Re: ssh session times out annoyingly fast, why?

[Britton Kerin]

On Wed, Sep 23, 2020 at 11:49 AM Greg Wooledge  wrote:
>
> On Wed, Sep 23, 2020 at 10:30:15AM +0300, Anssi Saari wrote:
> > Britton Kerin  writes:
> > > I'm using ssh from a debian box to a rasberry pi (sorta debian also :).
> > >
> > > For some reason ssh sessions seem to time out pretty quickly.
>
> How quickly, exactly?  What is the actual message/behavior you see when
> it happens?  Are they both on the same LAN, or is there some complexity
> in between them (especially a NAT router)?
>
> > Well, the keepalives themselves can cause a disconnect if the keepalive
> > messages are not reaching the other end due to bad connection for
> > example. Looks like by default in Debian client sends keepalives if
> > server is quiet but server doesn't send keepalives to a client.
>
> The normal reason people need to use ServerAlive or ClientAlive is NAT.
> If your connection from ssh client to ssh server goes through a NAT
> router, the router may keep track of activity on that connection, and
> drop the translation when it goes idle for 5 minutes or so.  Forcing the
> *Alive packets to happen every few minutes prevents a NAT timeout.
>
> If there is no NAT involved, then I agree with the previous suggestion
> that this might be a shell's TMOUT variable.  Are you sitting at a shell
> prompt when the "timeout" occurs?  Does the timeout stop occurring when
> you're inside a text editor, for example?

Looks like NAT was the culprit, because top kept it alive.  Internet has bogus
advice on this one because it suggests ServerAliveInterval 1200 or something
which I guess is larger than most firewall timeout.

Thanks for all help good to see debian community still so good.

Britton

ssh session times out annoyingly fast, why?

[Britton Kerin]

I'm using ssh from a debian box to a rasberry pi (sorta debian also :).

For some reason ssh sessions seem to time out pretty quickly.  I've
tried setting ClientAliveInterval and ClientAliveCountMax and also
ServerAliveInterval  and ServerAliveCountMax, but it doesn't seem to
make any difference.  Is there some other setting somewhere that
affects this?

Thanks,
Britton

looking for a replacement for debian since systemd

[Britton Kerin]

I see from below vote that we're working on dumping other init systems
now as expected.  Luckily I've given up on debian since systemd in the
first place and am in long process of finding a replacement.

Britton

> - - -=-=-=-=-=- Don't Delete Anything Between These Lines =-=-=-=-=-=-=-=-
> 7b77e0f2-4ff9-4adb-85e4-af249191f27a
> [ 3 ] Choice 1: F: Focus on systemd
> [ 1 ] Choice 2: B: Systemd but we support exploring alternatives
> [   ] Choice 3: A: Support for multiple init systems is Important
> [   ] Choice 4: D: Support non-systemd systems, without blocking progress
> [ 2 ] Choice 5: H: Support portability, without blocking progress
> [   ] Choice 6: E: Support for multiple init systems is Required
> [   ] Choice 7: G: Support portability and multiple implementations
> [ 4 ] Choice 8: Further Discussion
> - - -=-=-=-=-=- Don't Delete Anything Between These Lines =-=-=-=-=-=-=-=-

Re: make ping executable by normal users?

[Britton Kerin]

On Thu, Jun 2, 2016 at 2:33 PM, Santiago Vila <sanv...@unex.es> wrote:
> On Thu, Jun 02, 2016 at 01:56:08PM -0800, Britton Kerin wrote:
>> On my old debian system I could ping as a normal user.  The ping
>> binary had the suid bit set.  Now I get:
>>
>> $ ping www.google.com
>> ping: icmp open socket: Operation not permitted
>> 2 $
>>
>> presumably because the bit isn't set.
>>
>> What's the right fix?  I could setuid it but then if I understand
>> correctly it might get changed back by an upgrade.  Does it use
>> capabilites or something?
>
> Yes, it uses capabilities. The simple fix is to do this:
>
> dpkg-reconfigure iputils-ping

Well, that works, thanks.  But I really don't get the overall behavior.
It says this:

 root@debian:/home/bkerin# dpkg-reconfigure iputils-ping
 Setcap worked! Ping(6) is not suid!
 root@debian:/home/bkerin#

And then ping works for non-root users.

How, just by executing dpkg-reconfigure, did I tell it this is what
I wanted?  If that's the default, why wasn't it that way to begin with?

More generally, is it somehow possible to still run debian without
capabilities?  I hate them.  The simple root-or-not security model
is much simpler and doesn't promise more than it can really
deliver.  I'm sad to see capabilities now as the default.

Britton

make ping executable by normal users?

[Britton Kerin]

On my old debian system I could ping as a normal user.  The ping
binary had the suid bit set.  Now I get:

$ ping www.google.com
ping: icmp open socket: Operation not permitted
2 $

presumably because the bit isn't set.

