• Strange cabled network fault when rebooting from Windows (any version)

    From Java Jive@2:250/1 to All on Saturday, November 15, 2025 14:39:35
    Subject: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    In case it's helpful to anyone else ...

    I have some Dell Precision M6800s & M6700s dual-booting via Grub between various versions of Windows and Ubuntu 24, all cabled to the LAN. If I
    choose to move between any version of Windows and Ubuntu by rebooting,
    Ubuntu 24 cannot make a reliable cabled LAN connection, it keeps trying
    to connect but never succeeds. However, if I choose to move between any version of Windows and Ubuntu by shutting down the PC in between, Ubuntu
    24 'just works', no networking problem at all. Reboot from U24 and go
    back into U24, the situation is unchanged - if it was working before
    the reboot, it's working now, if it wasn't working before the reboot, it
    isn't working now. Reboot from U24 and go back into Windows, no network problems in Windows at all. Reboot from Windows and go back into any
    version of Windows, no network problems at all.

    In summary, when rebooting ...
    Ubuntu 24 -> Ubuntu 24 No change
    Ubuntu 24 -> Windows No problem
    Windows -> Ubuntu 24 U24 cannot make cabled LAN connection
    Windows -> Windows No problem

    It seems to me from this that the cabled network hardware is not being properly initialised when booting Ubuntu 24. The cure is simply to
    actually power down the PC before launching Ubuntu 24

    --

    Fake news kills!

    I may be contacted via the contact address given on my website: www.macfh.co.uk

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Andy Burns@2:250/1 to All on Saturday, November 15, 2025 15:04:44
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    Java Jive wrote:

    It seems to me from this that the cabled network hardware is not being properly initialised when booting Ubuntu 24.  The cure is simply to actually power down the PC before launching Ubuntu 24

    I had some older Dells (Precision and Dimension) which disabled the NIC
    if it wasn't physically connected at the time it booted Windows, there
    was a utility to re-enable it, probably doesn't exist for Linux (though
    Dell were pretty good with their Linux support via Fedora last time I
    kept up with it).

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Air Applewood, The Linux Gateway to the UK & Eire (2:250/1@fidonet)
  • From Java Jive@2:250/1 to All on Saturday, November 15, 2025 15:17:26
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    On 2025-11-15 15:04, Andy Burns wrote:
    Java Jive wrote:

    It seems to me from this that the cabled network hardware is not being
    properly initialised when booting Ubuntu 24.  The cure is simply to
    actually power down the PC before launching Ubuntu 24

    I had some older Dells (Precision and Dimension) which disabled the NIC
    if it wasn't physically connected at the time it booted Windows, there
    was a utility to re-enable it, probably doesn't exist for Linux (though
    Dell were pretty good with their Linux support via Fedora last time I
    kept up with it).

    Interesting piece of info ... However, during the testing these were in E-Dock stations - from which they are only undocked when I travel, so weren't undocked during testing - so seemingly there wouldn't be any
    reason for either OS to disable the netcard, as it was permanently
    connected during the testing.

    --

    Fake news kills!

    I may be contacted via the contact address given on my website: www.macfh.co.uk


    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From vallor@2:250/1 to All on Saturday, November 15, 2025 16:17:58
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    At Sat, 15 Nov 2025 15:17:26 +0000, Java Jive <java@evij.com.invalid>
    wrote:

    On 2025-11-15 15:04, Andy Burns wrote:
    Java Jive wrote:

    It seems to me from this that the cabled network hardware is not
    being properly initialised when booting Ubuntu 24.  The cure is
    simply to actually power down the PC before launching Ubuntu 24

    I had some older Dells (Precision and Dimension) which disabled the
    NIC if it wasn't physically connected at the time it booted
    Windows, there was a utility to re-enable it, probably doesn't
    exist for Linux (though Dell were pretty good with their Linux
    support via Fedora last time I kept up with it).

    Interesting piece of info ... However, during the testing these were
    in E-Dock stations - from which they are only undocked when I
    travel, so weren't undocked during testing - so seemingly there
    wouldn't be any reason for either OS to disable the netcard, as it
    was permanently connected during the testing.

    Weird problem. If you want to troubleshoot, you might get
    it into failure mode, then run ethtool on the ethernet device
    to see if auto-negotiation is happening properly, if the
    Linux driver thinks link is detected, and so forth.

    What chipset does your Ethernet use, and is it part
    of the dock? It appears that Intel driver modules
    have a "debug" option that can be set, not sure if
    that helps...

    --
    -v System76 Thelio Mega v1.1 x86_64 NVIDIA RTX 3090Ti 24G
    OS: Linux 6.17.8 D: Mint 22.2 DE: Xfce 4.18
    NVIDIA: 580.105.08 Mem: 258G
    "Ethernet n.: something used to catch the etherbunny."

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: The Unnamed Newswriter (2:250/1@fidonet)
  • From Java Jive@2:250/1 to All on Saturday, November 15, 2025 17:19:49
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    On 2025-11-15 16:17, vallor wrote:
    At Sat, 15 Nov 2025 15:17:26 +0000, Java Jive <java@evij.com.invalid>
    wrote:

    On 2025-11-15 15:04, Andy Burns wrote:
    Java Jive wrote:

    It seems to me from this that the cabled network hardware is not
    being properly initialised when booting Ubuntu 24.  The cure is
    simply to actually power down the PC before launching Ubuntu 24

    I had some older Dells (Precision and Dimension) which disabled the
    NIC if it wasn't physically connected at the time it booted
    Windows, there was a utility to re-enable it, probably doesn't
    exist for Linux (though Dell were pretty good with their Linux
    support via Fedora last time I kept up with it).

