rather poor GUI experience - you might as well stay
 in text console.

I do not know what he mean by the core system.  I hope it meant up to
simple X.  gnome-core package is already huge :-)

Installing task-gnome-desktop package and its all depends and recommends
is good idea (which I think tasksel menu will do.)

Osamu



 
 1. http://wiki.debian.org/Gnome#task
 
 Regards,
 -- 
 Andrej
 
 
 -- 
 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/20120113225335.388635b5@penny


-- 
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/20120114030957.GB6823@localhost

Re: pure debian testing or a distro based on debian testing

[Osamu Aoki]

Hi,

On Mon, Jan 09, 2012 at 07:02:01PM +0530, Amrish Purohit wrote:
 Hi all,
 I am using debian stable as my primary OS. I want to involve in
 debian and as a first step I would like to install debian testing. I
 have tried to install debian testing with various snapshot of weekly
 build, but my system get stuck when gui come up ,keyboard and mouse
 is not working.

Sometime installer does not work well for testing/unstable.

 I think I should go with a distro based on debian testing.
 (considering mint on debian testing). what should I do?

Maybe ... if such thing exist.  But do you know most of us install
stable system and upgrade it to　testing/unstable.

 Should I go with pure debian testing or a distro on debian testing?

I think you should at least try upgrading system after installing stable
system on separate partition as dual boot.

If you want to make you life easy, install non-desktop small system
first.

Then edit /etc/apt/souces.list to replace stable or squeeze with wheezy and run
 # apt-get update
 # apt-get dist-upgrade

Then run tasksel or with aptitude, install desktop task and you are
done.

You should at least read this section:

http://www.debian.org/doc/manuals/debian-reference/ch02.en.html#_debian_package_management_prerequisites

Osamu


-- 
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/20120109140432.GA24617@localhost

Re: resolving dependencies in aptitude

[Osamu Aoki]

Hi,

On Wed, Dec 28, 2011 at 02:02:30PM +, Hendrik Boom wrote:
...
 Painfully aware.  I fled to xfce.  I may go back when it stabilizes.  In 
 the meantime, it's getting in the way of smooth upgrades.  I guess I 
 could uninstall gnome, but that seems drastic.

Hey, same here ... a refugeee user of xfce.

uninstalling gnome is becoming almost ritural for me for major upgrade.
Done that for several releases.  Aptitude makes it very easy.

Here is what you do with aptitude.
/^gnome$

Now you see gnome package.

Down to Depends and Recommends.
Press m to killl A markings for each.

Now you can remove gnome without having major issue.  Remove package as
you need.

Cheers,

Osamu


-- 
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/20120106130644.GB4147@localhost

Re: problems with having two DHCP servers...

[Osamu Aoki]

Hi

On Thu, Jan 05, 2012 at 04:26:20AM -0500, Rick Thomas wrote:
 On 12/27/11 22:04, Scott Ferguson wrote:
 
 It sounds like you are running two DHCP servers - in which case you
 have four options (none of which involve preseeding).
 
 If you have multiple DHCP servers the problem is *easily* fixed - please
 tell me the make and model of the primary (router/firewall) DHCP server
 and I'll give you instructions.
 
 Hi Scott,
 
 I'm not the OP, but I do have this problem.  When I try to do an
 install (wheezy) on a network with two DHCP servers, the installer's
 dhcp-client never seems to get an IP address -- even though the two
 servers are both responding and both giving the same IP address.

Then why have 2?

 Both DHCP servers are dnsmasq.

Why having 2 in a network.  Keeep them separated is the way to go.

 When I kill off one of the servers temporarily, all goes well.

I thought this is the right solution.

 Is there a solution that isn't so drastic?

Make a sane network configuration, I think...

Osamu


-- 
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/20120105150747.GA7425@localhost

Re: lenny hosting wheezy chroot

[Osamu Aoki]

Hi,


On Mon, Dec 26, 2011 at 11:15:29PM -0800, Ross Boylan wrote:
 I am looking for advice on how to run a wheezy chroot from a lenny host,
 in particular how to handle dev, udev, and X.
 
 BACKGROUND
 
 My system is running lenny and I wanted to use python 2.7.  I upgraded
 debootstrap from backports and then created a wheezy (actually,
 testing) chroot with --variant=minbase.  After some rough  spots (I
 needed to bind mount sys and proc, and eventually /dev/pts, and install
 a package so debconf could ask questions during the install) thing seem
 to be running.
 