What's the right fix?  I could setuid it but then if I understand
correctly it might get changed back by an upgrade.  Does it use
capabilites or something?

Thanks,
Britton

Re: wireless without network-manager... is it still possible?

[Britton Kerin]

On Thu, May 26, 2016 at 11:29 AM, Brian <a...@cityscape.co.uk> wrote:
> On Wed 25 May 2016 at 21:21:14 -0800, Britton Kerin wrote:
>
>> On Mon, May 23, 2016 at 9:32 AM, Brian <a...@cityscape.co.uk> wrote:
>> >
>> > The OP deposited his FUD in -user (twice), completely ignored the
>> > invitation to explore a technical solution using ifupdown and has now
>>
>> Regardless of ifupdown, I want network-manager to work
>> like it's supposed to, i.e. the same way in a dbus-launch-spawned
>> session as under gnome.  It doesn't.  That a bug, not FUD.
>
>   https://lists.debian.org/debian-user/2016/05/msg00939.html
>
> You might want to pursue this response, either on -user or in your own
> time; it appears very useful. It is not something I want to do but looks
> to be right up your street. The poster doesn't see any bug there.
>
> For myself, I'll have to remember that subject titles might bear no
> resemblance to the problem in search of a solution. ifupdown is used
> throughout the North American continent (Canada, Mexico, the U.S.A etc),
> and, excepting hardware problems, it and the supplicant do a good job.
>
> Life is too short to worry about whether network-manager delivers the
> goods. Just use ifupdown and wpa_supplicant and you will live happily
> ever after.

Thanks for this response.  Modulo some issues with rfkill it's the correct
approach.  I'm sorry for the insults in my original post and for being so
combative with you.

Regarding network-manager, I think you're entirely right.  It was flaky
10 years ago, it still seems flaky, time to just avoid it forever.

Britton

Re: wireless without network-manager... is it still possible?

[Britton Kerin]

On Mon, May 23, 2016 at 9:32 AM, Brian  wrote:
> On Mon 23 May 2016 at 09:07:36 +0200, to...@tuxteam.de wrote:
>
>> On Sun, May 22, 2016 at 11:38:39PM +0100, Lisi Reisz wrote:
>> > On Sunday 22 May 2016 22:56:36 Richard Owlett wrote:
>> > >
>> > > So what ever B says goes?
>> > > Please *NO* !
>> >
>> > Richard - this snipping is very misleading.  It was not Tomas whom Brian 
>> > was
>> > calling incompetent - in fact, quite the opposite.  I can't quite see why
>> > Tomas has taken such fright, except that both the others were putting 
>> > things
>> > rather strongly.  But *nothing* that Brian said could reasonably have been
>> > interpreted, as you appear to have done, as Brian telling Tomas to leave 
>> > the
>> > thread.
>>
>> Exactly, Lisi. I'm too old to watch people slinging mud at each other
>> ("asshole", "incompetent") like kids. I mean, I feel strongly about
>> software, that's why I'm here, and I feel strongly about free software
>> and its social value, but my time is just too precious for that nonsense.
>>
>> So you got it right.
>
> The OP deposited his FUD in -user (twice), completely ignored the
> invitation to explore a technical solution using ifupdown and has now

Regardless of ifupdown, I want network-manager to work
like it's supposed to, i.e. the same way in a dbus-launch-spawned
session as under gnome.  It doesn't.  That a bug, not FUD.

> taken himself off to -devel, which is not renowned for suffering the
> spreading of misinformation.

There's agreement on devel that it's not FUD.  I posted a detailed
description of the problem there.  Go read it and send me something
useful instead of more groundless claims that bug reports on
network-manager are FUD.

> One hopes his transition to there was not
> due to anything *I* said and he gets a glimmer of clue, :)

It was partly due to you, I'm hoping for more substantive responses there.

Britton

Re: wireless without network-manager... is it still possible?

[Britton Kerin]

On Sun, May 22, 2016 at 1:56 PM, Richard Owlett  wrote:
> On 5/22/2016 3:23 PM, to...@tuxteam.de wrote:
>>
>> -BEGIN PGP SIGNED MESSAGE-
>> Hash: SHA1
>>
>> On Sun, May 22, 2016 at 08:35:22PM +0100, Brian wrote:
>>>
>>> [...] it merely indicates your incompetence.
>>
>>
>> Folks, I'm out of this thread.
>>
>
> So what ever B says goes?
> Please *NO* !

Not a problem, since he didn't actually say anything worth responding to.

Britton

Re: wireless without network-manager... is it still possible?