    Interesting piece of info ... However, during the testing these were
    in E-Dock stations - from which they are only undocked when I
    travel, so weren't undocked during testing - so seemingly there
    wouldn't be any reason for either OS to disable the netcard, as it
    was permanently connected during the testing.

    Weird problem. If you want to troubleshoot, you might get
    it into failure mode, then run ethtool on the ethernet device
    to see if auto-negotiation is happening properly, if the
    Linux driver thinks link is detected, and so forth.

    What chipset does your Ethernet use, and is it part
    of the dock? It appears that Intel driver modules
    have a "debug" option that can be set, not sure if
    that helps...

    Thanks for the suggestions, but now I know what causes it and the simple
    fix, probably I shall just try to remember always to shut down the PC
    before going into Ubuntu.

    Back now in Windows, the cabled net card is described as 'Intel Ethernet Connection I217-LM', and this doesn't change when docking and undocking
    the PC. Similarly, another PC's net card, again regardless of whether
    it is docked or not, is described as 'Intel 82579LM Gigabit Network Connection' (I haven't tested the Ubuntu 24 problem on that one, as it
    wasn't the PC I noticed the problem on). Given this, I think the dock
    is merely an alternative connection between the PC's network interface
    and the outside world, and doesn't have it's own separate network
    interface in the dock itself.

    --

    Fake news kills!

    I may be contacted via the contact address given on my website: www.macfh.co.uk


    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Dan Purgert@2:250/1 to All on Saturday, November 15, 2025 17:27:04
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    On 2025-11-15, Java Jive wrote:
    In case it's helpful to anyone else ...

    I have some Dell Precision M6800s & M6700s dual-booting via Grub between various versions of Windows and Ubuntu 24, all cabled to the LAN. If I choose to move between any version of Windows and Ubuntu by rebooting,

    Win10, 11 have that "FastBoot" thing that mucks with hardware releases
    on "reboot" (win basically goes into hibernate).

    Not sure if / how that'd work *between* windows versions (e.g. you had
    both 10,11 installed) though ...


    --
    |_|O|_|
    |_|_|O| Github: https://github.com/dpurgert
    |O|O|O| PGP: DDAB 23FB 19FA 7D85 1CC1 E067 6D65 70E5 4CE7 2860

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From VanguardLH@2:250/1 to All on Saturday, November 15, 2025 17:27:07
    Subject: Re: Strange cabled network fault when rebooting from Windows (any version) into Ubuntu 24
    Keywords: VanguardLH,VLH

    Java Jive <java@evij.com.invalid> wrote:

    In case it's helpful to anyone else ...

    I have some Dell Precision M6800s & M6700s dual-booting via Grub between various versions of Windows and Ubuntu 24, all cabled to the LAN. If I choose to move between any version of Windows and Ubuntu by rebooting, Ubuntu 24 cannot make a reliable cabled LAN connection, it keeps trying
    to connect but never succeeds. However, if I choose to move between any version of Windows and Ubuntu by shutting down the PC in between, Ubuntu
    24 'just works', no networking problem at all. Reboot from U24 and go
    back into U24, the situation is unchanged - if it was working before
    the reboot, it's working now, if it wasn't working before the reboot, it isn't working now. Reboot from U24 and go back into Windows, no network problems in Windows at all. Reboot from Windows and go back into any version of Windows, no network problems at all.

    In summary, when rebooting ...
    Ubuntu 24 -> Ubuntu 24 No change
    Ubuntu 24 -> Windows No problem
    Windows -> Ubuntu 24 U24 cannot make cabled LAN connection
    Windows -> Windows No problem

    It seems to me from this that the cabled network hardware is not being properly initialised when booting Ubuntu 24. The cure is simply to
    actually power down the PC before launching Ubuntu 24

    Don't know what are the "networking problems". In each scenario, can
    you connect to the internal web server in the cable modem, router, or
    whatever you have upstream of the problematic host? Check if you can do intranet connections before testing if you can get to the Internet.

    Does the upstream web server (cable modem, router) report the same MAC
    each time you reboot? When you have network problems, does it even see
    the MAC address for your problematic host? I'm wondering if outside
    your problematic host if anything else sees it.

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Usenet Elder (2:250/1@fidonet)
  • From Java Jive@2:250/1 to All on Saturday, November 15, 2025 17:51:25
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    On 2025-11-15 17:27, Dan Purgert wrote:
    On 2025-11-15, Java Jive wrote:
    In case it's helpful to anyone else ...

    I have some Dell Precision M6800s & M6700s dual-booting via Grub between
    various versions of Windows and Ubuntu 24, all cabled to the LAN. If I
    choose to move between any version of Windows and Ubuntu by rebooting,

    Win10, 11 have that "FastBoot" thing that mucks with hardware releases
    on "reboot" (win basically goes into hibernate).

    Not sure if / how that'd work *between* windows versions (e.g. you had
    both 10,11 installed) though ...

    And I've disabled it anyway, because it's not compatible with using
    imaging software to back up the OS.

    --

    Fake news kills!

