Created attachment 557482 [details] patch Hi, as you maybe know net-tools has been deprecated for some time. We (as well as other distributions) try to eviscerate net-tools from as many packages as possible and replace it with iproute. I see that libguestfs-1.17.1/appliance/init script uses ifconfig, route and netstat. Would it be possible to use 'ip' command instead (see attached patch) ? And therefore also replace (Build)Requires: net-tools with (Build)Requires: iproute in Fedora's spec file ? Thanks !
Is it possible there is a mistake in the patch (I'm not very familiar with the 'ip' commands)? With the patch applied, the network doesn't work. Here is the old working config: ><rescue> cat /etc/resolv.conf # Generated by NetworkManager domain home.annexia.org search home.annexia.org nameserver 192.168.0.254 ><rescue> ip r default via 169.254.2.2 dev eth0 169.254.0.0/16 dev eth0 proto kernel scope link src 169.254.2.10 ><rescue> ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 52:54:00:12:34:56 brd ff:ff:ff:ff:ff:ff inet 169.254.2.10/16 brd 169.254.255.255 scope global eth0 inet6 fe80::5054:ff:fe12:3456/64 scope link valid_lft forever preferred_lft forever Here is the new non-working config: ><rescue> cat /etc/resolv.conf # Generated by NetworkManager domain home.annexia.org search home.annexia.org nameserver 192.168.0.254 ><rescue> ip r default via 169.254.2.2 dev eth0 169.254.0.0/16 dev eth0 proto kernel scope link src 169.254.2.10 ><rescue> ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 brd 127.255.255.255 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 52:54:00:12:34:56 brd ff:ff:ff:ff:ff:ff inet 169.254.2.10/16 brd 169.254.255.255 scope link eth0 inet6 fe80::5054:ff:fe12:3456/64 scope link valid_lft forever preferred_lft forever I cannot see any significant different, but in the second (non-working) case, the VM cannot connect to anything, not even by IP address.
Actually I can see a difference ... < inet 169.254.2.10/16 brd 169.254.255.255 scope global eth0 > inet 169.254.2.10/16 brd 169.254.255.255 scope link eth0
I changed your patch so that: -ip addr add 169.254.2.10/16 brd + dev eth0 scope link +ip addr add 169.254.2.10/16 brd + dev eth0 scope global and the network is working again. Obviously 169.254.* addresses probably shouldn't be global, but that may be another bug. It is now upstream here: https://github.com/libguestfs/libguestfs/commit/501561742b43edb044072ec85048bb35b0b8201a and I have made the corresponding change to libguestfs.spec (but not pushed it yet).
Yes, I forgot to mention that I wasn't sure about the link vs. global scope. Thanks !
Upstream for a while now and well tested, so closing.