• Re: Dual wifi connections in Bookworm

    From The Natural Philosopher@3:770/3 to bp@www.zefox.net on Sat Nov 30 09:07:38 2024
    On 29/11/2024 23:13, bp@www.zefox.net wrote:
    The Natural Philosopher <tnp@invalid.invalid> wrote:
    On 27/11/2024 20:22, bp@www.zefox.net wrote:
    It turns out that letting the system stand for ~4 hrs reproduced some
    of the problems. Ping times with both interfaces up eventually rose to
    7-10 ms but didn't fail completely. Several of the ssh connections
    open from the Pi5 to other hosts either froze or disconnected.

    That's exactly what I got with an inadequate PSU.


    Since a software update this morning it appears that both interfaces
    are co-existing without problems. Re-checking the supply voltage
    reveals 5.05 volts, 10 mV below the value seen earlier, measured on
    the GPIO header.

    I got sag when USB was drawing current I think.

    That's not to claim the problem is resolved: It took about four hours
    to misbehave in the previous instance, now it might merely take longer.

    Fingers crossed.

    uname -a now reports
    Linux raspberrypi 6.6.62+rpt-rpi-2712 #1 SMP PREEMPT Debian 1:6.6.62-1+rpt1 (2024-11-25) aarch64 GNU/Linux

    Thanks for writing,

    bob prohaska

    Reminds me of that old BT advert where an old lady discovers her phone
    isn't working. Cut to engineers in vans climbing poles, remaking
    connections and so on, cut to old lady lifting the handset 'Oh, its
    fixed itself!'...



    --
    β€œIt is dangerous to be right in matters on which the established
    authorities are wrong.”

    ― Voltaire, The Age of Louis XIV

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From bp@www.zefox.net@3:770/3 to The Natural Philosopher on Sat Nov 30 17:41:53 2024
    The Natural Philosopher <tnp@invalid.invalid> wrote:
    On 29/11/2024 23:13, bp@www.zefox.net wrote:
    The Natural Philosopher <tnp@invalid.invalid> wrote:
    On 27/11/2024 20:22, bp@www.zefox.net wrote:
    It turns out that letting the system stand for ~4 hrs reproduced some
    of the problems. Ping times with both interfaces up eventually rose to >>>> 7-10 ms but didn't fail completely. Several of the ssh connections
    open from the Pi5 to other hosts either froze or disconnected.

    That's exactly what I got with an inadequate PSU.


    Since a software update this morning it appears that both interfaces
    are co-existing without problems. Re-checking the supply voltage
    reveals 5.05 volts, 10 mV below the value seen earlier, measured on
    the GPIO header.

    I got sag when USB was drawing current I think.

    That's not to claim the problem is resolved: It took about four hours
    to misbehave in the previous instance, now it might merely take longer.

    Fingers crossed.

    uname -a now reports
    Linux raspberrypi 6.6.62+rpt-rpi-2712 #1 SMP PREEMPT Debian 1:6.6.62-1+rpt1 (2024-11-25) aarch64 GNU/Linux


    Through this morning the networking functioned normally. Then I noticed a
    batch of new system updates. After installing and rebooting the trouble returned, slightly worse. Now the internal wifi simply fails to connect.
    The external usb-wifi adapter works as expected. I'm again seeing failures reported for networkmanager-wait-online, but don't understand what they mean.

    Looking through journalctl output, filtered for wlan0, I see a torrent of output, some of which states:

    Nov 30 09:19:00 raspberrypi NetworkManager[852]: <warn> [1732987140.7820] device (wlan0): Activation: (wifi) association took too long
    Nov 30 09:19:00 raspberrypi NetworkManager[852]: <info> [1732987140.7821] device (wlan0): state change: config -> failed (reason 'no-secrets', sys-iface-state: 'managed')
    Nov 30 09:19:00 raspberrypi NetworkManager[852]: <warn> [1732987140.7826] device (wlan0): Activation: failed for connection 'Wi-Fi connection 1'
    Nov 30 09:19:00 raspberrypi NetworkManager[852]: <info> [1732987140.7828] device (wlan0): state change: failed -> disconnected (reason 'none', sys-iface-state: 'managed')

    The mismash of reasons looks inconsistent to me.

    The only changes have been disconnecting and reconnecting the usb-wifi
    dongle, right now it's connected and the gpio is showing 5.08 volts. I
    suspect the small fluctuations seen reflect line voltage drift more
    than load.

    In any case, I've now navigated successfully from the frying pan back
    to the fire 8-)

    Thanks for reading,

    bob prohaska

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)
  • From Lawrence D'Oliveiro@3:770/3 to The Natural Philosopher on Sat Nov 30 23:26:44 2024
    On Fri, 29 Nov 2024 13:27:40 +0000, The Natural Philosopher wrote:

    The point being wifi is as shit as coaxial ethernet was, and should be avoided if at all possible

    I find it hard to disagree. ;)

    I have Ethernet connections running into both my bedroom and living room.
    The wi-fi is fine for basic web browsing (not including watching videos),
    but anything more traffic-intensive, and I plug the nearest handy cable
    into my laptop.

    --- SoupGate-Win32 v1.05
    * Origin: Agency HUB, Dunedin - New Zealand | Fido<>Usenet Gateway (3:770/3)