    I may be contacted via the contact address given on my website: www.macfh.co.uk


    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From VanguardLH@2:250/1 to All on Saturday, November 15, 2025 17:56:58
    Subject: Re: Strange cabled network fault when rebooting from Windows (any version) into Ubuntu 24
    Keywords: VanguardLH,VLH

    Java Jive <java@evij.com.invalid> wrote:

    On 2025-11-15 16:17, vallor wrote:
    At Sat, 15 Nov 2025 15:17:26 +0000, Java Jive <java@evij.com.invalid>
    wrote:

    On 2025-11-15 15:04, Andy Burns wrote:
    Java Jive wrote:

    It seems to me from this that the cabled network hardware is not
    being properly initialised when booting Ubuntu 24.  The cure is
    simply to actually power down the PC before launching Ubuntu 24

    I had some older Dells (Precision and Dimension) which disabled the
    NIC if it wasn't physically connected at the time it booted
    Windows, there was a utility to re-enable it, probably doesn't
    exist for Linux (though Dell were pretty good with their Linux
    support via Fedora last time I kept up with it).

    Interesting piece of info ... However, during the testing these were
    in E-Dock stations - from which they are only undocked when I
    travel, so weren't undocked during testing - so seemingly there
    wouldn't be any reason for either OS to disable the netcard, as it
    was permanently connected during the testing.

    Weird problem. If you want to troubleshoot, you might get
    it into failure mode, then run ethtool on the ethernet device
    to see if auto-negotiation is happening properly, if the
    Linux driver thinks link is detected, and so forth.

    What chipset does your Ethernet use, and is it part
    of the dock? It appears that Intel driver modules
    have a "debug" option that can be set, not sure if
    that helps...

    Thanks for the suggestions, but now I know what causes it and the simple fix, probably I shall just try to remember always to shut down the PC
    before going into Ubuntu.

    Back now in Windows, the cabled net card is described as 'Intel Ethernet Connection I217-LM', and this doesn't change when docking and undocking
    the PC. Similarly, another PC's net card, again regardless of whether
    it is docked or not, is described as 'Intel 82579LM Gigabit Network Connection' (I haven't tested the Ubuntu 24 problem on that one, as it wasn't the PC I noticed the problem on). Given this, I think the dock
    is merely an alternative connection between the PC's network interface
    and the outside world, and doesn't have it's own separate network
    interface in the dock itself.

    Were doing a cold or warm reboot between switching OSes? A cold reboot
    has the CPU send a reset signal to all hardware to ensure the hardware
    is at a known starting state. A warm boot doesn't. For a cold reboot,
    (power cycle) the keyboard's LEDs should flash letting you know it got
    the reset.

    I've had DSP modem card (dial-up) that would get hung, and a warm reboot
    aka restart wouldn't cure it, but a cold boot with its reset got the
    modem back into a known and usable state, so the drivers could then
    communicate to the modem.

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Usenet Elder (2:250/1@fidonet)
  • From Java Jive@2:250/1 to All on Saturday, November 15, 2025 18:00:12
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    On 2025-11-15 17:27, VanguardLH wrote:

    Don't know what are the "networking problems". In each scenario, can
    you connect to the internal web server in the cable modem, router, or whatever you have upstream of the problematic host? Check if you can do intranet connections before testing if you can get to the Internet.

    I can't connect to anything in the problem scenario of rebooting from
    Windows into Ubuntu 24, I can connect to everything when there is no
    problem.

    --

    Fake news kills!

    I may be contacted via the contact address given on my website: www.macfh.co.uk


    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Simon@2:250/1 to All on Saturday, November 15, 2025 19:15:12
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    On 15/11/2025 14:39, Java Jive wrote:
    It seems to me from this that the cabled network hardware is not being properly initialised when booting Ubuntu 24.  The cure is simply to actually power down the PC before launching Ubuntu 24

    You can try this trick to see if it connects without rebooting Ubuntu:

    <https://i.postimg.cc/q7sgR5M3/2025-11-15-19-05-40.png>

    This is all done in Ubuntu.


    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: To protect and to server (2:250/1@fidonet)
  • From Carlos E.R.@2:250/1 to All on Saturday, November 15, 2025 19:43:33
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    On 2025-11-15 15:39, Java Jive wrote:
    In case it's helpful to anyone else ...

    I have some Dell Precision M6800s & M6700s dual-booting via Grub between various versions of Windows and Ubuntu 24, all cabled to the LAN.  If I choose to move between any version of Windows and Ubuntu by rebooting, Ubuntu 24 cannot make a reliable cabled LAN connection, it keeps trying
    to connect but never succeeds.  However, if I choose to move between any version of Windows and Ubuntu by shutting down the PC in between, Ubuntu
    24 'just works', no networking problem at all.  Reboot from U24 and go
    back into U24, the situation is unchanged  -  if it was working before
    the reboot, it's working now, if it wasn't working before the reboot, it isn't working now.  Reboot from U24 and go back into Windows, no network problems in Windows at all.  Reboot from Windows and go back into any version of Windows, no network problems at all.

    In summary, when rebooting ...
        Ubuntu 24 -> Ubuntu 24    No change
        Ubuntu 24 -> Windows      No problem
        Windows -> Ubuntu 24      U24 cannot make cabled LAN connection
        Windows -> Windows        No problem

    It seems to me from this that the cabled network hardware is not being properly initialised when booting Ubuntu 24.  The cure is simply to actually power down the PC before launching Ubuntu 24

    Similar problems have been described long before.

