On Sat 20 Dec 2025 at 00:35:33 (-0600), Mike McClain wrote:
> After installing trixie without a desktop, configuring it to my
> tastes, installing my prefered tools I rebooted to see if I'd broken
> something. Let me point out that I removed nothing, only added
> packages I normally use like gpm, jed, lynx and localc.
> Upon reboot I have no internet connection though the cable is
> still hooked to my ISPs modem.
> I've spent several hours reading man pages about systemctl,
> networkctl, systemd.network, systemd-networkd.service, ad nauseum.
> Most of those man pages gave a little sketch and said 'See Also:'.
> I have a simple system with static IP addressing of my computer, a
> router, printer and Windows computer that is seldom on.
> It has worked well through several versions of Debian but this
> upgrade has me stumped.
> I read a post a while back relating a conversation from DebCon
> where professional system managers pushed NetworkManager to be the
> default in Debian. While I don't mind that it should be included this
> added complexity does me no good at all and I'd be very surprised
> if the professional network managers aren't a very small minority of
> Debian users.
> More and more Debian is looking like Windows and less like the
> unix of old, a collection of small tools that do one thing and do
> it well.
> Pardon the rant, but if someone could tell me how to get the
> network back up in trixie I'd appreciate it.
If you installed trixie without a desktop, the d-i should have left
you with ifupdown running the network. The main difference, AIUI, from
bookworm is that it installs dhcpcd-base in place of isc-dhcp-client.
So I'm not sure why you're reading all those man pages above.
When you last tackled this problem seven weeks ago, you wrote that
you use little scripts with ip commands to configure the network.
I don't understand what that was all about, nor why you were tracing
system calls and quoting random error messages without context. On the
face of it, the narrative in the first paragraph above suggests you
just broke the system with inappropriate configuration changes.
As you wrote there 'ip link set dev eth0 up', I suppose the most
obvious question is whether the name of the interface in
/etc/network/interfaces matches the actual interface's name:
$ my-listings /etc/network/interfaces
/etc/network/interfaces :
# This file describes the network interfaces available on your system
# and how to activate them. For more information, see interfaces(5).
source /etc/network/interfaces.d/*
# The loopback network interface
auto lo
iface lo inet loopback
# The primary network interface
allow-hotplug enp2s0
iface enp2s0 inet dhcp
→→ ↑↑↑↑↑↑
$
$ ip link
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: enp2s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode
DEFAULT group default qlen 1000
link/ether 00:13:72:00:00:00 brd ff:ff:ff:ff:ff:ff
altname enx001372000000 [redacted]
$
/etc/hosts contains lines like:
127.0.0.1 localhost
192.168.1.1 router.corp router
192.168.1.12 brother.corp brother
127.0.1.1 alum.corp alum # 192.168.1.19
# The following lines are desirable for IPv6 capable hosts
::1 localhost ip6-localhost ip6-loopback
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
I use static addresses, dished out by the router; Above, I've included
this host, the router and a printer, but eliminated all the other
hosts/TVs etc.
The /etc/resolv.conf file gets written by dhcpcd-base, thanks to the
last line of /etc/network/interfaces.
$ my-listings /etc/resolv.conf
/etc/resolv.conf :
# Generated by dhcpcd from enp2s0.dhcp
# /etc/resolv.conf.head can replace this line
nameserver 192.168.1.1
# /etc/resolv.conf.tail can replace this line
$
Cheers,
David.