 Now I want to try mythtv in the chroot, which requires X, which I
 believe requires /dev to be hooked up properly.  The Debian Reference
 for lenny provides some instructions about setting up using a different
 virtual terminal 

Yes since it worked on earlier days, I wrote it.  Maybe, by lenny it may
have been outdated content.

 (the current version does not:
 http://www.debian.org/doc/manuals/debian-reference/ch09.en.html#_chroot_system.
  The lenny reference suggests running MAKEDEV.  Most other sources suggest 
 mount -bind is the way to go, but some people caution this can lead to 
 trouble (and there were some Debian bugs 
 (http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=623060_ in this scenario).
 

Anyway, some X start up codes and other packages had major changes which
made it imossible for me to do this.  I then said, heck .. too much
troble.  it is much easier using kvm or other newer technology.

In order to avoid false expectation, I removed such contents.

So I do not bother using chroot for X application any more.

Please note we DD always use chroot for building pakcages.  so chroot
with bind mounting devices are valid for many situation.

 FYI I have a single core P4 without hardware virtualization.  This is
 one reason I'm trying a chroot and not a VM.

virtualbox is likely good choice.

 I also thought a chroot would be simpler :)  

If it works :-)

 I am not especially concerned about chroot
 security, except in the sense that I don't want to inadvertently mess up
 the host from inside the chroot.

You can read through scripts such as pbuilder and schroot to see how
people handle some issues.  But they usually do not do X.

I think use of or making your own backported package is the way to use
such old box semi-stably.

Osamu


-- 
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/20120103121859.ga26...@goofy.lan

Re: qemu KVM: can't boot a qemu machine with liveCD (system rescue)

[Osamu Aoki]

On Tue, Jan 03, 2012 at 11:09:06AM +0100, Ekkard Gerlach wrote:
 Hi, 
 
 there is no special forum/mailinglist existing for qemu-KVM, right? 

I do not know the answer but have you read:
  http://wiki.debian.org/KVM
  http://wiki.debian.org/QEMU
 and links from there.
  Good luck.

Osamu 


-- 
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/20120103124356.gb26...@goofy.lan

Re: displaying Japanese in the virtual consoles/terminals in squeeze

[Osamu Aoki]

Hi,

On Sun, Dec 25, 2011 at 09:10:48PM +0900, Joel Rees wrote:
 For reasons I don't care to debate, I have set my system's default
 language to Japanese. When I go to work in the virtual console windows
 (for instance, apt-get), I tend to get messages that should be in
 Japanese, but the Japanese is replaced by the Unicode replacement
 characters (?), which is no problem unless I hit some errors. Which,
 of course is a problem..

Normal Linux console is only good for UTF-8 within latin-like
characters.
 
 Trying to search the web for this, and all I can see is old stuff,
 stuff about game consoles, and stuff about setting up X11. I know I
 should be able to get better results, but I'm not very imaginative.

For example, Debian console installer runs on virtual console with
japanese display or any fancy characters.  This is because they run
special terminal program.

 Should be the same set of issues, except for the specific parameters,
 for any large-character-set language.

  jfbterm
  fbterm-ucimf
  fbterm
  bogl-bterm

Try one of these with appropriate font packages.  I think you can get
Japanese display.

 Anyone with a clue for me?

But for all practical purpose, Use X or run aptitude on normal Linux
console under any Locale with:

$ sudo LANGUAGE=fr_FR.UTF-8 aptitude
 or 
$ sudo LANGUAGE=en_US.UTF-8 aptitude

Then you can read it :-)

Osamu


-- 
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/20111225164741.ga18...@goofy.lan

Re: Find which package has installed some file under /etc

[Osamu Aoki]

Hi,

On Tue, Dec 20, 2011 at 11:06:04PM +0200, Andrei Popescu wrote:
 On Ma, 20 dec 11, 20:56:11, Osamu Aoki wrote:
  On Tue, Dec 20, 2011 at 12:29:01AM +0100, Vincent Lefevre wrote:
   The point is what the Debian Policy says. 
  
  Anyway, if you feel strong to enforce this ipart of policy, most
  effective thing to do is file a wishlist bug with patch to lintian to
  enforce package_name for both init.d script and default file.
 
 AFAIK init.d scripts always belong to a package, so not the same thing 
 (name clashes would be detected by dpkg).

Yah ... that is why no DD cares about hypothetical name crash and live
with common sense.

Osamu



-- 
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/20111221151544.gb8...@goofy.lan

Re: Find which package has installed some file under /etc

[Osamu Aoki]

On Tue, Dec 20, 2011 at 12:29:01AM +0100, Vincent Lefevre wrote:
 The point is what the Debian Policy says. 

Anyway, if you feel strong to enforce this ipart of policy, most
effective thing to do is file a wishlist bug with patch to lintian to
enforce package_name for both init.d script and default file.

I realy do not see any major issue with current state relying common
sense and some huristics to handle common case such as apache2.

Osamu


-- 
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/20111220115611.ga3...@goofy.lan

Re: Find which package has installed some file under /etc

[Osamu Aoki]

Hi,

On Mon, Dec 19, 2011 at 01:02:22AM +0100, Vincent Lefevre wrote:
 On 2011-12-18 13:18:02 -0700, Bob Proulx wrote:
  The namespace is defined by Debian Policy.  The filename should be
  named after the package name.  Since the package names must be
  different the file name derived from it must be different.  (I think
  it is okay for /etc/default/foo to be part of a foo-common package
  though.  The namespace intention has been preserved.)
 
 It's not always the case:

did you check apt-cache showsrc ...
 
 /etc/default/alsa created by alsa-base
Package: alsa-driver
Binary: linux-sound-base, alsa-base, alsa-base-udeb, alsa-source

Not exactly in any way now but close enough

 /etc/default/apache2 created by apache2.2-common
Package: apache2
Binary: apache2.2-common, apache2.2-bin, apache2-mpm-worker,
apache2-mpm-prefork, apache2-mpm-event, apache2-mpm-itk, apache2-utils,
apache2-suexec, apache2-suexec-custom, apache2, apache2-doc,
apache2-prefork-dev, apache2-threaded-dev, apache2-dbg

So this is following source package name

 /etc/default/bluetooth created by bluez
Package: bluez
Binary: libbluetooth3, libbluetooth3-dbg, libbluetooth-dev, bluetooth,
bluez, bluez-dbg, bluez-alsa, bluez-gstreamer, bluez-cups,
bluez-pcmcia-support, bluez-compat, bluez-utils, bluez-audio

So this is following a associated package name from the same source
package

 /etc/default/bootlogd created by initscripts

Package: sysvinit
Binary: sysvinit, sysvinit-utils, sysv-rc, initscripts
sysvinit-utils: /sbin/bootlogd
initscripts: /etc/default/bootlogd

...

and the rests are almost same situation.  Some are shorthand name ... others
are one of those associated package name.


-- 
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/20111219140207.ga10...@goofy.lan

Re: Safety while network install.

[Osamu Aoki]

On Mon, Nov 21, 2011 at 04:34:26PM +, Curt wrote:
 On 2011-11-21, Osamu Aoki os...@debian.org wrote:
 
  But seriously, Debian is configured as a quite secure system at any time
  unless you make stupid configuration yourself.  So it is quite safe.
 
 
 Would you be so kind as to explain to me what ports/services are
 open and listening on a default install of Debian Squeeze (if any) and if 
 there
 are any security implications for the novice user or hardening to be
 performed on a default install (in relation to listening daemons)?

Anyway, read good source.
http://www.debian.org/doc/user-manuals#securing
http://www.debian.org/doc/manuals/securing-debian-howto/index.en.html
 
 I got rid of avahi and another daemon that opened listening ports because I
 didn't know what those services were exactly, what purpose they served,
 or whether they presented any kind of danger to my security. (I think
 the other service was portmap, now that I come to think of it).

http://en.wikipedia.org/wiki/Avahi_(software)
http://www.debian.org/doc/manuals/debian-reference/ch05.en.html#_the_hostname_resolution

It gives you  multicast DNS/DNS-SD service discovery.  hostname IP
resolution on LAN without DNS.

Just do not run if you do not need it.  task-desktop pulled in via
recommends to avahi-daemon so you can remove it easily. 
 
 I remember in the past the xserver would listen by default (at least on
 some linuxes), but now the default apparently is not to listen, which is
 a good thing, though it took a while to get there.

Anyway, checking it yourself for your system is good idea than asking
such thing in general sense.

Osamu


-- 
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/2022181612.ga19...@goofy.lan

Re: Safety while network install.

[Osamu Aoki]

Hi,

On Mon, Nov 21, 2011 at 12:13:41AM +0700, Sthu Deus wrote:
 Good time of the day.
 
 
 I'm concerned on safety of install over network (the netinst) - what
 techniques are used to protect the installed system during the very
 process of installation?

instalation data is transmitted via http.  But its data integrity is
checked by GPG key and hush values.

 Or it includes some risk for the install and therefore media (kind of
 disk) is a preferred installation method?

Oh, practically the same.

 Thanks for Your time.

The only extra risk you expose your system via your network installation
is possibility of attack via network to the system.  If you download a
CD image, you are exposing a PC which downloads image but not the PC
installing Debian.

But seriously, Debian is configured as a quite secure system at any time
unless you make stupid configuration yourself.  So it is quite safe.

Osamu

 
 -- 
 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/4ec93549.879acc0a.43d1.c...@mx.google.com


-- 
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/2021135226.ga16...@goofy.lan

Re: gnome 3 refugee: obstructive panel in xfce

[Osamu Aoki]

On Fri, Nov 18, 2011 at 08:52:46PM +, Hendrik Boom wrote:
 I've fled from gnome 3 and am mostly happy with xfce.  But there's the 
 panel with six icons at the bottom of the screen that obscures the bottom 
 of a lot of windows.  How can I get rid  of it, or move it elsewhere?  Is 
 it possible to move it to be part of the strip at the top, for example?  
 In a pinch, I suppose I could get rid of it, because all those tools seem 
 to be available as part of the applications menu anyway. 
 
 The screen savers are pretty, but would it be possible to have them just 
 go blank and save power when the laptop in not plugged in?

I put launchers on the top panel and removed the lower panel.  Right click and
play :-)  Clean and meat.

Osamu


-- 
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/2019154905.ga16...@goofy.lan

Re: Xfce: GIMP as default PDF viewer ?

[Osamu Aoki]

On Thu, Nov 17, 2011 at 04:45:19PM +0100, Jerome BENOIT wrote:
 Hello List:
 
 Since I migrated from Gnome3 to Xfce, I have noticed that thdefault viewer 
 for PDF files is now GIMP:
 I would rather expect acrobat or evince.
 
 Where is the place to fix this ?

Install epdfview.

Please check all recommends of task-xfce-desktop is installed using
aptitude.

Osamu


-- 
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/2017163738.gb7...@goofy.lan

Re: USB boot from HardDisk

[Osamu Aoki]

Hi,

On Fri, Oct 21, 2011 at 11:14:02AM +0200, Jonas Geiregat wrote:
 Hello,
 
 I'm aware the subject may sound confusing. 
 So let me elaborate the subject some more.
 
 I have a computer currently running Ubuntu linux. I'm quite fed up with 
 this distrobution and would like to switch to Debian GNU/Linux.

Ubuntu is not so bad.  Which Debian you are thinking? Ubuntu LTS and
Debian stable tend to be slightly older but very stable.  Anytime when
you chase the latest desktop, you may hit little problems which ever
distribution you use.  Anyway, welcome to Debian.
 
 This computer has no cdrom-drive also, for some reason (probably
 something went wrong with my motherboard) I can't boot directely from an
 external cdrom-drive or a usb pendrive.

That is odd.  Have you checked BIOS setting.
 
 But if the computer starts I do get a grub menu from the Ubuntu
 distrobution. Would it be possible to use this grub setup from Ubuntu to
 boot my pendrive usb ?

Theoretically YES by pressing e in the GRUB menu and point it to USB
image properly arranged.  But this is too much hussle even for me.

If I were you, I install Debian to USB disk manually and swap hard disk
with the one in PC. (tricks with dbootstrap etc...)

But the easiest thing to do is fix BIOS.  If BIOS is right and problem
is due to half broken MB, you may wish to replace it before it breaks
down badly.

Osamu


-- 
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/20111021131840.ga11...@goofy.lan

Re: Cannot login to GNOME after sid update

[Osamu Aoki]

Hi,

On Tue, Oct 18, 2011 at 05:26:05PM +0900, Miles Bader wrote:
 Frank debianl...@videotron.ca writes:
  Another update came through laterabout 8 new packages and now at
  least the desktop loads. Half the stuff is still missing. We'll see what
  happens in the days ahead.  I rarely use Gnome anyway. But now I am
  curious.
 
 My main complaint is various settings seem to now be ignored...
 (e.g., fonts are different, my display background is the default, not
 what I set, etc).

That seems to be things to expect for such major upgrade: gnome2 - gnome3.

It is practically impossible to take care a non-regular transition of
such minor details.
 
 Dunno if this is a bug with the current debian state of things, or a
 general problem with gnome3.  Downgrading everything to gnome2 seems
 to fix the problems tho.

If you wish to move to gnome3 while on tesing/unstable system, you need
to deal such transition yourself for each account on the system as I
understand.  If you do not like to deal with such complication, please
keep using Debian stable.

Of course, if you can come up with bug-free robust transition script from
gnome2 to gnome3, please file wishlist bug with your proposed script.  I
am sure that the maintainer will include it as an example script or so.

Regards,

Osamu


-- 
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/20111018152339.ga20...@goofy.lan

Re: docbook2html error

[Osamu Aoki]

Hi,

On Tue, Oct 18, 2011 at 03:34:32PM +0530, Chandrabhanu Mahapatra wrote:
 After I installed docbook-utils sudo apt-get docbook-utils running
 docbook2html  gives error

I just installed docbook-utils to my Debian.

 jw: There is no frontend called
 /usr/local/share/sgml/docbook/utils-0.6.14/frontends/docbook.

AS others said, local is strange.  Did you install tar source by
yourself?  

$ dpkg -S docbook2html
docbook-utils: /usr/share/man/man1/docbook2html.1.gz
docbook-utils: /usr/bin/docbook2html
$ head  /usr/bin/docbook2html
#! /bin/sh
jw -f docbook -b html $@

Do you get the same result?
 
 there is no folder utils-0.6.14 in  /usr/local/share/sgml/docbook/

Who cares if you installed Debian package.

 How can I install it properly in Ubuntu 10.10.

What? Ubuntu??? You should be able to get ported Debian packages there.  If not
ported, do it yourself.  You should ask Ubuntu but here is a hint:

  
http://www.debian.org/doc/manuals/debian-reference/ch02.en.html#_porting_a_package_to_the_stable_system

Of course, you should read all the other packaging section ...

Osamu


-- 
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/20111018153802.gb20...@goofy.lan

Re: SCIM not working after upgrade

[Osamu Aoki]

Hi,

On Tue, Oct 04, 2011 at 06:59:25AM +0530, Sridhar M.A. wrote:
 Hi,
 
 Presently running debian testing. The latest xserver-xorg-core update
 did cause certain problems with my nvidia card. I could find a
 solution for starting X with the new nvidia binary driver (275.28). 
 
 Strangely (or maybe not), scim has stopped working. I cannot change to a
 different language. Tried im-switch to reset the default. No change.
 Clicking on the keyboard icon, does not bring up the language list as it
 used to earlier. Right clicking brings up the menu, though.
 
 An identical setup on my laptop with intel graphics has no isses with
 the new xserver-xorg-core and scim.
 
 Any pointers to get the scim working again?

http://bugs.debian.org/cgi-bin/pkgreport.cgi?repeatmerged=nosrc=scim
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=641140
(I think dropping scim for wheezy may be an option these days.
 No activie upstream and there are better alternatives: ibus)

If you are looking for working input method, please install ibus.

Please make sure to manually pull in ibus-gtk3 and ibus-qt4 to make it
work with recent desktop.

OH... please switch from im-switch to im-config.  (Yes, I am upstream of
both.)

Osamu


-- 
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/20111004123135.gb31...@goofy.lan

Re: command to mv files folders to dir

[Osamu Aoki]

Hi,

On Sat, Sep 24, 2011 at 05:31:47PM +0700, Ivan Shmakov wrote:
  Camaleón  noela...@gmail.com writes:
  On Sat, 24 Sep 2011 11:40:48 +0200, Mark Panen wrote:
 
   Made a bit off a muck up off things when i backed up parts of my
   /home/mark directory to /mnt/deer
 
   In /mnt/deer i know have hundreds of files and folders which i
   rsynced on 22/09/2011.
 