    Booting Windows -> Linux, worked.
    Booting Linux from scratch, did not work.

    The reason was the Linux driver, developed by reverse engineering
    probably, did not initialize the network hardware properly, they did not
    know something that had to be done.

    After they knew this, the developer wrote something else to the driver,
    and it worked.

    In your case, Windows leaves the eth hardware in a state the Linux
    driver can not handle. Open a bug with the Ubuntu people, somehow.

    --
    Cheers, Carlos.
    ES🇪🇸, EU🇪🇺;

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Air Applewood, The Linux Gateway to the UK & Eire (2:250/1@fidonet)
  • From Anssi Saari@2:250/1 to All on Saturday, November 15, 2025 21:43:41
    Subject: Re: Strange cabled network fault when rebooting from Windows (any version) into Ubuntu 24

    Dan Purgert <dan@djph.net> writes:

    Win10, 11 have that "FastBoot" thing that mucks with hardware releases
    on "reboot" (win basically goes into hibernate).

    No. Reboot is always reboot, Windows would be completely useless without
    that. "FastBoot" aka fast startup happens when shutting down if not
    disabled. And it's hibernate without hibernating apps so fairly useless.

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: An impatient and LOUD arachnid (2:250/1@fidonet)
  • From Java Jive@2:250/1 to All on Saturday, November 15, 2025 23:26:40
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    On 2025-11-15 19:15, Simon wrote:
    On 15/11/2025 14:39, Java Jive wrote:
    It seems to me from this that the cabled network hardware is not being
    properly initialised when booting Ubuntu 24.  The cure is simply to
    actually power down the PC before launching Ubuntu 24

    You can try this trick to see if it connects without rebooting Ubuntu:

    <https://i.postimg.cc/q7sgR5M3/2025-11-15-19-05-40.png>

    This is all done in Ubuntu.

    I haven't got that settings dialog. Historically my Ubuntu has been
    upgraded from 18 to 20 to 22 to 24, and the original 18 was something of
    a hybrid, because I liked the xfce system for its light weight, but
    there aspects of XUbuntu at the time that I didn't like (can't remember
    what now), so I manually installed the xfce system onto a straight
    Ubuntu install and uninstalled some aspects of the original Ubuntu.

    However, while I can't use that dialog, I have spent the evening trying
    to find out more about the problem, which seems to hang on the failure
    of the NetworkManager-wait-online.service. Unfortunately, however, any further information beyond that has proved elusive. The official logs
    are appended below, and running the command-line for it without the -q
    (quiet) option merely displays a countdown of a 30s timeout. Here is
    all the information I was able to obtain without further research online
    which I've yet to do because I'm a bit busy with others things [beware unintended line wrap]:

    root@Charles-P2:home# systemctl --no-pager | grep -i net
    sys-devices-pci0000:00-0000:00:19.0-net-eno1.device
    loaded active plugged Ethernet
    Connection I217-LM
    sys-devices-pci0000:00-0000:00:1c.2-0000:03:00.0-net-wlp3s0.device
    loaded active plugged Wireless 7260
    (Dual Band Wireless-AC 7260 [Wilkins Peak 2])
    sys-subsystem-net-devices-eno1.device
    loaded active plugged Ethernet
    Connection I217-LM
    sys-subsystem-net-devices-wlp3s0.device
    loaded active plugged Wireless 7260
    (Dual Band Wireless-AC 7260 [Wilkins Peak 2])
    ntpsec-systemd-netif.path
    loaded active waiting
    ntpsec-systemd-netif.path
    ● NetworkManager-wait-online.service
    loaded failed failed Network Manager
    Wait Online
    NetworkManager.service
    loaded active running Network Manager
    ntpsec.service
    loaded active running Network Time
    Service
    tor.service
    loaded active exited Anonymizing
    overlay network for TCP (multi-instance-master)
    tor@default.service
    loaded active running Anonymizing
    overlay network for TCP
    xinetd.service
    loaded active running Xinetd A
    Powerful Replacement For Inetd
    network-online.target
    loaded active active Network is Online
    network-pre.target
    loaded active active Preparation for Network
    network.target
    loaded active active Network
    nss-lookup.target
    loaded active active Host and
    Network Name Lookups


    root@Charles-P2:home# systemctl status NetworkManager-wait-online.service
    × NetworkManager-wait-online.service - Network Manager Wait Online
    Loaded: loaded (/usr/lib/systemd/system/NetworkManager-wait-online.service; enabled;
    preset: enabled)
    Active: failed (Result: exit-code) since Sat 2025-11-15 22:59:41
    GMT; 2min 9s ago
    Docs: man:NetworkManager-wait-online.service(8)
    Process: 1391 ExecStart=/usr/bin/nm-online -s -q (code=exited, status=1/FAILURE)
    Main PID: 1391 (code=exited, status=1/FAILURE)
    CPU: 48ms

    Nov 15 22:58:41 Charles-P2 systemd[1]: Starting NetworkManager-wait-online.service - Network Manager Wait Online...
    Nov 15 22:59:41 Charles-P2 systemd[1]:
    NetworkManager-wait-online.service: Main process exited, code=exited, status=1/FAILURE
    Nov 15 22:59:41 Charles-P2 systemd[1]:
    NetworkManager-wait-online.service: Failed with result 'exit-code'.
    Nov 15 22:59:41 Charles-P2 systemd[1]: Failed to start NetworkManager-wait-online.service - Network Manager Wait Online.


