I've did a fresh install of Gentoo. I've set up network to be started at boot, together with ssh daemon. After restarting, kaboom, I could not connect to that machine. So I had to acquire physical access to it again and start investigating ... At first I thought I did some mistake configuring the kernel driver for the network card. But reading dmesg, I found it detected and initialised properly ... only after reading further I've found the problem is that it got renamed. This should not happen as it breaks networking completely. Reproducible: Always Steps to Reproduce: from http://www.gentoo.org/doc/en/handbook/handbook-amd64.xml?part=1&chap=8#doc_chap2 1. cd /etc/init.d 2. ln -s net.lo net.eth0 3. rc-update add net.eth0 default ... 4. reboot 5. ping [address of the machine in question] Actual Results: $ ping 192.168.1.12 PING 192.168.1.12 (192.168.1.12) 56(84) bytes of data. From 192.168.1.10: icmp_seq=1 Destination Host Unreachable From 192.168.1.10: icmp_seq=2 Destination Host Unreachable From 192.168.1.10: icmp_seq=3 Destination Host Unreachable ... Expected Results: $ ping 192.168.1.12 PING 192.168.1.12 (192.168.1.12) 56(84) bytes of data. 64 bytes from 192.168.1.12: icmp_seq=1 ttl=64 time=0.283 ms 64 bytes from 192.168.1.12: icmp_seq=2 ttl=64 time=0.306 ms 64 bytes from 192.168.1.12: icmp_seq=3 ttl=64 time=0.301 ms ... sorry, but documenting this as "If you later find out the assumption about the network interface name (which we currently document as eth0) was wrong, then" is no solution (and, besides, when I installed, I'm pretty sure the Handbook hasn't included this text ...) I'm not interested in playing "guess what the interface name will be this week". I need to know the name in advance, before rebooting, to have the things working after the reboot.
This was mostly fixed already by documenting the new udev interface naming in the handbook from bug 466262 In the "Automatically Start Networking at Boot" section it says: <quote> If you later find out the assumption about the network interface name (which we currently document as eth0) was wrong, then - update the /etc/conf.d/net file with the correct interface name (like enp3s0 instead of eth0), - create new symbolic link (like /etc/init.d/net.enp3s0), - remove the old symbolic link (rm /etc/init.d/net.eth0), - add the new one to the default runlevel, and - remove the old one using rc-update del net.eth0 default. </quote> But it should say something like: <quote> If your virtual manager is sys-fs/udev then you can safely expect the new network interface names are used which are explained here" </quote> Then link to http://www.gentoo.org/doc/en/handbook/handbook-amd64.xml?part=4&chap=2#doc_chap4
swift, can you add a small link to the page that explains it plus mention it's the default with the default virtual manager udev so it's clear if he wants eth0 he'll need something simple as net.ifnames=0 when configuring grub's config
I'll add in the link in the note: """ Note: This assumes that your network interface will be called eth0. This is, however, very system dependent. It is recommended to assume that the interface is named the same as the interface name when booted from the installation media if the installation media is sufficiently recent. More information can be found in [Network Interface Naming]. """ udev is the default during the installation and we don't document switching to another one during the installation, so I think it would be wrong to start talking about it; otherwise we'll have to document as well how to deviate from the standard and that would clutter the documentation imo. From what I heard, recent installation media should use the proper naming automatically (so if during the installation your interface is called enp3s0 then it is save to assume this will be the case after booting as well). Ok for you?
(In reply to Sven Vermeulen from comment #3) > Ok for you? sure
Committed in CVS
