Re: Source file bind9

[Mathieu Trudel-Lapierre]

On Sat, Apr 8, 2017 at 1:44 AM, Nils Kassube <kass...@gmx.net> wrote:

> Alexandre Vilarinho wrote:
> > I'm trying to get the source file for bind9,
>
> This is a sopport question. you should better ask on the ubuntu-users
> list <https://lists.ubuntu.com/mailman/listinfo/ubuntu-users>.
>


Actually, if this is about the source for a bind9 that was uploaded to the
archive; then it's quite on topic for ubuntu-devel-discuss@; it does
qualifies as "discussing the development of Ubuntu".

You have the 'pull-lp-source' command that allows you to get the right
source package for the release you'll want; to use it, you need to have
'ubuntu-dev-tools' installed:

pull-lp-source bind9 xenial

(assuming what you want is the source for the latest upload of bind9 to
Ubuntu 16.04)

Kindly,


Mathieu Trudel-Lapierre <cypher...@ubuntu.com <mathieu...@gmail.com>>
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/65B58DA1 818A D123 0992 275B 23C2  CF89 C67B B4D6 65B5 8DA1
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Fwd: Usb tethering

[Mathieu Trudel-Lapierre]

On Dec 19, 2015 08:40, "Anca Emanuel"  wrote:
>
> Please make usb tethering work 'out of the box'
> Tutorial here: https://youtu.be/K2QQHsC2jiA
> On Ubuntu Xenial:
> (from dmesg) rndis_host 1-5:1.0 enxfe28313119dc: renamed from usb0
> IPv6: ADDRCONF(NETDEV_UP): enxfe28313119dc: link is not ready
>
> dhcpd enxfe28313119dc
>

We ship isc-dhcp-client, which has dhclient, which is more likely to be
what you want for USB tethering. I assume what you are looking to do is
share the connection from your phone to your computer, which means the
phone should act as a dhcp server.

Seeing the messages you've shared, typically the connection should be
brought up automatically by NetworkManager as soon as the link is ready
(which probably only depends on your phone having signal and having USB
tethering enabled).

/ Matt
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: License issues with libblkid1

[Mathieu Trudel-Lapierre]

On Fri, Jul 10, 2015 at 10:49 AM, Mishari Alarfaj malar...@bluebelttech.com
 wrote:


 Hi.  I just wanted to mention that libblkid1 is currently distributed via
 ubuntu 14.04 as an LGPL libarary.  However, when viewing the source, a
 file: version.c is distributed as GPL.  I believe this either breaks the
 LGPL licensing, and the license either needs to be updated or the version.c
 file (which appears to be irrelevant) needs to be removed.  Thanks for your
 time


A quick look at the debian/copyright file shows that it is listed (though
not very clearly, I must admit) as having multiple different source files
with different licenses; GPL-2+ is listed [1]. This would include version.c
as under the GPL. (I don't know whether this is correct or not, if it
should be GPL-2+ or just GPL-2).

The debian/copyright file has been updated in later releases to more
clearly list the copyright for each source file. The first stanza is a
catch-all (*) which includes any file not explicitly listed later in the
file [2].

This make me feel like while util-linux might benefit a good look w.r.t
licensing compliance, it's likely already fine (that also depends on local
copyright law...)

Do you have specific concerns about that file or about the general
licensing for util-linux? Are there issues specific to your location I'm
not aware of?

[1]
http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/trusty/util-linux/trusty/view/head:/debian/copyright
[2]
http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/wily/util-linux/wily/view/head:/debian/copyright

Regards,

/ Matt
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Window Controls on the Right Side

[Mathieu Trudel-Lapierre]

Le mercredi 20 mai 2015 à 11:18 -0400, Rodney Dawes a écrit :
[...]
 There is no sponsorship of bugs in Launchpad.

I think it's vitally important here to make sure things are clear: yes,
there is such as thing as sponsorship. Once someone has a fix ready
that is both appropriate and well-executed (as reviewed by some person,
upstream or a domain expert), then developers can upload these fixes
for a contributor who has no upload access. It seems to me like this is
what Raphael was pre-emptively asking for.

However, it's *way* too early for this...

 If there is already a
 bug about the window controls, then simply make sure that ubuntu-ux is
 an affected project as well, and feel free to discuss details on
 implementation and design in there. However, please keep such discussion
 objective and technical, rather than filling it with the subjective
 commentary as is common with these sorts of polarizing religious topics.