    root@Charles-P2:home# journalctl -xeu NetworkManager-wait-online.service
    | tail -30
    Nov 15 22:58:41 Charles-P2 systemd[1]: Starting NetworkManager-wait-online.service - Network Manager Wait Online...
    ░░ Subject: A start job for unit NetworkManager-wait-online.service has begun execution
    ░░ Defined-By: systemd
    ░░ Support: http://www.ubuntu.com/support
    ░░
    ░░ A start job for unit NetworkManager-wait-online.service has begun execution.
    ░░
    ░░ The job identifier is 126.
    Nov 15 22:59:41 Charles-P2 systemd[1]:
    NetworkManager-wait-online.service: Main process exited, code=exited, status=1/FAILURE
    ░░ Subject: Unit process exited
    ░░ Defined-By: systemd
    ░░ Support: http://www.ubuntu.com/support
    ░░
    ░░ An ExecStart= process belonging to unit NetworkManager-wait-online.service has exited.
    ░░
    ░░ The process' exit code is 'exited' and its exit status is 1.
    Nov 15 22:59:41 Charles-P2 systemd[1]:
    NetworkManager-wait-online.service: Failed with result 'exit-code'.
    ░░ Subject: Unit failed
    ░░ Defined-By: systemd
    ░░ Support: http://www.ubuntu.com/support
    ░░
    ░░ The unit NetworkManager-wait-online.service has entered the 'failed' state with result 'exit-code'.
    Nov 15 22:59:41 Charles-P2 systemd[1]: Failed to start NetworkManager-wait-online.service - Network Manager Wait Online.
    ░░ Subject: A start job for unit NetworkManager-wait-online.service has failed
    ░░ Defined-By: systemd
    ░░ Support: http://www.ubuntu.com/support
    ░░
    ░░ A start job for unit NetworkManager-wait-online.service has finished with a failure.
    ░░
    ░░ The job identifier is 126 and the job result is failed.


    --

    Fake news kills!

    I may be contacted via the contact address given on my website: www.macfh.co.uk


    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From VanguardLH@2:250/1 to All on Sunday, November 16, 2025 02:06:48
    Subject: Re: Strange cabled network fault when rebooting from Windows (any version) into Ubuntu 24
    Keywords: VanguardLH,VLH

    Java Jive <java@evij.com.invalid> wrote:

    VanguardLH wrote:

    Don't know what are the "networking problems". In each scenario, can
    you connect to the internal web server in the cable modem, router, or
    whatever you have upstream of the problematic host? Check if you can do
    intranet connections before testing if you can get to the Internet.

    I can't connect to anything in the problem scenario of rebooting from Windows into Ubuntu 24, I can connect to everything when there is no problem.

    While the following are Windows/DOS commands, there is probably
    something very similar for *NIX:

    ipconfig /release
    ipconfig /release6
    ipconfig /flushdns
    ipconfig /renew
    ipconfig /renew6

    That will do a soft reset of the NIC. All bindings are released, and
    then renewed (reinstated) for both IPv4 and IPv6 (the command args
    ending with "6"). The DNS flush is optional. A cold reboot (not a warm
    reset) will send a reset signal to the NIC, and other hardware, to put
    them into a known initialized state. A warm restart does not. Perhaps
    the above (but in *NIX commands) might reinitialize the NIC. If the
    commands work, you could have them run at startup.

    For the above renew commands to work means your computer must be able to
    reach the upstream DHCP server to get its IP binds from there. However,
    if you configured the OS to use static IP addresses then the computer
    isn't getting IP binds from the upstream DHCP server. The same for DNS
    server assignment unless the NICs (IPv4 and IPv6) in the OS config are configured to use specified DNS servers. In the past, I configured the
    hosts to use DHCP IP assigned, but configured the router (in the cable
    modem) to assigned static IP addresses based on the MAC of my hosts. I
    wanted some intranet hosts in one subnet, and others in a different
    subnet, and then blocked one subnet from another (to keep the family
    from getting at my host). By using static IP ranges for its DHCP
    server, I could put the intranet hosts in different subnets. It also
    let me block any access through the router by any host that didn't match
    on the MACs that were allowed in the router. Most users don't change
    the MAC address of their host, but some users will configure the OS to
    report a different MAC address; i.e., the hardware (NIC) uses the same manufacture-assigned MAC, but the OS sends a different MAC address. Now
    that I mentioned MAC addresses, and because the OS will cache the ones
    it finds, maybe try clearing the arp cache, too:

    arp -a

    lists the arp cached MAC addresses, and their IP address at the time the entries were cached. ARP (Address Resolution Protocol) is used in
    Windows, Mac, and Linux to connect to an intranet host on the same
    network.

    To clear the ARP cache, run:

    netsh interface ip delete arpcache

    Think of ARP as an equivalent of the DNS client (local DNS caching) for
    quickly finding hosts, but ARP is intranet versus DNS cache is mostly
    for Intranet hosts. Or think of ARP like a media control method of
    quickly finding intranet hosts versus using the 'hosts' file. I don't
    know if you have multiple subnets, how traffic is controlled between
    them, or not, dynamic IP addressing or static, etc. Some more info on
    ARP is here (and many other online articles):

    https://www.simplehomelab.com/ways-to-clear-arp-cache/

    As new network connections are established, the arp cache repopulates.
    The above is a Windows/DOS command. The ipconfig and netsh commands may
    need to run inside a command shell with administrative priveleges.

