Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.user > #167348 > unrolled thread

USB tethering and interface naming on Debian/unstable

Back to article view | Back to linux.debian.user

Contents

  USB tethering and interface naming on Debian/unstable Vincent Lefevre <vincent@vinc17.net> - 2016-05-02 11:30 +0200
    Re: USB tethering and interface naming on Debian/unstable David Wright <deblis@lionunicorn.co.uk> - 2016-05-02 16:00 +0200
      Re: USB tethering and interface naming on Debian/unstable Vincent Lefevre <vincent@vinc17.net> - 2016-05-06 01:10 +0200
        Re: USB tethering and interface naming on Debian/unstable David Wright <deblis@lionunicorn.co.uk> - 2016-05-06 04:00 +0200
          Re: USB tethering and interface naming on Debian/unstable Vincent Lefevre <vincent@vinc17.net> - 2016-05-06 11:30 +0200

#167348 — USB tethering and interface naming on Debian/unstable

What is the rule for interface naming when doing USB tethering
on Debian/unstable (with systemd)?

In December, I had enx02060b0e3333, but yesterday, I had enp0s20u2.
Isn't the interface supposed to be fixed?

-- 
Vincent Lefèvre <vincent@vinc17.net> - Web: <https://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)

[toc] | [next] | [standalone]

#167356

On Mon 02 May 2016 at 11:28:47 (+0200), Vincent Lefevre wrote:
> What is the rule for interface naming when doing USB tethering
> on Debian/unstable (with systemd)?
> 
> In December, I had enx02060b0e3333, but yesterday, I had enp0s20u2.
> Isn't the interface supposed to be fixed?

Not fixed; but predictable. See:
https://www.freedesktop.org/wiki/Software/systemd/PredictableNetworkInterfaceNames/

It would appear you have moved from scheme 4 to scheme 3, ie from
4. Names incorporating the interfaces's MAC address (example: enx78e7d1ea46da)
to
3. Names incorporating physical/geographical location of the connector of the hardware (example: enp2s0)

Cheers,
David.

[toc] | [prev] | [next] | [standalone]

#167516

On 2016-05-02 08:49:20 -0500, David Wright wrote:
> On Mon 02 May 2016 at 11:28:47 (+0200), Vincent Lefevre wrote:
> > What is the rule for interface naming when doing USB tethering
> > on Debian/unstable (with systemd)?
> > 
> > In December, I had enx02060b0e3333, but yesterday, I had enp0s20u2.
> > Isn't the interface supposed to be fixed?
> 
> Not fixed; but predictable. See:
> https://www.freedesktop.org/wiki/Software/systemd/PredictableNetworkInterfaceNames/
> 
> It would appear you have moved from scheme 4 to scheme 3, ie from
> 4. Names incorporating the interfaces's MAC address (example: enx78e7d1ea46da)
> to
> 3. Names incorporating physical/geographical location of the connector of the hardware (example: enp2s0)

Any reason why? I haven't changed anything in my configuration.

How can I move back to scheme 4? With a .link file? Scheme 3 doesn't
make sense for USB tethering, because there are various USB ports,
and anyone can be used.

-- 
Vincent Lefèvre <vincent@vinc17.net> - Web: <https://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)

[toc] | [prev] | [next] | [standalone]

#167519

On Fri 06 May 2016 at 00:57:35 (+0200), Vincent Lefevre wrote:
> On 2016-05-02 08:49:20 -0500, David Wright wrote:
> > On Mon 02 May 2016 at 11:28:47 (+0200), Vincent Lefevre wrote:
> > > What is the rule for interface naming when doing USB tethering
> > > on Debian/unstable (with systemd)?
> > > 
> > > In December, I had enx02060b0e3333, but yesterday, I had enp0s20u2.
> > > Isn't the interface supposed to be fixed?
> > 
> > Not fixed; but predictable. See:
> > https://www.freedesktop.org/wiki/Software/systemd/PredictableNetworkInterfaceNames/
> > 
> > It would appear you have moved from scheme 4 to scheme 3, ie from
> > 4. Names incorporating the interfaces's MAC address (example: enx78e7d1ea46da)
> > to
> > 3. Names incorporating physical/geographical location of the connector of the hardware (example: enp2s0)
> 
> Any reason why? I haven't changed anything in my configuration.

Difficult for me to say. The only jessie I'm running at present is an
upgrade from squeeze/wheezy and Debian policy is AFAIK to leave things
as they are. You have to install jessie afresh for Debian to use the
new methods automatically.

> How can I move back to scheme 4? With a .link file? Scheme 3 doesn't
> make sense for USB tethering, because there are various USB ports,
> and anyone can be used.

The last paragraph under the heading "What precisely has changed in v197?"
says "This combined policy is only applied as last resort. That means,
if the system has biosdevname installed, it will take precedence. If
the user has added udev rules which change the name of the kernel
devices these will take precedence too. Also, any distribution
specific naming schemes generally take precedence." So I'm guessing
that that is what locks in my "eth0"/"wlan0" names. I assume the file
concerned is /etc/udev/rules.d/70-persistent-net.rules which has the
lines:

# PCI device 0x14e4:0x1600 (tg3)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:1c:23:12:34:56", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth0"

# PCI device 0x8086:0x4222 (iwl3945)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:1c:bf:12:34:56", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="wlan*", NAME="wlan0"

The machine which uses a netgear dongle has:

# USB device 0x:0x (ndiswrapper)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="c4:3d:c7:12:34:56", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="wlan*", NAME="wlan0"

So you might want a line that looks like (using my mangled addresses):

SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="c4:3d:c7:12:34:56", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="wlan*", NAME="enxc43dc7123456"

Cheers,
David.

[toc] | [prev] | [next] | [standalone]

#167527

On 2016-05-05 20:50:18 -0500, David Wright wrote:
> On Fri 06 May 2016 at 00:57:35 (+0200), Vincent Lefevre wrote:
> > On 2016-05-02 08:49:20 -0500, David Wright wrote:
> > > On Mon 02 May 2016 at 11:28:47 (+0200), Vincent Lefevre wrote:
> > > > What is the rule for interface naming when doing USB tethering
> > > > on Debian/unstable (with systemd)?
> > > > 
> > > > In December, I had enx02060b0e3333, but yesterday, I had enp0s20u2.
> > > > Isn't the interface supposed to be fixed?
> > > 
> > > Not fixed; but predictable. See:
> > > https://www.freedesktop.org/wiki/Software/systemd/PredictableNetworkInterfaceNames/
> > > 
> > > It would appear you have moved from scheme 4 to scheme 3, ie from
> > > 4. Names incorporating the interfaces's MAC address (example: enx78e7d1ea46da)
> > > to
> > > 3. Names incorporating physical/geographical location of the connector of the hardware (example: enp2s0)
> > 
> > Any reason why? I haven't changed anything in my configuration.
> 
> Difficult for me to say. The only jessie I'm running at present is an
> upgrade from squeeze/wheezy and Debian policy is AFAIK to leave things
> as they are. You have to install jessie afresh for Debian to use the
> new methods automatically.

Yes, it was jessie that was installed in the first place, in June 2015.
Then I immediately tracked unstable.

But the new methods are not in jessie yet (that's why I still have
eth0, because it was seen before I switched to unstable).

> > How can I move back to scheme 4? With a .link file? Scheme 3 doesn't
> > make sense for USB tethering, because there are various USB ports,
> > and anyone can be used.
> 
> The last paragraph under the heading "What precisely has changed in v197?"
> says "This combined policy is only applied as last resort. That means,
> if the system has biosdevname installed, it will take precedence. If
> the user has added udev rules which change the name of the kernel
> devices these will take precedence too. Also, any distribution
> specific naming schemes generally take precedence." So I'm guessing
> that that is what locks in my "eth0"/"wlan0" names. I assume the file
> concerned is /etc/udev/rules.d/70-persistent-net.rules which has the
> lines:
> 
> # PCI device 0x14e4:0x1600 (tg3)
> SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:1c:23:12:34:56", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="eth*", NAME="eth0"
> 
> # PCI device 0x8086:0x4222 (iwl3945)
> SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:1c:bf:12:34:56", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="wlan*", NAME="wlan0"
> 
> The machine which uses a netgear dongle has:
> 
> # USB device 0x:0x (ndiswrapper)
> SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="c4:3d:c7:12:34:56", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="wlan*", NAME="wlan0"
> 
> So you might want a line that looks like (using my mangled addresses):
> 
> SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="c4:3d:c7:12:34:56", ATTR{dev_id}=="0x0", ATTR{type}=="1", KERNEL=="wlan*", NAME="enxc43dc7123456"

This was mainly for jessie. I thought that this file was more or less
obsolete for unstable (and the next stable). So, what is better? This
/etc/udev/rules.d/70-persistent-net.rules file? A .link file[*]?
Something else? Is there official documentation?

[*] https://www.freedesktop.org/software/systemd/man/systemd.link.html

-- 
Vincent Lefèvre <vincent@vinc17.net> - Web: <https://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.user

csiph-web