I agree. This is a contentious issue that polarizes people, even though,
in the end, the actual location of the controls doesn't matter. We've
changed them once already and there was a lot of criticism. It seems
like now there'd be just as much, since people have gotten accustomed to
it, like it this way, etc.; just like there are others who are just
coming to Ubuntu and feel their placement is wrong.

If you want change to happen, the best way is to provide concrete
technical proof (studies?) that it's a better location -- anything else
boils down to personal opinion.

If what you're after is providing a setting so that users can customize
their systems, then you probably should bring this up on the appropriate
technical list (unity-design or unity-dev I guess?), so that domain
experts can say that it has already been considered, and why it wasn't
done yet.

Finally, if I can share a bit: when I concentrate on a window for an
extended period of time, it's maximized. This means I will have the menu
in the title bar, which is integrated in the top panel. Having the
window controls on the left in this case is fine since it would
otherwise be unbalanced to have even more icons on the right (plus these
icons have a vastly different purpose. Some are menu-like to effect an
action on the current window, the others provide global information
about my system. White space generally separates the two, unless the
window title is very long). The window controls also push the title
right just enough that it's almost lined up with the actual window,
rather than being above the Launcher.

-- 
Mathieu Trudel-Lapierre mathieu...@gmail.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/DC95CA5A 36E2 CF22 B077 FEFE 725C  80D3 C7DA A946 DC95 CA5A




signature.asc
Description: This is a digitally signed message part
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: openssl performance delta built-in vs custom compiled

[Mathieu Trudel-Lapierre]

On Dec 21, 2014 7:59 AM, Marcus Pollice marcus.poll...@gmail.com wrote:

use the Debian/Ubuntu patchset, which is what I will look at more closely
now. I suspect that maybe one of these patches changes something that
affects performance for small data sizes. Is there some documentation
available somewhere which patch changes what specifically?

I don't think there would be very many patches that are meant to improve
performance in our patch set (note, I did not check). However, the details
of what each patch changes should be in the patch description as well as in
debian/changelog. You may also which to check what are all the compiler
options coming from dpkg-buildflags used in the build... So if what you are
trying to do is build your own copy of openssl to enable a particular
feature, I would suggest using the source package as a base and using
debuild / chroots / PPAs to build your custom package, that way you'd
benefit from the same performance unless your custom changes impact them in
some way, with the least amount of effort.

/ Matt
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Could someone merge usb-modeswitch from Debian - currently usb-modeswitch from Ubuntu 14.10 (and 14.04) doesn't work with latest usb-modeswitch-data releases and doesn't support lots of Huawei (an

[Mathieu Trudel-Lapierre]

On ven, 2014-09-05 at 13:30 +0300, UAB „Bona Mens“ wrote:
[...]
 Lots of other Huawei related issues, like bug #1308895 are fixed in 
 updated usb-modeswitch (2.2.0+repack0-2) and usb-modeswitch-data
 (20140529-1) packages in Debian unstable.
 
 Please merge usb-modeswitch 2.2.0+repack0-2 package from Debian unstable
 to Ubuntu Utopic. 
 

I'd say this is in progress right now: the merge is pretty much done;
but before I go ahead and upload, I'd like to do a bit more testing on
the package. I've uploaded it to my PPA:

https://launchpad.net/~mathieu-tl/+archive/ubuntu/nv-build/+sourcepub/4397785/+listing-archive-extra

Note, this means you'll also need to install the usb-modeswitch-data
package from utopic-proposed yourself.

If you've been having issues with generic Huawei devices and that's
the piece you need from usb-modeswitch 2.2.0; then please see if you can
help me out with testing the above package. I'll get back to it this
evening.

Regards,

-- 
Mathieu Trudel-Lapierre mathieu...@ubuntu.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/DC95CA5A 36E2 CF22 B077 FEFE 725C  80D3 C7DA A946 DC95 CA5A




signature.asc
Description: This is a digitally signed message part
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Could someone merge usb-modeswitch from Debian - currently usb-modeswitch from Ubuntu 14.10 (and 14.04) doesn't work with latest usb-modeswitch-data releases and doesn't support lots of Huawei (an

[Mathieu Trudel-Lapierre]

On Sep 5, 2014 4:42 PM, Jackson Doak nosk...@ubuntu.com wrote:

 Have you made an FFe bug asking for the update? That would help.



