News for hackers. Stuff that matters.
Re:Debian reconsiders init-system diversity by kresten on Tuesday November 12, 2019 @01:46AM (#21512048)
Yep agreed - solid assessment.

Apart from networking. That’s an f*ing mess without systemd.

Re:Debian reconsiders init-system diversity by isodude on Tuesday November 12, 2019 @02:11AM (#21512130)
Been trying out systemd-networkd for a couple of years now and it is a mess. Especially when you want to do something like ipsec/vpn/vxlan. Also, no possible way to reload a specific interface?
Re:Debian reconsiders init-system diversity by brmgb on Tuesday November 12, 2019 @10:52AM (#21515020)
> Especially when you want to do something like ipsec/vpn/vxlan.

I have a wireguard tunnel setup by networkd with custom rules to only send traffic from a slice through it. It works fine. Take a look at the netdev documentation [1].

What are you trying to do exactly ?

[1] https://www.freedesktop.org/software/systemd/man/systemd.net...

Re:Debian reconsiders init-system diversity by isodude on Thursday November 14, 2019 @01:51PM (#21538172)
The problem is rather how the troubleshooting works.

You don't know how the network will be parsed or which statements that works with your version. Very difficult to debug when it doesn't behave as it should.

After a lot of fiddling with source routing on e.g. Debian Strech I was able to something working. But only half of the times I restarted networkd.

I would like some better tooling to assist with more difficult setups.

Re:Debian reconsiders init-system diversity by pas on Friday November 15, 2019 @03:42PM (#21548553)
Hm, Stretch has an old version, no? Or do you use a backport repo?

But, yes, turning on debug mode is harder than it should be, and it's usually not the right level. (I usually just go for strace first, but with netdev I guess it's using netlink to talk to the kernel, so that'll not help much :/ )

Re:Debian reconsiders init-system diversity by isodude on Saturday November 16, 2019 @02:22AM (#21551482)
Backport, because all the needed statements are not there.

I use strace etc (and debug level in networkd), but then you realise that this should make it easier not tougher. I did learn about ifup -v though which gave me all I needed. But both tools silently ignores flags that is wrong.

I'm writing my own tool now that is more strict in that sense instead.

Sorry about the rant but it's at least three trips to get everything working flawless in systemd-networkd now..

Re:Debian reconsiders init-system diversity by pas on Saturday November 16, 2019 @09:50AM (#21552876)
When it comes to Linux networking I find that tooling is very, very, veery bad, compared to what it should/could be.

Of course it's free and open source, so we can at least try to make it better besides complaining, but that part is harder than it should be (otherwise the whole problem would have been "solved" by now).

Re:Debian reconsiders init-system diversity by isodude on Saturday November 16, 2019 @05:20PM (#21555325)
Totally agree. I find network management bad wherever you go.

Systemd-networkd is fine layout wise, it's just the tooling that could be improved upon. Also here you see the problem with systemd-networkd being shipped together with systemd deb, if it were to be standalone it would be so easy to backport just that.

Re:Debian reconsiders init-system diversity by isodude on Sunday November 17, 2019 @04:28PM (#21560150)
And I just realised that it should not be that hard to control the interface via dbus to networkd.

Coreos even made a complete systemctl dbus integration.

Re:Debian reconsiders init-system diversity by noodlesUK on Tuesday November 12, 2019 @02:18AM (#21512161)
Funnily enough I found it much easier setting up networking in weird configurations using Gentoo... For simple DHCP stuff yeah but iirc you can use NetworkManager without systemd.
Re:Debian reconsiders init-system diversity by justinclift on Tuesday November 12, 2019 @03:49PM (#21518036)
For servers which get set up once and are expected to be left alone from then on, that's not the case.

With anything dynamic though - laptops, VMs - the various solutions can be pretty hit and miss. NetworkManager was terrible for many years, but seems relatively ok these days.