[Britton Kerin]

On Sat, May 21, 2016 at 10:44 PM,  <to...@tuxteam.de> wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> On Sat, May 21, 2016 at 09:02:15PM -0800, Britton Kerin wrote:
>> somehow network-manager makes it work.  But I've had it with gnome, and none
>> of the command line tools or references I've found work.  That
>> includes /etc/network/interfaces,
>> direct use of ifconfig,iw,ip,rfkill,wpa_supplicant,dhclient, and
>> wicd-client (though it's not really what I want.
>
> I have wireless working (right now) without Network Manager, via
> ifupdown, on jessie/sid.
>
>> Is this even still possible or have the systemd assholes decided we
>> shouldn't be doing it?
>
> Now, now. I don't like systemd myself (and manage to avoid it, I'm
> still using SysV init), but treating free software developers as

Nice to hear that's still possible, the general tone I've seen so far as
I look at this is that its a huge pain even on gentoo and almost
impossible elsewhere.

> "assholes" seems highly inappropriate.

Well it's the customary term for people with the attitude that they constantly
and deliberately exhibit.

>> I believe the problem with the traditional wpa_supplicatn approach is
>> related to syslog entries like
>> these:
>>
>> May 21 17:00:06 debian kernel: iwlwifi :02:00.0: can't access the
>> RSA semaphore it is write protected
>
> Sorry I can't help you with that specific one. But search engines give
> some hits which look relevant, like e.g.
>
>   <https://bbs.archlinux.org/viewtopic.php?id=210306>

I saw that, but if it needs a new firmware, it seems that network-manager
is pulling it in somehow, so it looks like the problem is that the old
interfaces don't and no one has noticed.

>> But as I say somehow network-manager makes it work.
>>
>> I just got a new laptop after 10 years, I didn't realize how bad the
>> situation has become with systemd udev gnome etc.
>
> Then help others keeping the alternatives up and running. No clients?
> No product!

I'm trying.  I'll report when it works.  Trying to run nmcli fails from other
window managers with dbus errors, which it absolutely  shouldn't do since
network-manager is running.  Same with nm-connection-editor and wicd-client,
they are all apparently dependent on dbus now, and dbus doesn't appear to
work correctly without a gnome session going.

So I need gnome running to talk to my network card.  How fcking ridiculous.  It
should be a kernel function but it looks like the only tested configuration
involves all of this wobbly stack of garbage.  Isn't this exactly what we were
promised wasn't going to happen when debian went with systemd?

Britton

wireless without network-manager... is it still possible?

[Britton Kerin]

somehow network-manager makes it work.  But I've had it with gnome, and none
of the command line tools or references I've found work.  That
includes /etc/network/interfaces,
direct use of ifconfig,iw,ip,rfkill,wpa_supplicant,dhclient, and
wicd-client (though it's not really what I want.

Is this even still possible or have the systemd assholes decided we
shouldn't be doing it?

I believe the problem with the traditional wpa_supplicatn approach is
related to syslog entries like
these:

May 21 17:00:06 debian kernel: iwlwifi :02:00.0: can't access the
RSA semaphore it is write protected

But as I say somehow network-manager makes it work.

I just got a new laptop after 10 years, I didn't realize how bad the
situation has become with systemd udev gnome etc.

Britton

easiest way to shut down all network services besides ssh?

[Britton Kerin]

I have a system that I would like to make accessible only by ssh.

No apache telnet ftp anything else.

What is the easiest way to achieve this?  It came from a vendor with
a slew of package of all sorts, so I don't even know everything that
I want to remove.

Thanks,
Britton


-- 
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/cac4o8c_s_bqvye980fj2ok+qdeeinkmcp3u0aqcw4hjsri5...@mail.gmail.com

how to make gnome SHUT DOWN when I say SHUT DOWN

[Britton Kerin]

Sometimes firefox doesn't really exit (despite all its windows being closed)
so when I say shutdown gnome pops up this dialog asking if I want to
shutdown despite a running process.  Then my laptop gets put in its bag
and tries to cook itself to death.

Is there a way to explain to gnome that when I say shutdown, I mean
SHUTDOWN NOW, REGARDLESS OF STUPID BROKEN PROCESSES?

Thanks,
Britton


-- 
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/CAC4O8c_-aROgbWT_rFL=XcSyAyyJ5ey0U41O+mkO_A=fosf...@mail.gmail.com

debian install wireless fails on laptop with Ralink RT3290 wireless, what to do?

[Britton Kerin]

The installer correctly detects a Ralink RT3290 and seems to think its
going to be
able to work with it, but then it always fails (I think at the DHCP
stage) to actually
manage to connect to the network.

Has anyone else encountered this issue?  Workarounds?

As usually everyone assumes it would work on Ubuntu and as usual I'd rather not
go there.

What I'm hoping to do is install a gnome system from CD image, then follow these
instructions:

  https://wiki.debian.org/rt3290

Is this probably workable?  Is it the best way?  If there was some way to get
wireless working at the point in the installation where the installer
tries to set
it up that seem like it would be best, but maybe it isn't required then if you
have one of the big (non-netinst) CDs?

