Please read this before reporting a bug:
https://wiki.archlinux.org/title/Bug_reporting_guidelines
Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.
REPEAT: Do NOT report bugs for outdated packages!
https://wiki.archlinux.org/title/Bug_reporting_guidelines
Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.
REPEAT: Do NOT report bugs for outdated packages!
FS#79020 - [linux] Realtek RTL8111/8168/8411 non-funcitoning on 6.4.1 and later
Attached to Project:
Arch Linux
Opened by twixt (twixt) - Sunday, 09 July 2023, 06:35 GMT
Last edited by Toolybird (Toolybird) - Thursday, 24 August 2023, 22:31 GMT
Opened by twixt (twixt) - Sunday, 09 July 2023, 06:35 GMT
Last edited by Toolybird (Toolybird) - Thursday, 24 August 2023, 22:31 GMT
|
DetailsDescription:
After installing kernel version 6.4.1 and 6.4.2 the drivers for the Realtek GigE RTL8111/8168/8411 would cease to function approximately 1-2 hours after boot. I've experienced it as early as half an hour after boot. Downgrading back down to 6.3.9 and below, the issue never comes up again. No issues with the latest kernel for Ethernet controllers from Intel, both I225-V and I219-LM. First noticed when network went down, and any attempts to fix did not work. Plugging in a USB NIC and taking the link down would cause traffic to flow normally, but RTNETLINK would always fail to find the device to attempt to bring the link back up. After reboot, once net went down, trying to run "ip a" to see if DHCP address was still available would cause a hang for around 5 minutes. Following this, the DHCP route and address would be gone from the interface, and taking it down/up would result in behavior previously mentioned. Following this, I had installed the 6.4.2 kernel to repeat issues. Unfortunately, I do not have logs since I was more focused on bringing the machine back online to continue work. I can attempt to replicate and get logs if really necessary, but this is a machine in use almost every day. Forcing it into downtime isn't ideal. |
This task depends upon
Closed by Toolybird (Toolybird)
Thursday, 24 August 2023, 22:31 GMT
Reason for closing: Upstream
Additional comments about closing: If still happening with latest kernels, please report it upstream. This is not an Arch packaging issue.
Thursday, 24 August 2023, 22:31 GMT
Reason for closing: Upstream
Additional comments about closing: If still happening with latest kernels, please report it upstream. This is not an Arch packaging issue.
[1] https://wiki.archlinux.org/title/Kernel#Debugging_regressions
[2] https://linux-regtracking.leemhuis.info/regzbot/mainline/
[3] https://lore.kernel.org/netdev/c3465166-f04d-fcf5-d284-57357abb3f99%40freenet.de/
I'll try and schedule some debugging in low-load hours to present there, checking out mainline to see if the issue is still present sounds like a plan.
Will update once more after leaving system up for a day or more, then we can be certain this is resolved.