I will take care of the merge shortly, I simply haven't had time to get
to it yet. Thanks for the reminder :-)

/ Matt
-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Please update usb-modeswitch packages in Trusty

[Mathieu Trudel-Lapierre]

Thanks, I'll take a look into this.

Mathieu Trudel-Lapierre mathieu...@ubuntu.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93


On Thu, Mar 27, 2014 at 8:04 AM, Mateusz Stachowski
mateusz.stachow...@wp.pl wrote:
 I opened a bug for that but I should probably add [FFe] to it because after
 all we are past the Feature Freeze. Also sorry for adding needs-packaging I
 probably shouldn't add that right because those packages are already in
 ubuntu archives they just need updating.

 https://bugs.launchpad.net/ubuntu/+source/usb-modeswitch/+bug/1296090

 I used the PPA of Debian usb-modeswitch maintainer and it didn't cause any
 troubles. My modem was already supported by the old packages available in
 Ubuntu repos.

 Below is a copy of text from Bug Description:

 The upstream released new versions of usb-modeswitch and usb-modeswitch-data
 on January 28th and 29th 2014.

 http://www.draisberghof.de/usb_modeswitch/

 This version supports many more usb modems than the one currently available
 in Trusty besides those packages haven't been updated for a very, very long
 time in Ubuntu.

 The packages are already available in Debian Testing and Unstable.

 https://packages.debian.org/search?keywords=usb-modeswitch

 Besides that the maintainer of the USB_ModeSwitch Debian package has set up
 a PPA providing this most recent release.

 https://launchpad.net/~odyx/+archive/usb-modeswitch

 NOTE: there is also a similar bug report from users that want the latest
 usb-modeswitch on Trusty.

 https://bugs.launchpad.net/ubuntu/+source/usb-modeswitch/+bug/1270499

 --
 Ubuntu-devel-discuss mailing list
 Ubuntu-devel-discuss@lists.ubuntu.com
 Modify settings or unsubscribe at:
 https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss


-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: could you add this feature or discuss it at 13.04 Developer Summit?

[Mathieu Trudel-Lapierre]

On Mon, Oct 15, 2012 at 1:25 PM, Brian labishi bni1...@live.com wrote:

 Hi. I'm new to Ubuntu and like it very much. Overall I like Ubuntu better
 than what I used to use, Windows. But one thing that I really miss from
 Windows is the ability to know what applications and services are connecting
 to the internet. In Windows I could log this kind of information. But I've
 asked some very knowledgeable computer people for help with Ubuntu and I'm
 told this can't be done on ubuntu.

 I was hoping that Ubuntu developers might address this shortcoming at the
 summit? I was told this is where these kind of things are discussed.

You're suggesting a very interesting project, yet one that is likely
to depend on a fair amount of new development.

Do we have other instances of this being asked by people, such as on
Ubuntu Brainstorm (I'll look too)? It would be important to know,
before committing time to work on such a thing, how important it's
perceived to be by our users.

Keeping in mind that there can be a very large number of connections
happening on a machine at any point in time, what kind of information
are you looking for? Is it to see everything that attempts to make a
connection or just what gets blocked by a firewall? Do you want to see
notifications on the desktop or are you looking for this at the server
level?

All the above are information that would be best to flesh out a bit in
advance before starting discussion just so that work items could be
derived from the resulting discussion.

Obviously, you don't *need* to discuss a project like this at UDS.
Perhaps it's just something people can start working on as a project,
and ask for specific things needed in Ubuntu to support using such an
application/service

Kind regards,

Mathieu Trudel-Lapierre mathieu...@ubuntu.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: DNS caching disabled for 12.10...still

[Mathieu Trudel-Lapierre]

On Sun, Oct 7, 2012 at 11:35 PM, Daniel J Blueman dan...@quora.org wrote:
[...]
 Good tip on the workaround, Mathieu. Looks like this doesn't work in
 Ubuntu 12.10 pre-release here:

 # echo cache-size=400 /etc/NetworkManager/dnsmasq.d/cache
 reboot
 $ ps -ef | grep dnsmasq
 nobody2057  1128  0 11:29 ?00:00:00 /usr/sbin/dnsmasq
 --no-resolv --keep-in-foreground --no-hosts --bind-interfaces
 --pid-file=/var/run/sendsigs.omit.d/network-manager.dnsmasq.pid
 --listen-address=127.0.1.1 --conf-file=/var/run/nm-dns-dnsmasq.conf
 --cache-size=0 --proxy-dnssec
 --enable-dbus=org.freedesktop.NetworkManager.dnsmasq
 --conf-dir=/etc/NetworkManager/dnsmasq.d


You can't see it on the command-line. Things are evaluated in order;
command-line parameters first, up to the --conf-dir parameter, and
then the files in that directory will be looked at and configuration
taken into account. However, it won't change the actual command-line
for the application, since it's indeed how it was started.