    You might want to visit the internal web server in the router to check
    how it is configured to handle addressing and management of your
    intranet hosts.

    You said you cannot connect anywhere which means you also cannot connect
    to the web server inside your router to configure its network management functions, but only after booting into Ubuntu. When booted into
    Windows, and since you can get Internet connections, I'd start digging
    into the config server settings in the router, or maybe just reset the
    router to its default settings if you ever customized them away from the defaults.

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Usenet Elder (2:250/1@fidonet)
  • From VanguardLH@2:250/1 to All on Sunday, November 16, 2025 02:18:21
    Subject: Re: Strange cabled network fault when rebooting from Windows (any version) into Ubuntu 24
    Keywords: VanguardLH,VLH

    Anssi Saari <anssi.saari@usenet.mail.kapsi.fi> wrote:

    Dan Purgert <dan@djph.net> writes:

    Win10, 11 have that "FastBoot" thing that mucks with hardware releases
    on "reboot" (win basically goes into hibernate).

    No. Reboot is always reboot, Windows would be completely useless without that. "FastBoot" aka fast startup happens when shutting down if not
    disabled. And it's hibernate without hibernating apps so fairly useless.

    Actually Fast Starup is a full hibernate (all memory copied into the hyberfil.sys file). The computer then goes into sleep mode. When
    brought out of sleep, the computer resumes from sleep. If, however, the computer ever lost power during sleep, the computer resumes using the
    hibernate file. Because a memory image is reinstated or resumed from a
    Fast Startup mode, there is no re-initialization of hardware. This is
    the same as a Windows restart which is a warm boot.

    Fast Startup causes problems with hardware, because the hardware is not
    sent a reset signal to ensure it is initialized to a known initial
    state. A restart using Fast Startup will not fix hardware problems
    caused by hardware getting into a state the devs did not plan upon in
    their driver, or their driver simply cannot perform an initialize the
    way the hardware was designed. Even using hibernate mode without Fast
    Startup can prevent troubleshooting problems when booting the OS,
    because the memory image gets restored, and you don't want that when troubleshooting. You don't want to get back to the state that had
    problems, but load a new memory image.

    Fast Startup can be enabled or disabled by itself. If hibernate mode is disabled, so is Fast Startup. If you have an SSD for the OS partitions,
    Fast Startup provides little assist in shortening cold boot (which only
    happens if the computer lost power during sleep). Fast Startup can make
    it impossible to get at the boot menu, because Fast Startup begins so
    quickly that keyboard input is ignored, and you cannot get the boot menu
    to, say, elect to boot from a CD or USB drive instead of the BIOS
    configured primary boot device. Fast Startup was intended for use with
    old slow computer, especially those still using HDDs.

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Usenet Elder (2:250/1@fidonet)
  • From Anssi Saari@2:250/1 to All on Monday, November 17, 2025 10:14:08
    Subject: Re: Strange cabled network fault when rebooting from Windows (any version) into Ubuntu 24

    VanguardLH <V@nguard.LH> writes:

    Anssi Saari <anssi.saari@usenet.mail.kapsi.fi> wrote:

    Dan Purgert <dan@djph.net> writes:

    Win10, 11 have that "FastBoot" thing that mucks with hardware releases
    on "reboot" (win basically goes into hibernate).

    No. Reboot is always reboot, Windows would be completely useless without
    that. "FastBoot" aka fast startup happens when shutting down if not
    disabled. And it's hibernate without hibernating apps so fairly useless.

    Actually Fast Starup is a full hibernate (all memory copied into the hyberfil.sys file).

    Source? For example here: https://learn.microsoft.com/en-us/windows-hardware/test/weg/delivering-a-great-startup-and-shutdown-experience
    "Starting with Windows 8.x, the default shutdown and restart scenario
    has been updated and named fast startup. Fast startup begins with the
    shutdown process and includes writing data to disk similar to the
    hibernate process. A key difference is that all user sessions (Session
    1) are logged off and the remaining information is written to the
    hiberfile."

    When user sessions are logged off, all user apps die and so aren't
    written to the hiberfil.sys, which, as I stated, makes this "fast
    startup" fairly useless.

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: An impatient and LOUD arachnid (2:250/1@fidonet)
  • From Carlos E.R.@2:250/1 to All on Monday, November 17, 2025 11:36:55
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    On 2025-11-16 03:18, VanguardLH wrote:
    Anssi Saari <anssi.saari@usenet.mail.kapsi.fi> wrote:

    Dan Purgert <dan@djph.net> writes:

    Win10, 11 have that "FastBoot" thing that mucks with hardware releases
    on "reboot" (win basically goes into hibernate).

    No. Reboot is always reboot, Windows would be completely useless without
    that. "FastBoot" aka fast startup happens when shutting down if not
    disabled. And it's hibernate without hibernating apps so fairly useless.

    Actually Fast Starup is a full hibernate (all memory copied into the hyberfil.sys file). The computer then goes into sleep mode. When
    brought out of sleep, the computer resumes from sleep. If, however, the computer ever lost power during sleep, the computer resumes using the hibernate file. Because a memory image is reinstated or resumed from a
    Fast Startup mode, there is no re-initialization of hardware. This is
    the same as a Windows restart which is a warm boot.

