 |
info.zip (17,023 bytes) |
|
|
I'm seeing similar issues with the r8169 driver. I filed https://bugzilla.redhat.com/show_bug.cgi?id=1753350 |
 |
What is the device ID [xxxx:yyyy] as reported by 'lspci -nn' ? |
|
|
lspci and dmesg are in the file attached. Here it is again: 03:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 01) I guess this is the one because the name is enp3s0 |
|
|
Sorry. Here it is with -nn option: 03:00.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [10ec:8168] (rev 01) |
|
|
With that device ID [10ec:8168], I can confirm it is supported by the r8169 driver: $ grep -i 10ec /lib/modules/3.10.0-1062.el7.x86_64/modules.alias | grep 8168 alias pci:v000010ECd00008168sv*sd*bc*sc*i* r8169 Is this module loaded? 'lsmod' If not, can you load it? 'modprobe r8169' |
|
|
I don't think the problem is missing module. The device is listed in "ip link show" with the new kernel. It just shows as if it is not connected. The LEDs are off. When I reboot to old kernel without touching the cable the LEDs light up and the card works. But I'm not an expert there. Here the output of these commands: grep 8169 /lib/modules/3.10.0-1062.1.1.el7.x86_64/modules.alias|grep -i 10ec alias pci:v000010ECd00008169sv*sd*bc*sc*i* r8169 alias pci:v000010ECd00008168sv*sd*bc*sc*i* r8169 alias pci:v000010ECd00008167sv*sd*bc*sc*i* r8169 alias pci:v000010ECd00008161sv*sd*bc*sc*i* r8169 alias pci:v000010ECd00008136sv*sd*bc*sc*i* r8169 alias pci:v000010ECd00008129sv*sd*bc*sc*i* r8169 alias pci:v000010ECd00002600sv*sd*bc*sc*i* r8169 alias pci:v000010ECd00002502sv*sd*bc*sc*i* r8169 # lsmod.|grep 8169 r8169 84623 0 |
|
|
OK, the r8169 driver is loaded. The 7.6 kernel has r8169 version 6.020.00-NAPI. I don't know the version number of this driver in the 7.7 kernel. Presumably it is different. ELRepo has kmod-r8168 (NOT r8169) that supports the device ID [10EC:8168]. https://elrepo.org/linux/elrepo/el7/x86_64/RPMS/kmod-r8168-8.046.00-1.el7_5.elrepo.x86_64.rpm Its version is 8.046.00-NAPI. Try installing it and see if that works. |
|
|
installing the above rpm did the job. lsmod shows r8168 loaded, r8169 is not loaded now, the network card works fine. Thank you very much! Is this going to be investigated or maybe reported to redhat? I can't report to redhat because I don't have redhat linux installed, but the problem could be in the module version they used. # rpm -qa|grep 816 kmod-r8168-8.046.00-1.el7_5.elrepo.x86_64 # uname -r 3.10.0-1062.1.1.el7.x86_64 # ip link show dev enp3s0 4: enp3s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP mode DEFAULT group default qlen 1000 link/ether 00:1d:7d:d0:2a:52 brd ff:ff:ff:ff:ff:ff # lsmod |grep 816 r8168 552036 0 |
|
|
You can safely assume CentOS 7 kernels are binary compatible with the RHEL counterpart. So, yes, I suggest you file a bug report with them. |
|
|
A similar problem led me here. I just wanted to add some info in case it assists. This appears to be a regressive problem, and may have been discussed previously here: https://bugzilla.redhat.com/show_bug.cgi?id=1650984 The pertinent bits being: ---------------------------------------------------- lsmod reveals that module r8169 is loaded, but module realtek is not. # rmmod r8169; modprobe r8169 # systemctl restart network.service fixes the problem temporarily, but this does not survive a reboot. # echo realtek >/etc/modules-load.d/realtek.conf fixes the problem and survives a reboot. ---------------------------------------------------- h/t to Crystal Cowboy @ http://scientificlinuxforum.org Further reading suggests the fix is not permanent but by applying it at every boot it solves the problem for the time being. |
 |
@crnz, that helps a lot!!! @toracat, what do you think of this https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?id=11287b693d03830010356339e4ceddf47dee34fa ? |
|
|
@pgreco, sounds like a useful patch. Can go to the plus kernel and the kmod. |
|
|
Turns out that patch was added to the kernel as of 3.10.0-1024.el7. Therefore it is in the CentOS 7.7 kernel. |
|
|
the same happened with me:-( |
- Anonymous
