On Wed, Oct 29, 2025 at 5:23 PM Stephen Morris <[email protected]> wrote:

    [...]
    The only thing I've between yesterday when the device wasn't
    active is I did a system update yesterday which updated the kernel
    to 6.17.4 so I am now running on that instead of 6.16.10, but I
    checked the 6.16.10 config yesterday and it was showing the it had
    support for the mt7925 chipset, and the driver and firmware were
    already installed. From what I've read support for the chipset was
    patched into the kernel with kernel 6.7, so it should have been
    active in later kernels.
    The other thing I did yesterday was turn off random mac address
    generation for wifi in the NetworkManager config, which I had also
    done earlier as a trial which did nothing relative to the issue.
    As I said a bit earlier, now that it seems to be active I'll keep
    monitoring it and see if it stops continually connecting and
    disconnecting, but irrespective of that thankyou for all your help.


Sometimes wifi hardware gets into a "confused" state. Because wifi is often enabled even when a system is "off" to support wake-on-wlan, and vendors try to reduce startup time, they tend to avoid delays resetting wifi hardware.  It is hard to know what is needed to force a reset.

Hi George,
    With the device continually resetting plugged into the usb port I was using it from under Windows, I changed to a different usb port which then provided what seemed to be a stable state in Fedora, but when I boot into Windows from a cold start, the Windows boot continually crashes until I unplug the device. I switched the device to a third usb port which then enabled the device to function again under Windows, but now when I boot into Fedora the wifi connection is now connecting and disconnecting continually again. What I don't know at the moment is I have both ethernet and wifi active and I am using ethernet for my network connection (just on that front I thought for the network definitions if one specified a priority of 0 and one specified a priority of 10 the definition with the priority of 0 is the one that is used first, is that correct? I'm just asking because it seems to be the other way around).

I have found an article on the net (not sure I can find it again) from another linux user who was using a device with the same chipset under Arch linux and he was getting the same connection/disconnection issues I'm getting, and he tried a lot of different things to try to get the device to work without success. He eventually changed NetworkManager to use iwd as its back end and then to stop the connect/disconnect he had to add threshold parameters of -90 (not sure what the means) for each of the 3 bands, but he was still not happy that after doing all that it was still not working properly (not sure what he was still not happy with).

Just looking at the popup messages shown on the desktop for the disconnection/connection it seems to me the disconnection is being caused because potentially wpa_supplicant is trying to authenticate to the device and that authentication is timing out (I have also seen those messages in dmesg and journalctl as well). If I switch to iwd as the network backend with that stop the wpa_supplicant authorization issue?

regards,


--
George N. White III



BEGIN:VCARD
VERSION:4.0
N:Morris;Stephen;;;
FN:Steve
NICKNAME:Steve
EMAIL;PREF=1;TYPE=home:[email protected]
TZ:Australia/Melbourne
END:VCARD
-- 
_______________________________________________
users mailing list -- [email protected]
To unsubscribe send an email to [email protected]
Fedora Code of Conduct: 
https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/[email protected]
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Reply via email to