   I need a command line option to put them all In one shot in
   /mnt/deer/zebra.
 
   Not a command line but Midnight Commander is very good for such
   day-to- day tasks.
 
   Following the suggestion of Victor Wagner (in
   news:fido7.ru.unix.linux, I believe), I've dropped
   Midnight Commander in favor of Bash something like a decade ago.
 
   I've never regret the change.
 
   Midnight Commander may simplify the simple things, but
   whatever's your experience with it, the complex tasks are often
   impossible.  

??? It is possible but repeating it is a bit of hussle.

 On the contrary, Shell lets one to benefit from the
   experience, and the solutions to the simpler tasks could usually
   be re-used to solve the more complex ones.

But task described does not seem to be much of repeat action. MC can do
this by:
 * Open old and new directory in each panel.
 * Mark all directory then unmark new directory in the old directory panel
 * press F6
== DONE!

   Not to mention that the use of Shell keeps one's mind “in
   shape.”

Hmmm...

MC is good for interactive complex task while shell command with find
command is good for repeated complex task.  Many shell scripts tends
to become very complex which requires too much time for debugging.

When there is some exclusion rule etc., most commands with rsync etc.
becomes nealy impossible to debug for people like me.

Anyway, here are typical tricks used to copy files.

  
http://www.debian.org/doc/manuals/debian-reference/ch10.en.html#_idioms_for_the_copy

If you combine this with find as described in the following link, it can
really do any complicated selection and copy.

  
http://www.debian.org/doc/manuals/debian-reference/ch10.en.html#_idioms_for_the_selection_of_files

This is good if you are making backup script for regular backups.  It
can be told to pick file by date or size  Or you can create target
directory name automatically by date.  That is when you need such shell
script.

(I would not bother doing this just for simple task like what you
described.  If you configure MC to display hidden files, it is very
powerful.)

Regards,

Osamu


--
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/20110924171831.gb21...@goofy.lan

Re: sharing one r/w unix filesystem between different machines and users

[Osamu Aoki]

On Fri, Aug 26, 2011 at 12:21:18AM +0200, Christoph Groth wrote:
 Hi,
 
 I'd like to share the data saved on an external USB drive between
 different (GNU/Linux) machines, each having different users.  Each user
 should be able to mount the drive and read and write any files as he or
 she pleases.  The users aren't necessary root themselves.
 
 Is there a way to implement such a scheme with a non-windows filesystem
 like ext3?
 
 I understand how Unix file permissions work.  However, for a removable
 drive which might be connected to different systems (with completely
 unrelated uids/gids), assigning fixed uids/gids to files just doesn't
 make any sense.

This is untested ...

Use:
 user = nobody 65534
 group = users 100 or nogroup 65534
 
Then use BSD type file permission schemr using set GID trick to the
mount point directory as root only when you start using it.  

$ sudo chmod 5775 /mount/point/

Let each system mount it automatically.
 
Does not this work for you?


 What's the best FS for sharing data between unrelated Linux systems?  Is
 it really FAT or NTFS?
 
 Thanks,
 Christoph
 
 
 -- 
 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/87hb55ku8x@falma.de


-- 
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/20110905152202.ga15...@debian.org

Re: MTU and Postfix

[Osamu Aoki]

HI,

On Sun, Sep 04, 2011 at 10:40:29AM +, Camaleón wrote:
 On Wed, 31 Aug 2011 16:09:23 +, Camaleón wrote:
 
 (...)
 
  I'm still monitoring this but if this is the cure to prevent such
  errors, are there any expected drawbacks for lowering MTU system-wide?
 
 (...)
 
 Mmm, no replies yet...
 
 Does it mean then that there are no gotchas to care about in setting a 
 MTU value of 1400 for a bonded interface? :-)
 
 Greetings,

Some router between you and outside have issue, I guess.

It is a bit complicated.  My best effort information is here:

http://www.debian.org/doc/manuals/debian-reference/ch05.en.html#_finding_optimal_mtu

Osamu


--
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/20110904115651.ga8...@debian.org