Comment 75 for bug 1752772

Revision history for this message
craig p hicks (craigphicks) wrote :

I've been having the same problem after resume. However, although de-modprobe-ing and re-modprobe-ing enables the 'eth0' to be renamed as 'enp2s0', finally it fizzles out with

Jul 06 02:13:24 craig-desktop NetworkManager[5649]: <debug> [1530868404.9510] device[0x2840aa0] (enp2s0): emit RECHECK_ASSUME signal
Jul 06 02:13:24 craig-desktop NetworkManager[5649]: <debug> [1530868404.9510] device[0x2840aa0] (enp2s0): ip4-config: update (commit=0, routes-full-sync=0, new-config=0x285f8f0)
Jul 06 02:13:24 craig-desktop NetworkManager[5649]: <debug> [1530868404.9511] device[0x2840aa0] (enp2s0): ip6-config: update (commit=0, routes-full-sync=0, new-config=0x282a440)
Jul 06 02:13:24 craig-desktop NetworkManager[5649]: <debug> [1530868404.9511] device[0x2840aa0] (enp2s0): device has no existing configuration
Jul 06 02:13:24 craig-desktop NetworkManager[5649]: <debug> [1530868404.9511] manager: (enp2s0): can't assume; no connection

and connectivity fails. Only a reboot fixes it.

Version:
Linux craig-desktop 4.15.0-24-generic #26~16.04.1-Ubuntu SMP Fri Jun 15 14:35:08 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux

Past similar reports related to r8169 over the years:

2017 - https://askubuntu.com/a/893823/723997

2011 - https://ubuntuforums.org/showthread.php?t=1661489
 - This 2011 one is interesting because the author points out that ::

   # lspci | grep Realtek
   02:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411
    PCI Express Gigabit Ethernet Controller (rev 0c)

   so he replaced r8169 with r8168 and succeeded.

On the other hand, the following set of reports seems to indicate that it is independent of the driver, because of multiple cases where unloading and loading the driver solved the problem although the drivers are different:

2016 - https://forum.ubuntuusers.de/topic/nach-16-10-installation-kein-internet-mehr-nac/2/#post-8652643
     - https://ubuntuforums.org/showthread.php?t=2355314

In conclusion, there might be a deeper bug which is appearing and disappearing with different versions depending upon the random module layout in the compile.