To see the result, you'll want to kill dnsmasq with the SIGUSR1 signal
-- this will force it to write out statistics to syslog. This is also
the way to list the nameservers used by dnsmasq.

Regards,

Mathieu Trudel-Lapierre mathieu...@ubuntu.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: DNS caching disabled for 12.10...still

[Mathieu Trudel-Lapierre]

On Sun, Oct 7, 2012 at 3:19 PM, Stéphane Graber stgra...@ubuntu.com wrote:
 On 10/07/2012 04:32 AM, Benjamin Kerensa wrote:

 On Oct 7, 2012 12:28 AM, Daniel J Blueman dan...@quora.org
 mailto:dan...@quora.org wrote:

 DNS caching was previously disabled [1] when dnsmasq was introduced in
 12.04 (one of the benefits), to prevent privacy issues, and to
 prevent local users from spying on source ports and trivially
 performing a birthday attack in order to poison the cache.

 Since dnsmasq eg introduced the standard port-randomisation
 mitigations [2] for Birthday attacks in 2008 and related hardening,
 what are the other technical reasons we should still keep this
 disablement, despite upstream keeping DNS caching enabled? (ie should
 upstream also disable DNS caching?)

 Of course, the impact of disabling DNS caching is considerable.
[...]

 Good points it does look like hardening and addressing some of the
 concerns has occurred it is possible perhaps that enabling caching was
 just overlooked but either way it would be nice to see it enabled in 13.04.

 dnsmasq still doesn't support per-user caching so it still doesn't meet
 the criteria we discussed with the security team last cycle and as such
 as kept in its current configuration.


With the small difference that you can now actually enable caching
should you choose to disregard the security implications. You can do
so by adding a file in /etc/NetworkManager/dnsmasq.d containing
cache-size=n where n is the size you want to use (default in dnsmasq
is 150, and set to 400 in NM upstream). The name of the file doesn't
matter.


Mathieu Trudel-Lapierre mathieu...@ubuntu.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Network Manager dependencies

[Mathieu Trudel-Lapierre]

On Wed, Aug 22, 2012 at 3:24 AM, Tom H tomh0...@gmail.com wrote:
 IMO, we'll end up sooner or later using NM on X-less boxes by default

It might be the case eventually, but we're not there yet.

 NM on Fedora can now handle bonding and bridging by reading
 /etc/sysconfig/netwok-scripts/ifcfg-* files.

 I was curious about whether NM could do the same by reading
 /etc/network/interfaces so I've just tried to install NM in an
 X-less Quantal VM.

Not by reading /e/n/i; but bonding and bridging support was added in
NetworkManager 0.9.4.0.

There is currently no way to configure it besides the rather opaque
files in /etc/NetworkManager/system-connections. This will be fixed
eventually, but you'll still see the configuration methods come up in
nm-applet first, before it's added to the command-line configuration
tool or to the /e/n/i parser (and parsing that is difficult, if not
dangerous anyway).


 apt-get install network-manager results in:

[...]
 and

 apt-get install network-manager -o APT::Install-Recommends=false results in:

[...]
 So, even an installation of NM without its Recommends (which isn't
 something that I like to do or, AFAIK, something that's recommended)
 results in the installation of packages that are only needed by a DE,
 like dconf-gsettings-backend, dconf-service, glib-networking,
 glib-networking-common, glib-networking-services,
 gsettings-desktop-schemas.

[...]
 Could NM's Depends and Recommends be pared down for an X-less
 use-case? Thanks.

We'll burn that bridge when we get there. :)

Seriously though; I agree that some of these requirements are
unfortunate, especially with all this coming from glib-networking
(which I'll look to make sure is really required). Any help you can
provide to reducing these requirements and making NetworkManager
suitable on X-less systems is definitely welcome. I recommend you come
hang around on #nm on Freenode if it's the case; this is where the
upstream NetworkManager development gets done.

Regards,

Mathieu Trudel-Lapierre mathieu...@ubuntu.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: How to install Precise without getting screwed?

[Mathieu Trudel-Lapierre]

On Tue, Apr 10, 2012 at 11:46 PM, Dane Mutters dmutt...@gmail.com wrote:
[...]
 So, now that we've gotten some matters of conduct out of the way (we have,
 haven't we?), does anyone care to suggest what to do about making the GUI(s)
 of Ubuntu more usable for those who aren't OK with the current offerings?

Have you considered trying the other window manager that are available
for installation? Between Kubuntu, Lubuntu, Xubuntu which each install
their own different window manager by default; and being able to
install GNOME Shell (gnome-shell) to replace Unity (or
gnome-session-fallback for a GNOME2-like look), there's a fair amount
of choice.

No matter which option you'll choose, there is bound to be some amount
of change in the look and feel, since even GNOME is moving away from
what you're used to seeing in 10.04 with the two panels. That will
mean some amount of relearning, with a varying transition period
depending on your choice.

As far as I can tell, from an LTS to LTS upgrade perspective it's all
a matter of choosing whether you want to spend increasing amounts of
time figuring out how to get the same look you were used to, or
spending a (relatively) finite amount of time relearning interface to
familiarize yourself with new window manager of choice. That's true
for all other distros at this point in time, the difference is that
Ubuntu has chosen to go with Unity as the default window manager for
Ubuntu Desktop installs (as opposed to Kubuntu or others).

Regards,

Mathieu Trudel-Lapierre mathieu...@ubuntu.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Drop Gwibber from default install

[Mathieu Trudel-Lapierre]

On Sun, Mar 11, 2012 at 10:25 PM, Benjamin Kerensa bkere...@ubuntu.com wrote:

 * Performance Improvements (Gwibber seems very laggy)

I can't say I've never seen gwibber suffer a bit from lagginess,
though it's hard to say how much or when that was happening exactly,
and pinpointing it to a specific release or action.

Have you opened a bug already about this particular issue? There's
still time to address specific issues despite some of the emails on
this thread may have lead people to believe. If there's a bug open,
and if that bug has data to support that a specific part of gwibber
could benefit in slight rework, then we'd have an area on which to
focus our efforts.

In other words, maybe there's some low-hanging fruit there which could
make the visible performance much better?

Mathieu Trudel-Lapierre mathieu...@gmail.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: tcp_mtu_probing on by default?

[Mathieu Trudel-Lapierre]

On Mon, Feb 6, 2012 at 6:15 PM, Martin Pool m...@canonical.com wrote:
 I have helped a few people recently who were having path MTU discovery
 problems, causing bulk TCP transfers to hang quasi-intermittently.
 Once you know the likely cause it's fairly easy but it's a fairly
 annoying problem for someone who doesn't recognize it.

 There is a kernel sysctl sudo sysctl -w net.ipv4.tcp_mtu_probing=1
 that seems fairly effective at detecting when the problem is occurring
 and automatically fixing it.   This implements RFC 4821.  It is off by
 default in the kernel.  I haven't seen any reports of problems caused
 by turning it on, but there may be some.

 I wonder if Ubuntu should turn it on in /etc/sysctl.d?

Admittedly I haven't really looked much into this and whether it's
likely to cause issues in some environments, but setting it to 1
indeed seems relatively safe.

  0 - Disabled
** 1 - Disabled by default, enabled when an ICMP black hole detected
  2 - Always enabled, use initial MSS of tcp_base_mss.

This s hould help those network paths for which fragmentation is
required.On the other hand, enabling this will cause more
retransmissions of segments in this case, which would mean an increase
in traffic. I don't think it's likely to be huge, but just something
to keep in mind.

The question would be how many people would benefit from this change?
I'd be tempted to say it probably doesn't affect all that many people
in general. If you've found a lot of people who had this issue, maybe
it's worth also trying to figure out if they have the same ISP, if
they try to connect to the same place, etc. in case it's an issue
outside their network.

Mathieu Trudel-Lapierre mathieu...@gmail.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Network Configuration

[Mathieu Trudel-Lapierre]

On Thu, Sep 15, 2011 at 3:03 PM, Kai Mast kai.m...@freakybytes.org wrote:

 Oneiric is already in Beta and we still have two different apps to
 configure network connections. One in the System Settings and one when
 using the indicators. Will that be the way Oneiric is shipped or will
 they be merged?

As far as I know, both will be shipped, and as for my own opinion,
both are needed.

Part of the reason for this is that they both deal with the exact same
backend; and not all variants of Ubuntu ship the gnome-control-center
pieces (I can already think of Xubuntu which is currently not shipping
it).

Furthermore, not all the settings are covered by both applications.
The gnome-control-center stuff includes proxy configuration which is
not something available (yet) in nm-connection-editor. This is partly
a decision coming from Gnome and a side effect of the move to Gnome
Shell, which ships its own module for the network indicator, while
still using some parts of nm-applet.

And all this brings me back to the fact that the gnome-control-center
module for Network uses dialogs coming from nm-applet, so for now they
are not really possible to separate.

Regards,

Mathieu Trudel-Lapierre mathieu...@ubuntu.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Disconnect via NetworkManager should send DHCPRELEASE when applicable

[Mathieu Trudel-Lapierre]

On Wed, May 11, 2011 at 9:10 AM, Thomas Novin tho...@xyz.pp.se wrote:
 Hello

 I think NetworkManager should send a DHCPRELEASE when the user chooses
 disconnect on a DHCP-connection.

 This is important when using a service that only allows one IP, then the
 MAC/IP-combo is locked for whatever the DHCP lease-time is. Sending a
 DHCPRELEASE frees up the lease making the service available for another
 computer.

 Any reason why not to do this? Should I file a Launchpad bug?

Hi Thomas,

I'm not sure if there is a specific reason. However, seems to me like
this would need careful consideration, given that what we usually want
is for NM to retain IP addresses at least on wired, if it gets
shutdown (e.g. the daemon is stopped or killed, etc.). Then you get
into the question of how to deal with differenciating a
user-initiatied *disconnect* rather than a shutdown of the daemon.

You'll also get into the cases where you really want to disconnect for
whatever reason, but want/need to retain your IP if possible. I think
this is the most likely use case.

I think this would probably be better served in your case by a
dispatcher script: you can drop files in
/etc/NetworkManager/dispatcher.d to have them run on various events;
just take a look at 01ifupdown for inspiration; you should easily be
able to do something like 'dhclient -r' from there to achieve the same
thing.

Otherwise, I guess you might want to ask on the NetworkManager list,
which would be better suited for this kind of targetted question about
this project.

Regards,

Mathieu Trudel-Lapierre mathieu...@gmail.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Ubuntu-devel-discuss Digest, Vol 54, Issue 1

[Mathieu Trudel-Lapierre]

On Sun, May 1, 2011 at 12:59 PM, Nedas Pekorius nedaspekor...@gmail.com wrote:
 Vodafone k4505-z modem on Natty is detected (not every time just
 instant after fresh install) it detects network and connects just for
 first time. Every each time It fails to connect or system fails to
 detect it. Everything was fine in pre-release versions like beta/beta2

Nedas,

Sorry to hear you're having issues with this modem, especially if it
was working during the development cycle.

If you haven't already done so, please file a bug in Launchpad with
the following command:

ubuntu-bug modemmanager

Then reply here with the bug number -- bug reports are usually a
better medium for handling this type of issue.

One thing I'd suggest testing if it works the first time you try is
would be to unplug and replug the modem; it may be put in a
power-saving mode that we're not able to recover from once the initial
connection is terminated.

If on the other hand you're booting with the device plugged in, the
same trick applies: boot with it disconnected, then connect after
booting, or unplug and replug it.

Regards,

Mathieu Trudel-Lapierre mathieu...@gmail.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: [Oneiric topic] IPv6

[Mathieu Trudel-Lapierre]

On Sat, Apr 30, 2011 at 1:38 PM, Jan Claeys li...@janc.be wrote:
 Etienne Goyer schreef op ma 18-04-2011 om 17:03 [-0400]:
 Any other major roadblock beside the above?  I am going to sift
 through the bugs tagged ipv6 on Launchpad, but if there's anything
 obvious I missed, please let me know.

 Finally, a question to consider is whether we want to address the
 IPv6-only use-case (ie, not dual-stack, no IPv4 configuration).  This
 has some implications, notably around d-i and NetworkManager.  Few
 networks are IPv6-only at this time, but it's bound to change in the
 near future.

 Also important here: the Ubuntu websites, the main archive/download
 server, and many mirrors don't support IPv6 currently...

