Hi,

i managed to workaround this issue by setting one interface to ipv4 dhcp and 
ipv6 static without configuration and another identical interface (in terms of 
vswitch and vlan) with ipv6 dhcp and ipv4 static without configuration.

This results in both interfaces getting their dhcp ip. Hope this gets fixed 
with a future pve update :)


Best regards


Felix Brucker


________________________________
Von: pve-user <[email protected]> im Auftrag von Felix Brucker 
<[email protected]>
Gesendet: Samstag, 29. Oktober 2016 16:55
An: PVE User List
Betreff: [PVE-User] lxc: could not get a link-local address

Hi,


i noticed this problem a few days ago when my lxc container running ubuntu 16 
and debian 8 where suddenly unable to reach over ipv6, their address was gone, 
and their link local address too.


I have verified this does not cohere with any changes made by myself by 
installing a fresh copy of the latest proxmox iso on this machine, issuing 
dist-upgrade, reboot and downloading the latest ubuntu lxc container available 
from "Templates".


The lxc container config is only changed to add one single network connection 
which is a standard linux bridge with vlan support. Also ipv4 and ipv6 are both 
set to "dhcp".


Upon starting the container it hangs with "waiting [..] raise network 
interfaces".


When its loaded i only have a ipv4 address and no link-local or ipv6 address.


I noticed /proc/sys/net/ipv6/conf/veth102i0/disable_ipv6 is set to 1 by default 
for every veth interface regardless if ipv6 is enabled or not, which results in 
only "iface eth0 inet dhcp" without "iface eth0 inet6 dhcp". Setting this as 
soon as possible after the launch of the container to zero does result in the 
/etc/network/interfaces to be updated to include the necessary line, but it 
still doesnt work.


In this situation, when issuing "/etc/init.d/networking restart" inside the 
container outputs:


Oct 29 14:37:34 test ifup[536]: DHCPACK of 10.223.4.221 from 10.223.0.1
Oct 29 14:37:34 test ifup[536]: RTNETLINK answers: File exists
Oct 29 14:37:34 test ifup[536]: bound to 10.223.4.221 -- renewal in 98 seconds.
Oct 29 14:37:40 test ifup[536]: Could not get a link-local address
Oct 29 14:37:40 test ifup[536]: Failed to bring up eth0.
Oct 29 14:37:40 test systemd[1]: networking.service: Main process exited, 
code=exited, status
Oct 29 14:37:40 test systemd[1]: Failed to start Raise network interfaces.
Oct 29 14:37:40 test systemd[1]: networking.service: Unit entered failed state.
Oct 29 14:37:40 test systemd[1]: networking.service: Failed with result 
'exit-code'.


Anybody has experienced the same?

Is this a know issue?


Best regards


Felix Brucker

_______________________________________________
pve-user mailing list
[email protected]
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
_______________________________________________
pve-user mailing list
[email protected]
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user

Reply via email to