There's a right way to enumerate network interfaces and a wrong way (which is dependent on the language you're using for enumeration - the wrong way is assuming that eth0 is the only network device).
Bear in mind that the new convention is not entirely dissimilar to how WiFi cards have worked on Linux for some time, so the only things that will break will be things that:
a) Assume eth0 instead of enumerating devices
b) Have a legitimate reason for using eth0 only (I can't think of one, but I wouldn't be surprised if one existed)
I got around to trying out Gentoo, which has amazing documentation that really drills down into what you're doing and why. That's when I ran ifconfig and iwconfig for the first time and realized I probably would have been using Linux years earlier if I had known how to find out that my NIC was wlan0.
So I'd say referencing eth0 throughout is fine, as long as at the beginning you take the time to point out that it may not be eth0, and more importantly, how to find out what it is.
If for example you're running Linux on a Macbook Air you have no eth0 by default. Ergo, documentation referencing it is technically incorrect - it would be possible to say badly written at this point but that's even more pedantic than your comment ;)
Everyone else has separate device namespaces for - you might have an fxp0 , rl0, hme0, etc. on any BSD or Solaris machine.
Rather than looking in /dev, they really ought to be parsing the output of ifconfig -a.
2. You mean (IIRC) 'ip link show'.
ifconfig on linux isn't maintained, and breaks in a bunch of situations (Aliases on VLANs on bonds, for one)
3. And you should of course use python-procfs or your preferred equiv, not parse the output of other tools.
On the article in general: can't you use /dev/by-manufacturer these days or similar? I haven't been working on OS stuff recently so can't remember if that became standard.
> 2. You mean (IIRC) 'ip link show'. ifconfig on linux isn't maintained, and breaks in a bunch of situations (Aliases on VLANs on bonds, for one)
This seemingly simple change seems to be taking forever to gain traction; "use `ip` for network config on linux".
There is so much googlable stuff instructing one to use ifconfig, route, etc. that it seems to self perpetuate. I often find people telling others that "you can't do that in Linux", when the real answer is along the lines of "you can't do that with the route command" (to give an example that I've seen a few times recently, route can't manipulate routing tables)
If you're going to deprecate a command, make running the old one either:
1. Not exist, and thus have the shell throw an "command not found" error
2. Print a useful message like "Use <newcommand> instead"
3. Work, but do #2 as well.
The current situation is like leaving a pair of rusty pliers out in the open, and hiding the brand new socket set when the goal is tightening a bunch of bolts.
When I worked at Red Hat (till 2005) their official RHCE documentation still told people to use ifconfig and route.
Not all add-in cards have a method to expose their Linux interface name(s) to external port mapping. biosdevname may provide incorrect names for such. Discussions are ongoing on the netdev mailing list to standardize a method of exposing such mapping.
Seems like this would be kind of a deal-breaker for any wholesale move like this. I'd be fairly surprised if this got any widespread adoption anytime soon.
em - Intel PRO/1000 10/100/Gigabit Ethernet device
Edit: Reading more of that thread I may be off-base here, but naming a network interface em that's not powered by certain Intel chipsets will be very confusing to BSD users. Just my 2 cents.
On other distributions, "udev" accomplishes the same thing.
The only point when interface naming is arbitrary is at installation time, then it stays the same forever.
Not only this will add confusion by introducing a bunch of new names for network interfaces, it will also break applications that rely on them being called "ethX".
It will also make it impossible to manually assign names to interfaces. For instance, if you have a configuration that uses an embedded interface and you want to add another interface on a card, you can manually assign the same "ethX" name to it, but not if the names depend on the physical characteristics of the hardware.
Ability to rename:
Why the eth prefix can't be used:
While this novelty certainly may appeal to the folks who are used to alphabet soup in the network adapters list, and I too think that the adapters that distinguish between the T568A and T568B cable layout have a degree of coolness in them by creating an extra value in being able to distinguish between the straight and crossover cables - overall it seems like a step back to me.
I would plug in the interfaces into a switch - and all I care is to find the same MAC address as I see on the switchports - not which bus they attach to.
If the machine is coldstarted then previous devices could be randomly reshufled. Imagine you have a datacenter with 100 machines and you have some kind of provisioning scheme where you coldstart them often. It would be a nightmare having to run ethtool -p <dev> and replug cables.
We have solved the problem by providing a consistent algorithm of sorting all the devices in a udev callback. We have a known set of hardware so we know which cards are add-ons (our machines have 4+ eth ports) and which ones are built-ins. So it works out, however, it would be nice to generalize it somehow.
The detection order would change, but the ifcfg file (RH) or udev rules (SuSE and others) would not change, so your NICs will stay the same these days.
> Subsequent cold starts only read the files
No they don't. Thare no files to read. You just wrote above in your commetn that the first time OS boots ifcfg are generated. So after a machine is coldstarted there are no ifcfg files! They are generated once per coldstart and saved. Then during each restart those files are read and everything is fine.
Now, I was saying, you can have a provisioning scheme where during coldstart, in your kickstart file you can fetch and write stable ifcfg files to prevent machines from creating their own. What we do instead is install a custom RPM with a udev script where an algorithm sorts the network interfaces in a stable way (since we only deal with a known and limited set of motherboards and network cards).
As long as you don't reinstall your OS, or lose your data, your config files are there. Cold starting hardware does not delete data.
Are you trolling?
Alright, figured it out. When you talk about coldstaring you really mean restarting the machine, as in pressing the reset button for example. When I talk about coldstarting I mean starting with a bare hardware box and installing an OS on it.
> Are you trolling?
Actually I thought you were. Sorry for the misunderstanding.
[EDIT: I won't edit my previous posts, otherwise your posts won't make sense, let others laugh at my bad English (it is a 3rd language so I don't mind)]
But that's because PCI bus walk order doesn't change across reboots.
But otherwise I understand that devices are detected in parallel and whichever ones return first are designated eth0, eth1 etc.
We had this problem on some machine that have 4+ eth0 devices. We use RHEL|CentOS 5.x (see in the post above how we solved the problem).
I once had a server (in the pre-udev era) that would occasionally shuffle its network devices after reboots. I don't remember how I fixed it but I do remember the short phase of amusement, followed by a longer phase of frustration.
While udev is more universal, iftab felt elegant and unixy.
~$ sudo egrep -Irl 'eth[0-9]' /etc /usr 2>/dev/null |wc -l
My hope is that consumer distributions will stop presenting the device name and configuration at all. When I'm trying to show new Ubuntu users how to connect to wired and wireless networks, it's annoying to have to explain what NetworkManager means by "Auto eth0".