    No, hardware has to be reinitialized "somehow". Hardware has been
    powered off, they have to be put back in the same status as they were
    when the machine hibernated. The driver needs adequate entries to
    restore status.



    --
    Cheers, Carlos.
    ES🇪🇸, EU🇪🇺;

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Air Applewood, The Linux Gateway to the UK & Eire (2:250/1@fidonet)
  • From VanguardLH@2:250/1 to All on Monday, November 17, 2025 12:19:18
    Subject: Disable FastBoot (was: Strange cabled network fault when rebooting from Windows (any version) into Ubuntu 24)
    Keywords: VanguardLH,VLH

    Anssi Saari <anssi.saari@usenet.mail.kapsi.fi> wrote:

    VanguardLH <V@nguard.LH> writes:

    Anssi Saari <anssi.saari@usenet.mail.kapsi.fi> wrote:

    Dan Purgert <dan@djph.net> writes:

    Win10, 11 have that "FastBoot" thing that mucks with hardware releases >>>> on "reboot" (win basically goes into hibernate).

    No. Reboot is always reboot, Windows would be completely useless without >>> that. "FastBoot" aka fast startup happens when shutting down if not
    disabled. And it's hibernate without hibernating apps so fairly useless.

    Actually Fast Starup is a full hibernate (all memory copied into the
    hyberfil.sys file).

    Source? For example here: https://learn.microsoft.com/en-us/windows-hardware/test/weg/delivering-a-great-startup-and-shutdown-experience
    "Starting with Windows 8.x, the default shutdown and restart scenario
    has been updated and named fast startup. Fast startup begins with the shutdown process and includes writing data to disk similar to the
    hibernate process. A key difference is that all user sessions (Session
    1) are logged off and the remaining information is written to the
    hiberfile."

    When user sessions are logged off, all user apps die and so aren't
    written to the hiberfil.sys, which, as I stated, makes this "fast
    startup" fairly useless.

    Depends on whether or not you chose to sleep (whether manually
    instigated, or by idle timeout) or shutdown while Fast Startup (aka
    FastBoot) was enabled.

    On sleep, Fast Startup saves an image of memory into the hiberfil.sys
    file. That's in case power is lost during sleep which means a reboot is needed, so the hiberfil.sys file is used to write back the memory image
    on the boot. If power is not lost during sleep mode, well, you just
    resume out of sleep mode.

    On shutdown, yep, you chose to exit your Windows session(s), but the
    reboot will still use the hiberfil.sys file to get more quickly get back
    to the logon state. Fast Startup does not alter that a shutdown logs
    you out. Fast Startup with shutdown does not waste the time nor disk
    space to write a hiberfil.sys file to record your Windows session. With
    Fast Startup enabled, it saves a memory image only for the OS kernel and drivers. The point of Fast Startup aka FastBoot is to get Windows
    booted faster, not restore your Windows session(s). You get to the
    logon screen faster. It is Fast*Boot*. Not FastSessionRestore.

    Perhaps I wasn't clear. Saving "all memory" into the hiber file is not
    while you have a Windows session. It is at the point of the logon
    screen to reload a memory image for that state instead of having to wait
    for the kernel to load, all drivers to load, all services to start, etc.

    Fast Startup may speed up booting, but it slows shutdown. When the
    computer goes into sleep mode, you don't notice the writes to
    hiberfil.sys. Users are more sensitive to how long it takes them to
    resume using Windows, because they're waiting there staring at the
    monitor. They are less sensitive to what happens going to sleep, or
    during shutdown, because typically they leave, so they are not waiting.
    Faster startup, slower shutdown.

    If you slept your computer (manually instigated, or by idle timeout),
    you can exit sleep mode to get back into your Windows session assuming
    the computer never lost power. If power was lost during sleep, yes,
    Fast Startup gets you more quickly up to the logon screen, because the
    kernel, driver, and services were reinstated from the memory image, not
    from having to load them from scratch.

    If you hibernated (manually instigated, or by timeout), that should be a regular hibernate, so you resume your Windows session when you boot.

    If you shutdown, why would you expect your Windows session(s) to get
    saved?

    If you lost power to the computer while in a Windows session (i.e., not sleeping), how could anything no longer running still write into the
    hiber file to restore a Windows session that was nuked by a power
    outage?

    FastBoot (aka Fast Startup) is for faster booting, not for Windows
    session resume. Use sleep mode to resume Windows sessions, and hope
    your computer doesn't lose power while sleeping. If power is lost while sleeping, Fast*Boot* gets to the logon screen faster.

    With SSDs, especially with m.2 NVMe SSDs, getting much more common for
    the OS disk, FastBoot provides little in a faster boot time, but causes
    lots of headaches when troubleshooting. I suspect FastBoot will
    eventually fade away (get deprecated), like how ReadyBoost has faded
    away (using USB flash drives or SD cards to add a faster drive-to-RAM
    cache for HDDs despite HDDs already have a small cache, but ReadyBoost
    could never be as fast as adding more real RAM). ReadyBoost was
    oriented to old computers still using HDDs for the OS disks. FastBoot
    was oriented toward old and slow computers. Little point using either
    today with typically larger RAM configs, and with SSDs, especially
    NVMEs, replacing HDDs.