Thanks,
Britton


-- 
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/CAC4O8c9OgobJZoJj28DBtR+CQaSsHX6KU=XR=koio_4pida...@mail.gmail.com

a browser for debian that will play pandora.com?

[Britton Kerin]

iceweasel doesn't seem to, I downloaded firefox and ran it but
it doesn't seem to either (just hangs forever).

Now my GF is saying just use Ubuntu blah blah is there any
non-horrible way to get a browser that will play internet radio?

Thanks,
Britton


-- 
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/cac4o8c8c_5qpnazqhtiu6gvuophxqcrgf15madrrzepvcd3...@mail.gmail.com

which package has a binary like this sendmail one in exim?

[Britton Kerin]

Hi folks,

I know exim sometimes contains a sendmail binary because on one system I
get this:

  britt...@brittonkerin.com [~]# sendmail --version
  Exim version 4.76 #1 built 26-Oct-2012 16:41:54
  Copyright (c) University of Cambridge, 1995 - 2007
  Berkeley DB: Berkeley DB 4.7.25: (April  4, 2012)
  Support for: crypteq iconv() IPv6 PAM Perl OpenSSL Content_Scanning
DKIM Old_Demime Experimental_SPF Experimental_SRS
  Lookups (built-in): lsearch wildlsearch nwildlsearch iplsearch dbm
dbmnz passwd
  Authenticators: cram_md5 dovecot plaintext spa
  Routers: accept dnslookup ipliteral manualroute queryprogram redirect
  Transports: appendfile/maildir autoreply pipe smtp
  Fixed never_users: 0
  Size of off_t: 8
  2012-11-28 11:17:54 non-existent configuration file(s): /etc/bh/exim/exim.conf

But on my debian box (current stable distribution), there is no
/usr/bin/sendmail binary even though I have exim installed.  There is a
/usr/sbin/sendmail binary, but it doesn't seem to be the same one because:

  $ /usr/sbin/sendmail --version
  exim abandoned: unknown, malformed, or incomplete option --version

I also found /usr/lib/sendmail but it does the same thing.

Does anyone know what debian package provides the /usr/bin/sendmail program
or its equivalent like on the first system described?


-- 
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/cac4o8c84zsxapm5jy6babwcsy5jdnjzcae7axcjky3h1qnk...@mail.gmail.com

manually reset/load network nameser configuration only?

[Britton Kerin]

I have some obnoxious problem with my laptop, such that
the NerworkManager Applet sometimes gets me on wireless
or wired network ok, but somehow the nameserver doesn't
start working right, despite working fine for other computers
on the same network.

I'm wondering if there is some way to specify the nameserver
to user and reset that part of the network infrastructure?

In effect override whatever strange problem NetWork manager
sometimes has?

Thanks,
Britton


-- 
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/cac4o8c8zf+psphpmfyqxc_1yaroqxxhtkejxzxgujtsn_ha...@mail.gmail.com

looking for the 'lockfile' program

[Britton Kerin]

I'm trying to determine if the 'lockfile' program is still available
or has been obsoleted for some reason.

I'm not looking for lockfile_create or friends, nor flock, but the
program described here:

 http://linux.about.com/library/cmd/blcmdl1_lockfile.htm

Thanks,
Britton


-- 
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/CAC4O8c-gw_2A+uEq0U=x8doadqs7hmpbkata93jhr7qkskj...@mail.gmail.com

Re: looking for the 'lockfile' program

[Britton Kerin]

Ok, its part of procmail apparently.

On Tue, Oct 2, 2012 at 4:29 PM, Britton Kerin britton.ke...@gmail.com wrote:
 I'm trying to determine if the 'lockfile' program is still available
 or has been obsoleted for some reason.

 I'm not looking for lockfile_create or friends, nor flock, but the
 program described here:

  http://linux.about.com/library/cmd/blcmdl1_lockfile.htm

 Thanks,
 Britton


-- 
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/CAC4O8c8koA1Zj_r75MWw7i9YhB=8ucugomromght1qe+43m...@mail.gmail.com

difficulty downloading packages, is ftp.us.debian.org having a problem?

[Britton Kerin]

Hi everyone,

I keep trying to download a big pile of packages and most of them keep failing.
Internet is working and I can ping things, but most of the packages always fail.

I'm wondering if ftp.us.debian.org is really overloaded or something?

Thanks,
Britton


-- 
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/cac4o8c9ponxxles_rp8uzjggpjtsrnwaxyzxs738uwgupyk...@mail.gmail.com

how to do something automatically at boot?

