I'm just verifying that on my end; note that the 99-default.link is
going to be sufficient to trigger a rename from eth0 to ens3; which
then prevents any rename once we mount root.
Here's the trigger:
% /usr/share/initramfs-tools/hooks/udev
# copy .link files containing interface naming definitions
mkdir -p $DESTDIR/lib/systemd/network/
find /lib/systemd/network -name '*.link' -execdir cp -pt
$DESTDIR/lib/systemd/network/ '{}' +
if [ -d /etc/systemd/network ]; then
find /etc/systemd/network -name '*.link' -execdir cp -pt
$DESTDIR/lib/systemd/network/ '{}' +
fi
If you create a .link file in /etc/systemd/network/ lower than 99-*
then those link rules will apply first.
On Fri, May 11, 2018 at 9:24 AM, Daniel Axtens
<daniel.axt...@canonical.com> wrote:
> Ok, so the bit I'm stuck on is how the link files and the netplan
> generator are getting pulled into the initramfs then.
>
> ubuntu@btest:~$ sudo update-initramfs -u
> update-initramfs: Generating /boot/initrd.img-4.15.0-20-generic
>
> ubuntu@btest:~$ lsinitramfs /boot/initrd.img-4.15.0-20-generic | grep \\.link
> lib/systemd/network/99-default.link
> ubuntu@btest:~$ lsinitramfs /boot/initrd.img-4.15.0-20-generic | grep netplan
> ubuntu@btest:~$ lsinitramfs /boot/initrd.img-4.15.0-20-generic | grep generate
> ubuntu@btest:~$ lsinitramfs /boot/initrd.img-4.15.0-20-generic | grep
> system-generators
>
> As you can see there's no generator and no link files in my initramfs -
> by what mechanism is it supposed to work? What package/script/tool is
> supposed to pull the link files in?
>
> --
> You received this bug notification because you are subscribed to
> netplan.
> Matching subscriptions: netplan
> https://bugs.launchpad.net/bugs/1770082
>
> Title:
> systemd-networkd not renaming devices on boot
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/netplan/+bug/1770082/+subscriptions
--
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to systemd in Ubuntu.
https://bugs.launchpad.net/bugs/1770082
Title:
systemd-networkd not renaming devices on boot
Status in netplan:
Incomplete
Status in systemd package in Ubuntu:
New
Bug description:
=== systemd issue ===
Renaming devices doesn't seem to work.
If I disable all other network configuration and create
/etc/systemd/network/10-network.link with:
[Match]
MACAddress=52:54:00:c1:c9:bb
[Link]
Name=myiface3
I expect this to cause the device with that MAC address to be renamed
to myiface3. However, when I reboot, I instead see:
$ ip l
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode
DEFAULT group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: ens3: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT
group default qlen 1000
link/ether 52:54:00:c1:c9:bb brd ff:ff:ff:ff:ff:ff
The device is not renamed.
This link file is pretty much identical to Example 2 in
https://www.freedesktop.org/software/systemd/man/systemd.link.html.
The renaming does work if I boot with net.ifnames=0, and oddly, it
also works if I unbind the device and rebind it as netplan apply does.
No setting of NamePolicy seems to help.
=== Original Bug ==
'set-name:' doesn't change the name of a network interface on boot, it
only works when you do netplan apply.
Say I take this 50-cloud-init.yaml file:
# This file is generated from information provided by
# the datasource. Changes to it will not persist across an instance.
# To disable cloud-init's network configuration capabilities, write a file
# /etc/cloud/cloud.cfg.d/99-disable-network-config.cfg with the following:
# network: {config: disabled}
network:
version: 2
ethernets:
ens3:
dhcp4: true
match:
macaddress: 52:54:00:de:bd:f6
set-name: ens3
Say I change set-name to 'myiface3' and reboot. I expect that the
device will be called myiface3 and brought up fine with dhcp. However,
instead I see:
$ ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group
default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: ens3: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default
qlen 1000
link/ether 52:54:00:de:bd:f6 brd ff:ff:ff:ff:ff:ff
The name has not been changed, and the device has not been brought up.
If I run netplan apply however, I see the following:
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group
default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
3: myiface3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state
UP group default qlen 1000
link/ether 52:54:00:de:bd:f6 brd ff:ff:ff:ff:ff:ff
inet 192.168.122.151/24 brd 192.168.122.255 scope global dynamic myiface3
valid_lft 3575sec preferred_lft 3575sec
inet6 fe80::5054:ff:fede:bdf6/64 scope link
valid_lft forever preferred_lft forever
So names are successfully changed with netplan apply.
This seems to be some udev-related timing or priority issue that I'm
still trying to hunt down.
This breaks some forms of migration in certain cloud environments.
To manage notifications about this bug go to:
https://bugs.launchpad.net/netplan/+bug/1770082/+subscriptions
--
Mailing list: https://launchpad.net/~touch-packages
Post to : touch-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~touch-packages
More help : https://help.launchpad.net/ListHelp