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Usenet Elder (2:250/1@fidonet)
  • From Paul@2:250/1 to All on Monday, November 17, 2025 12:23:54
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    On Mon, 11/17/2025 5:14 AM, Anssi Saari wrote:
    VanguardLH <V@nguard.LH> writes:

    Anssi Saari <anssi.saari@usenet.mail.kapsi.fi> wrote:

    Dan Purgert <dan@djph.net> writes:

    Win10, 11 have that "FastBoot" thing that mucks with hardware releases >>>> on "reboot" (win basically goes into hibernate).

    No. Reboot is always reboot, Windows would be completely useless without >>> that. "FastBoot" aka fast startup happens when shutting down if not
    disabled. And it's hibernate without hibernating apps so fairly useless.

    Actually Fast Starup is a full hibernate (all memory copied into the
    hyberfil.sys file).

    Source? For example here: https://learn.microsoft.com/en-us/windows-hardware/test/weg/delivering-a-great-startup-and-shutdown-experience
    "Starting with Windows 8.x, the default shutdown and restart scenario
    has been updated and named fast startup. Fast startup begins with the shutdown process and includes writing data to disk similar to the
    hibernate process. A key difference is that all user sessions (Session
    1) are logged off and the remaining information is written to the
    hiberfile."

    When user sessions are logged off, all user apps die and so aren't
    written to the hiberfil.sys, which, as I stated, makes this "fast
    startup" fairly useless.


    Fast startup changes the start from maybe 15 seconds to 5 seconds (fast machine).
    It is intended to get that screen lit up quickly, to impress your friends. (Some people have boot contests, and they would like this.)

    If used to be, you got a fast boot time, by owning a "mighty" processor.
    That was the purpose of the boot contest, was to show how you must
    have spent a lot of money on the machine. Fast start allows you to cheat.

    It's Windows kernel+drivers hibernation, the session is not saved.

    During the five second boot, the blob is read in from hiberfil.sys,
    and the drivers are warm started (data arrays initialized). If any
    new hardware was plugged in, it will eventually be discovered (perhaps
    after the desktop has been painted).

    But Fast Boot is also a "handcuffs", as it prevents multi-boot.

    Paul

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From VanguardLH@2:250/1 to All on Monday, November 17, 2025 12:26:20
    Subject: Disable FastBoot (was: Strange cabled network fault when rebooting from Windows (any version) into Ubuntu 24)
    Keywords: VanguardLH,VLH

    "Carlos E.R." <robin_listas@es.invalid> wrote:

    On 2025-11-16 03:18, VanguardLH wrote:
    Anssi Saari <anssi.saari@usenet.mail.kapsi.fi> wrote:

    Dan Purgert <dan@djph.net> writes:

    Win10, 11 have that "FastBoot" thing that mucks with hardware releases >>>> on "reboot" (win basically goes into hibernate).

    No. Reboot is always reboot, Windows would be completely useless without >>> that. "FastBoot" aka fast startup happens when shutting down if not
    disabled. And it's hibernate without hibernating apps so fairly useless.

    Actually Fast Starup is a full hibernate (all memory copied into the
    hyberfil.sys file). The computer then goes into sleep mode. When
    brought out of sleep, the computer resumes from sleep. If, however, the
    computer ever lost power during sleep, the computer resumes using the
    hibernate file. Because a memory image is reinstated or resumed from a
    Fast Startup mode, there is no re-initialization of hardware. This is
    the same as a Windows restart which is a warm boot.

    No, hardware has to be reinitialized "somehow". Hardware has been
    powered off, they have to be put back in the same status as they were
    when the machine hibernated. The driver needs adequate entries to
    restore status.

    It is the lack of initialization of hardware on a FastBoot startup why
    hardware that was hung, or in an inoperable state, remains so. FastBoot
    copies a memory image of the kernel and drivers in their state at that
    time, and restore those states on startup. Even if the hardware gets
    the CPU reset on a cold boot, FastBoot is going to reinstate its memory
    image of the kernel and drivers. The kernel and drivers are NOT loaded
    on a FastBoot. They are resumed from a saved state. That the memory
    image of states doesn't match hardware state is why FastBoot causes
    problems. When rebooting to attempt troubleshooting, FastBoot can make
    the boot-time menu disappear so fast the user has no chance of hitting a
    key to get it recognized to go into troubleshooting mode, or make a
    selection from a boot menu.

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Usenet Elder (2:250/1@fidonet)
  • From Frank Slootweg@2:250/1 to All on Monday, November 17, 2025 16:51:22
    VanguardLH <V@nguard.lh> wrote:
    Anssi Saari <anssi.saari@usenet.mail.kapsi.fi> wrote:

    VanguardLH <V@nguard.LH> writes:

    Anssi Saari <anssi.saari@usenet.mail.kapsi.fi> wrote:

    Dan Purgert <dan@djph.net> writes:

    Win10, 11 have that "FastBoot" thing that mucks with hardware releases >>>> on "reboot" (win basically goes into hibernate).

    No. Reboot is always reboot, Windows would be completely useless without >>> that. "FastBoot" aka fast startup happens when shutting down if not
    disabled. And it's hibernate without hibernating apps so fairly useless. >>
    Actually Fast Starup is a full hibernate (all memory copied into the
    hyberfil.sys file).

    Source? For example here: https://learn.microsoft.com/en-us/windows-hardware/test/weg/delivering-a-great-startup-and-shutdown-experience
    "Starting with Windows 8.x, the default shutdown and restart scenario
    has been updated and named fast startup. