| 1 |
-----BEGIN PGP SIGNED MESSAGE----- |
| 2 |
Hash: SHA256 |
| 3 |
|
| 4 |
Hi all, |
| 5 |
|
| 6 |
I respect both sides of the discussion, because: |
| 7 |
|
| 8 |
a) I once set up an old P3-700 with 5+1 eth cards in 6 different |
| 9 |
networks as (bridging)router and truly benefited from the ability to |
| 10 |
change a broken NIC - which happened quite often due scrap-metal |
| 11 |
hardware - without ending up with martian packages, dhcp service on |
| 12 |
the wrong places. But that was 1 incident in 10 years. |
| 13 |
|
| 14 |
b) I use multi-nic servers, some with onboard and extention NICs |
| 15 |
|
| 16 |
c) I tend to move my setups (esp. my laptop) around between different |
| 17 |
hardware (nearly identical thinkpad R61/X61), and I _share_ my |
| 18 |
installation with other/new users by cloning my disc (well rsync), |
| 19 |
lets call this stageN installation. |
| 20 |
|
| 21 |
d) I abuse an old multiport GBit card as GBit switch in my desktop, |
| 22 |
besides an onboard one. |
| 23 |
|
| 24 |
e) Some distro/driver constellations (archlinux?) tend to name their |
| 25 |
wireless lan eth*. |
| 26 |
|
| 27 |
This resulted in one decision per setup, whether or not to set |
| 28 |
/etc/conf.d/udev's |
| 29 |
|
| 30 |
> persistent_net_disable="yes" persistent_cd_disable="yes" |
| 31 |
|
| 32 |
Either to avoid random names due hardware replacement (a) or changed |
| 33 |
module loading order (b, inside debian initrd) |
| 34 |
or to just use kernel names (eth0, wlan0) because no other cards |
| 35 |
present (c) or the NIC drivers compiled into the kernel (d). |
| 36 |
e) never happened to me. |
| 37 |
|
| 38 |
It always bugged me to fix/reboot systems which needlessly end up with |
| 39 |
eth1/wlan1 because some stupid pre-persistent_net_disable did not |
| 40 |
recognize beeing run on an entirely different hardware. |
| 41 |
|
| 42 |
So can we just watch out for the disable="yes" setting and migrate it |
| 43 |
during udev's pkg_install phases __and__ post an big fat warning |
| 44 |
(elog, news item) on the wall? |
| 45 |
|
| 46 |
I assume most linux users do not operate |
| 47 |
servers/multi-nic/multi-networking setups, do not clone their setups |
| 48 |
to other hardware. |
| 49 |
Given that, these user will almost only see the 'my nics changed names |
| 50 |
and i cannot connect to the internet' errors due some moronic or |
| 51 |
unavoidable change in initrd/module loading. |
| 52 |
That might be the driving force behind udev persistence in the first |
| 53 |
place. |
| 54 |
|
| 55 |
I'd be glad if I we respect setups w/ custom-built kernels, w/o |
| 56 |
initrds, roots capable of choosing network-name-persistence iff |
| 57 |
needed, users adoring the possibility of just dd(1)'ing installations |
| 58 |
to new hardware without reinstalling or entering an new product code. |
| 59 |
|
| 60 |
rant=1; And I'd like to avoid dozends of conversations like "Yeah, |
| 61 |
your setup/firewall/rouing/... command no longer works, eth0 is no |
| 62 |
enp0xx2_at_home_lid_open or was it _bluetooth_turned_off. Didn't you |
| 63 |
read the post on some derps mailing list." with haunted people not |
| 64 |
knowing better than asking me about their problems. |
| 65 |
Not to mention all online documentation/forum posts referring to eth0. |
| 66 |
rant=0; |
| 67 |
|
| 68 |
Keep up the good work! |
| 69 |
|
| 70 |
Michael |
| 71 |
|
| 72 |
- -- |
| 73 |
Michael Weber |
| 74 |
Gentoo Developer |
| 75 |
web: https://xmw.de/ |
| 76 |
mailto: Michael Weber <xmw@g.o> |
| 77 |
-----BEGIN PGP SIGNATURE----- |
| 78 |
Version: GnuPG v2.0.19 (GNU/Linux) |
| 79 |
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ |
| 80 |
|
| 81 |
iF4EAREIAAYFAlD1FmAACgkQknrdDGLu8JA68wD/Vuw8mL7O0T398QR7OetqDoLN |
| 82 |
pQ7kJz9nveemDxw7o9MBAJSsyQ/DWIKLsqudXjlXhTPQEd0Od6vDBEL6IeFtXCjc |
| 83 |
=AfSI |
| 84 |
-----END PGP SIGNATURE----- |
Archives