[Britton Kerin]

If I understand right you make a script like this one:

  rhino:/etc/init.d# cat /etc/init.d/weird-at
  #!/bin/sh

  set -e

  touch /tmp/g
  echo 'touch /tmp/greeber' | at now + 3 minutes
  touch /tmp/fgg

  exit 0
  rhino:/etc/init.d#

  rhino:/etc/init.d# ls -l weird-at
  -rwxr-xr-x 1 root root 110 2011-12-28 23:31 weird-at

Then make a link to it like this link:

  rhino:~# ls -l /etc/rc5.d/S98weird-at
  lrwxrwxrwx 1 root root 18 2011-12-28 12:48 /etc/rc5.d/S98weird-at -
../init.d/weird-at
  rhino:~#

But this doesn't seem to work (/tmp/gooo and /tmp/fgg never show up,
and the at job isn't queued.  The script works fine  when run from the
command line.

Is there something I'm missing in this process?  Is there somewhere particular
to look for diagnostic output for things that are supposed to be done on boot?

Thanks,
Britton


-- 
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/CAC4O8c_8q_3mNUhcS2CNZiTivyOJ5fF1+g9OwpuRmw-SQ+aQ=w...@mail.gmail.com

nvidia xserver makes fonts big and icky

[Britton Kerin]

The nvidia packages in the latest debian work lovely, except for
one thing: all my fonts come out bigger.  The gnome stuff still
things that my screen is at the same resolution, and the fonts
the same, but they are all bigger so the editor, terminals etc.
are all nasty.

Any ideas how to fix?

Thanks,
Britton



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

Re: permissions problems when using libusb

[Britton Kerin]

Unfortunately someone on #debian on IRC said that the instructions in
that file are incorrect when then udev-created device is involved.  Or
something like that.  Sorry, I should have mentioned that originally.

Britton

On Mon, 10 Sep 2007 17:45:58 -0500, Michael Shuler
[EMAIL PROTECTED] said:
 On 09/10/2007 05:39 PM, Britton Kerin wrote:
  I have a program that uses libusb and it only works from root.  When I
  try
  to run as a normal user I get errors like this:
  
  avrdude: usb_open(): cannot read serial number error sending control
  message: Operation not permitted
  avrdude: usb_open(): cannot read product name error sending control
  message: Operation not permitted
  avrdude: usbdev_open(): error setting configuration 1: could not set
  config 1: Operation not permitted
  
  I didn't see a usb group in /etc/group or anything like that.  Can
  anyone
  tell me the preferred way to let users run programs that use libusb?
 
 /usr/share/doc/libusb-0.1-4/README.Debian has some mount options that
 might be helpful in your situation.
 
 -- 
 Kind Regards,
 Michael Shuler
 
 
 -- 
 To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
 with a subject of unsubscribe. Trouble? Contact
 [EMAIL PROTECTED]
 



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

permissions problems when using libusb

[Britton Kerin]

I have a program that uses libusb and it only works from root.  When I
try
to run as a normal user I get errors like this:

avrdude: usb_open(): cannot read serial number error sending control
message: Operation not permitted
avrdude: usb_open(): cannot read product name error sending control
message: Operation not permitted
avrdude: usbdev_open(): error setting configuration 1: could not set
config 1: Operation not permitted

I didn't see a usb group in /etc/group or anything like that.  Can
anyone
tell me the preferred way to let users run programs that use libusb?

Thanks,
Britton Kerin



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

getting debian fully working on System76 computers

[Britton Kerin]

On Thu, 30 Aug 2007 22:59:16 -0400, Michael Pobega [EMAIL PROTECTED]
said:
 -BEGIN PGP SIGNED MESSAGE-
 Hash: SHA1
 
 On Thu, Aug 30, 2007 at 11:07:08AM -0800, Britton Kerin wrote:
  
  On Thu, 30 Aug 2007 15:00:55 -0400, Michael Pobega [EMAIL PROTECTED]
  said:
   -BEGIN PGP SIGNED MESSAGE-

 I was just wondering because I have a System76 as well :D
 
 Debian seems to run pretty well on it, I have to say; The only problem
 is that the System76 driver installer is built not to run on anything
 but Ubuntu, so you have to hunt down and install all of the modules
 yourself. If you need any help getting things working give me an e-mail,
 anytime.

Ok.  I haven't started the hunt yet but I know its coming, since the
wireless card doesn't seem to be working automagicly.  Also, my GF
(whose computer this is) would like to be able to use YouTube, Java
web sites, and DVD player out of the box and apparently at least
some of these things don't work with the standard desktop task selection
for the latest debian stable.  Have you by any chance tried these
things?
Is it just a matter of knowing which packages you need (I don't) or do
all these things take custom builds?

 P.S. : You do know you replied directly to my inbox and not to the list,
 right?

No, I made a mistake there.  Thanks for telling me.