 When you do an IPv6-only install (and you have no local mirror), it
 would be nice if an IPv6-capable mirror is configured in
 sources.list  ;)

Well, I didn't mention this precisely because the mirrors and other
services aren't IPv6 ready, but I thought it would be nice if the UDS
infrastructure was at least v6 ready: that you can get addresses, etc.

Mathieu Trudel-Lapierre mathieu...@gmail.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: Keyboard mapping selection comes to late during install

[Mathieu Trudel-Lapierre]

On Wed, Apr 27, 2011 at 8:10 AM, Alain-Olivier Breysse
yahoe@gmail.com wrote:
 Bonjour,

 Keyboard mapping selection comes too late during installation. When one
 wants to manually partition the drives on, for example, a French-Canadian
 keyboard, the key mapping selection window hasn't been presented yet (it
 will be after the time zone selection) thus obliging the user to manually
 call for that change which is not even possible if Ubiquity as been started
 directly during the boot process without selecting the Try Ubuntu option.


 This is not a bug but a choice made by the developper that needs to be
 adresssed and corrected since it affects all international users.

Do you mean this happens when you select the French language (for
instance), as soon as the CD starts, once you've first pressed a key
to get to the menu?

I find it's often easier to install a system with the French-Canadian
settings if you first boot with everything in English (e.g. keep
english as the language at the boot menu, select Install Ubuntu), then
select French for the language once you're presented with Ubiquity's
language selection / welcome dialog.

What this effectively does is keep you in English until the language
is changed in Ubiquity, but also leaves the system with a qwerty
keyboard layout (e.g. US English) rather than azerty, as I guess is
the precise issue you're referring to. The side effect is that you
won't be able to type accents very easily.

Otherwise, at the boot menu, select French for the language, and press
F3 (IIRC) to get to the keyboard selection list. You should be able to
select another layout from there.

I don't think it's feasible to please everyone in this case: for the
French language (as it is the case here), people from France will most
likely want the French azerty layout; people from Canada will want one
of the French Canadian qwerty layouts or US/Intl; and people from
other french-speaking nations are likely to have different
requirements as well. Whatever default layout is choosen is bound to
make people unhappy ;)

Mathieu Trudel-Lapierre mathieu...@ubuntu.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss

Re: [Oneiric topic] IPv6

[Mathieu Trudel-Lapierre]

On Mon, Apr 18, 2011 at 5:03 PM, Etienne Goyer
etienne.go...@canonical.com wrote:
[...]
 - NetworkManager: It used to be that NetworkManager insisted upon
 getting an IPv4 address, but Matt Trudel just marked bug #307598 as
 fixed two weeks ago.  Presumably, NM now work fine in IPv6-only network,
 but I have not tested yet.

The default settings are still to only get IPv4 by default and require
that something responds to bring up the interface, but you can turn
this requirement off, enable IPv6, turns *its* requirement off, or
turn IPv4 address requests off altogether. For Natty enabling IPv6
addresses in NM still requires manual intervention: you need to set a
drop-down to Automatic or whatever you want it to be; and choose
whether it needs to respond to bring up the interface.

For Oneiric, my plan is to change that default for new interfaces to
Automatic IPv6 and not requiring IPv4 or IPv6 to bring up interfaces,
which should make almost everyone happy.


 Any other major roadblock beside the above?  I am going to sift through
 the bugs tagged ipv6 on Launchpad, but if there's anything obvious I
 missed, please let me know.

We used to see a few issues related to broken routers blocking on 
DNS requests. Note sure if there's really still a lot of those, but I
guess it's something to keep in mind:  requests could block and
turn into a big annoyance for users who don't care about IPv6.

 Finally, a question to consider is whether we want to address the
 IPv6-only use-case (ie, not dual-stack, no IPv4 configuration).  This
 has some implications, notably around d-i and NetworkManager.  Few
 networks are IPv6-only at this time, but it's bound to change in the
 near future.

I *think* this should be fine on the desktop side, but I haven't
tested IPv6-only, just dual-stack. The main problem is that if you're
relying on only IPv6, you will still not reach a whole lot of things
on the Internet, including some of our infrastructure: for instance,
cdimage.u.c doesn't appear to have  addresses.

Regards,

Mathieu Trudel-Lapierre mathieu...@ubuntu.com
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93

-- 
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at: 
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss