© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
bbfa235
demo site
(Get the code!)
Same here with 16.04.2 (4.8.0-58-generic x86_64, no VM) and Realtek8111 (PCI 10ec:8168 rev 06). Worked before with 12.04, now it's shutting down the interface every few days.
Regarding the Networkmanager restart: Not for me, the PID is still the same.
Maybe the timing from the latest event may help:
Jul 23 13:51:19 braindead5 dhclient[26468]: DHCPREQUEST of 192.168.0.5 on enp8s0 to 192.168.0.1 port 67 (xid=0x75dc0193)
Jul 23 13:51:19 braindead5 dhclient[26468]: DHCPACK of 192.168.0.5 from 192.168.0.1
Jul 23 13:51:19 braindead5 dhclient[26468]: bound to 192.168.0.5 -- renewal in 17447 seconds.
So the IP is valid for more than 4h. But about an hour later:
Jul 23 14:57:51 braindead5 avahi-daemon[907]: Withdrawing address record for 192.168.0.5 on enp8s0.
Jul 23 14:57:51 braindead5 avahi-daemon[907]: Leaving mDNS multicast group on interface enp8s0.IPv4 with address 192.168.0.5.
Jul 23 14:57:51 braindead5 avahi-daemon[907]: Interface enp8s0.IPv4 no longer relevant for mDNS.
Jul 23 14:57:51 braindead5 org.kde.
Jul 23 14:57:56 braindead5 kernel: [1879701.611525] nfs: server 192.168.0.1 not responding, timed out
The interface was shut down (no IP, only visible with ifconfig -a), but it had link state (green LED). There were no other messages regarding Networkmanager etc. at that time.
It's hard to tell if the NFS problems are caused by the avahi-stuff or both share the same root cause. In my experience, NFS usually "finds" missing connectivity very fast, so it should complain before avahi... Also on all other occurences the avahi message was before NFS.
I've now stopped the avahi-daemon. If it happens again, I will look if dhclient runs and if there are increasing RX packets and visible broadcast data from other hosts at the dead interface.