Thanks,
Britton

 
 - -- 
 If programmers deserve to be rewarded for creating innovative
 programs, by the same token they deserve to be punished if they
 restrict the use of these programs. 
  - Richard Stallman
 -BEGIN PGP SIGNATURE-
 Version: GnuPG v1.4.6 (GNU/Linux)
 
 iD8DBQFG14QEg6qL2BGnx4QRAhQiAKCGqrOScYJG2IdR57erHPv+DJZQvACgoqzz
 Xiwgq0Y311V4EBkTsZKh1cA=
 =8Rd8
 -END PGP SIGNATURE-



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

upgrading ubuntu to debian

[Britton Kerin]

I just bought a computer that came with ubuntu and would like to switch
it to pure debian.  Is there a standard way to do this that someone 
could point me to?

(Though I will say that little hack where the shell tells you which 
 package a program is in looks pretty cute and helpful :)

Thanks,
Britton



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

disableing gnome destop background?

[Britton Kerin]

I would like to be able to display pictures for my desktop background,
and change them every so often.  It seems that gnome doesn't do this,
so I though I'd just do it from a script with xsetbg, but I think for
this to work I need to somehow tell gnome to not do anything to the
X root window.  I couldn't figure out how to do this, can anyone tell
me?

Thanks,
Britton



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

ready to use debian or ubuntu laptop reccomendations?

[Britton Kerin]

I would like to buy my non-linux girlfriend a ready to 
use laptop with:
  
   wireless scan, falling back to CAT5 DHCP
   open office
   CD ROM that automounts
   SD card reader that automounts
   working sound card
   reasonable memory and disk
   working video acceleration (at least a bit)

The last item is nonessential, needed only to make
X more comfortable, not for games or anything.

I can spend about $1000, though it seems like they
should be available for less given the windows crud
that is out there and the supposed high cost of
Windows XP.

Is it possible to get such a system?  I tried with
linuxcertified but the system she got was broken in
a variety of ways.  I'd really appreciate it if
anyone has any reccomentations of a model that they
know from personnal experience will work.

Thanks,
Britton



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

perl based web site in a can type reccomendations?

[Britton Kerin]

I would like to be able to put together something like
what is seen on the linksys routers, i.e. tabs for 
configuring a system, popping up help, etc. without
having to monkey around with the details of html and
CSS any more than strictly necessary.  Perl is the lang
I know so I like it over php or others.  Anyone have
a recommendation?

Thanks,
Britton
-- 
  Britton Kerin
  [EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

cordless mice with USB connected base units

[Britton Kerin]

Do cordless non-bluetooth mice need weird drivers?
Can I take a USB only cordless mouse with a base
charger unit, plug it into some kind of USB_to_ps2
adaptor plug, plug that into the connector where 
my PS2 mouse currently lives, and just be happy?

I'm thinking about buying a cordless mouse, but I'm
a bit scared of it.  If it needs weird drivers its
probably not worth the hassle.

Thanks,
Britton
-- 
  Britton Kerin
  [EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Re: two head x server

[Britton Kerin]

On Thu, 19 Jan 2006 18:18:17 +0100, igor [EMAIL PROTECTED] said:
 I have an idea of setting up X server to use two video cards? Can anyone
 give me first direction hints about this? thanks.

It should be noted that another alternative is to use one of the newer
video chipsets from nvidia that supports dual monitors from a single
card.  I've generally had good luck with a GV-N52128, which should be
pretty cheap by now assuming you can find it, 6600, 6800, and 7800
series all have the same functionality.

Downside of course is you are stuck with the nonfree nvidia drivers,
glx library, etc., and X config is a bit different than you may 
read about in some places because they seem to have kind of done 
their own thing in terms of how dual head is described to the 
server using X options.  But nvidia seems to actually do a decent
job keeping up to date binary drivers available for linux, and the
install package they have works, and comes with a utility
(nvidia-xconfig
I think it is) to help you generate a config file, which you can use
or merge with your own, or I can send you mine.  One hates to have to
wonder what bits of the system its sticking things in, but, well, it
works.
For me.  As long as I don't upgrade my kernel too much.  *sigh*

Britton

 
 -- 
 igor [EMAIL PROTECTED]
 pletisan
 
 
 -- 
 To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
 with a subject of unsubscribe. Trouble? Contact
 [EMAIL PROTECTED]
 
-- 
  Britton Kerin
  [EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

pthread function in testing print weird messages to stderr

[Britton Kerin]

This minimal program, test_threads.c:

   #include assert.h
   #include pthread.h
   #include stdio.h
   #include stdlib.h

   int thread_return;

   void *
   thread_function (void *arg)
   {
 int idx;
 for ( idx = 0 ; idx  3 ; idx++ ) {
   sleep (1);
   printf (thread_arg: %d\n, *((int *)arg));
 }

 thread_return = 42;

 pthread_exit (thread_return);
   }


   int
   main (void)
   {
 int thread_arg = 2;
 pthread_t thread;
   
 int return_code;

 //  pthread_attr_t thread_attrs;
 //  return_code = pthread_attr_init (thread_attrs);

 return_code = pthread_create (thread, NULL, thread_function, 
(void *) thread_arg);
 assert (return_code == 0);

 int **return_location;
 return_code = pthread_join (thread, (void **) return_location);
 assert (return_code == 0);
 
 printf (thread_return: %d\n, **return_location);

 return 0;
   }

Compiled this way:

   gcc -Wall -D_REENTRANT test_threads.c -lpthread -o test

without any errors or warnings, produces this output when run:

   thread_arg: 2
   thread_arg: 2
   thread_arg: 2
   thread_return: 42
 1581:
 1581: runtime linker statistics:
 1581:final number of relocations: 139
 1581: final number of relocations from cache: 7

with the latter four lines being on stderr.  Is there some debugging
information that didn't get turned off or something?  I guess this 
might be a libc bug report?

Uncommenting the lines that set up the thread_attrs structure and
passing a pointer to it to pthread_create insteal of NULL as the 
second argument yields a different weird output:

   thread_arg: 2
   thread_arg: 2
   thread_arg: 2
   thread_return: 42
  1598: binding file /lib/tls/libpthread.so.0 to
  /lib/tls/libc.so.6: normal symbol `__cxa_finalize' [GLIBC_2.1.3]
  1598: binding file /lib/libgcc_s.so.1 to /lib/tls/libc.so.6:
  normal symbol `__cxa_finalize' [GLIBC_2.1.3]

Any clues as to what is going on here would be greatly appreciated.

Thanks,
Britton
-- 
  Britton Kerin
  [EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Re: Synching deeply nested directories Debian Server - Win XP

[Britton Kerin]

On Thu, 8 Dec 2005 02:00:44 +0100, Debian Users
debian-user@lists.debian.org said:
 Now for the interesting thing: our network does not allow SMB access from 
 outside (its the universiy's policy, I cannot change that): SMB ports are 
 blocked. I still would like to synchronize the data on e.g. my laptop and 
 the files on the server once in a while, even if not inside the server's 
 network. Since I cannot easily convince XP to use other ports for SMB 
 sharing (thumbs down for MS), I have to find other ways. For that end I 
 tried unison via ssh (available on Windows and Debian), but had to give 
 up because of the long path name bug in unison (or probably in OCAML).

Im surprised that unison would die on long file names.  The unison
people
say they still bug fix support unison, have you reported the problems to 
them?

Britton
-- 
  Britton Kerin
  [EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

dynamic linking of lib that uses other libs how?

[Britton Kerin]

I am trying to figure out if its possible to somehow insulate the user
of a library I wrote from the linking needs of the libraries my library
uses.  Here is the situation in detail:

mylib uses OpenGL, gtk, and gtkglext.

I would like to provide something like the pkg-config that comes with
gtk and gtkglext,
so that users of mylib can easily retrieve linking flags mylib needs,
without disturbing users ability to link with other libraries of their
choice.

At the moment I am using pkg-config for mylib, and accumulating the
linker flags required by libglgtk into variables that can be queried by
pkg-config, as gtkglext
does, but this does not seem to work the way I would like.

For example, trying to link myprog against mylib and a different version
of gtk than the one used by mylib leads to crashes related to mixing of
gtk functions from different versions.

Basically, my question is: what is the best way to package a dynamicly
linked C library for client consumption without causing trouble for the
clients?  

Any advice or pointers to information greatly appreciated,
Britton



-- 
  Britton Kerin
  [EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Re: two problems

[Britton Kerin]

On Mon, 12 Apr 1999, Sidney Brooks wrote:
 
 I recently had to replace my hard disk which meant reloading
 everything. I now have two problems.
 1. Although I had no trouble with Word Perfect before, when I
 downloaded it again, I could not use it because it can't locate libX..
 . I had this trouble with netscape originally, but the problem was
 solved by using the glibc version. There is no glibc version for Word
 Perfect.

Sounds like you might need a depricated compatability library.  There are
lots of these that are X-related.

 2. As long as I had to reload Debian, I decided to enable LILO although
 I used a boot floppy before. When I boot with LILO, ppp doesn't work,
 and I get a message saying that the ppp module is not loaded. When I
 boot with my floppy, ppp works fine. I haven't been able to find a
 difference in files when I compare the results of loading with the
 floppy and LILO. Everything else seems to be the same whichever I boot.

Is ppp compiled into the kernel on your new disk?

Re: email threat

[Britton Kerin]

 bruce writes:
  Because I know that Eric is a firearms enthusiast, for my own protection,
  I feel the best strategy is for me to publicize the threat widely.
 
 And quotes:
  Damn straight I took it personally.  And if you ever again behave like
  that kind of disruptive asshole in public, insult me, and jeopardize
  the interests of our entire tribe, I'll take it just as personally --
  and I will find a way to make you regret it.  Watch your step.
 
 While this is certainly objectionable, it is not a threat of violence.
 Notifying the police is totally unjustified.

True.  And however objectionable it may be it was at least private.  Now
we have notified not only the entire debian list of the scraggly side of
two of our most prominent figures, but also the police as well?  What
happens when the press latches onto this one?  Will we end up being looked
at as a crazed bunch of high-tech Montana Freeman?  When the leader or
representative of a group is stigmatized as a dangerous gun nut this rubs
off on the rest of the group, so please be careful what you say.

And both you and Eric should know better than to assume that anything you
say will remain private for long.  You both have movie star status in
software circles: more people know your names than don't.  I suggest to
both of you that you refrain from writing *anything* as inflammatory as
either of the messages in your latest exchange, public or private.


Britton

OFF TOPIC: is this abuse of mmap()?

[Britton Kerin]

I have been trying to read() audio data into an mmap()ed region (from a file 
just created and empty before the mmap call.  I know that read() normally
puts the data in a buffer and not a file, but with mmap() you get back a 
caddr_t pointer which I am hoping read() can use.  Here is the relevant chunk 
of code (full listing at bottom):

  /* create and memory map the target region */
  printf(0: I'm still here sir!);
  status = target = open(targetfile, O_CREAT | O_RDWR, 777);
  if (status  0) 
perror(error opening target file);
  printf(1: I'm still here sir!);
  region = mmap(NULL, (off_t) 512, PROT_WRITE, MAP_SHARED, target, 0);
  printf(2: I'm still here sir!);
  if (region == (caddr_t) (-1))
perror(mmap error);

  /* record sound into mmaped region */
  printf(3: I'm still here sir!);
  status = read(audio, (void *) region, 512); 
  if (status != sizeof(region))
perror(read screwed one way or another);

  msync(region, 512, MS_SYNC);

with the read(audio, (void *) region, 512);
line not commented out, when I run this program I get:

Bus error

and a file in the current directory:

-rx--t   1 gandalf  gandalf 0 Feb  4 17:04 targetfile

None of the little I'm still here sir! message print with the read in there, 
so I assume I'm doing something drastic.  My questions:

1.  Why does the file have those strange permissions?  I thought I specified 
mode 777.

2.  Is it possible to use read() into an mmap() this way?  Is there some other 
function I should be using instead of read()?

3.  Even if it is possible, is it a completely hairbrained approach?

The full source of the short program in question:

/*
 * rawrec.c Program to test recording principle of multiple mmaps,
 * forked into the background before syncing.
 */

#include unistd.h
#include fcntl.h
#include sys/types.h
#include sys/ioctl.h
#include sys/mman.h
#include stdlib.h
#include stdio.h
#include linux/soundcard.h

#define LENGTH 600  /* how many seconds of the CD to store */
#define RATE 44100  /* the sampling rate */
#define SIZE 16 /* sample size: 8 or 16 bits */
#define CHANNELS 2  /* 1 = mono 2 = stereo */

int main()
{  
  int audio;/* sound device file descriptor */
  int arg;  /* argumrnt for ioctl calls */
  int status;   /* return status of system calls */
  int target;   /* file descriptor to write to */
  void * region;/* for the pointer to the mapped file */

  /* open sound device */
  audio = open(/dev/dsp, O_RDWR);
  if (audio  0) {
perror(open of /dev/dsp failed);
exit(1);
  }

  /* set sampling parameters */
  arg = SIZE;   /* sample size */
  status = ioctl(audio, SOUND_PCM_WRITE_BITS, arg);
  if (status == -1)
perror(SOUND_PCM_WRITE_BITS ioctl fialed);
  if (arg != SIZE)
perror(unable to set sample size);

  arg = CHANNELS; /* mono or stereo */
  status = ioctl(audio, SOUND_PCM_WRITE_CHANNELS, arg);
  if (status == -1)
perror(SOUND_PCM_WRITE_CHANNELS ioctl failed);
  if (arg != CHANNELS)
perror(unable to set number of channels);

  arg = RATE;   /* sampling rate */
  status = ioctl(audio, SOUND_PCM_WRITE_RATE, arg);
  if (status == -1)
perror(SOUND_PCM_WRITE_RATE ioctl failed);

  /* create and memory map the target region */
  printf(0: I'm still here sir!);
  status = target = open(targetfile, O_CREAT | O_RDWR, 777);
  if (status  0) 
perror(error opening target file);
  printf(1: I'm still here sir!);
  region = mmap(NULL, (off_t) 512, PROT_WRITE, MAP_SHARED, target, 0);
  printf(2: I'm still here sir!);
  if (region == (caddr_t) (-1))
perror(mmap error);

  /* record sound into mmaped region */
  printf(3: I'm still here sir!);
  status = read(audio, (void *) region, 512); 
  if (status != sizeof(region))
perror(read screwed one way or another);

  msync(region, 512, MS_SYNC);
}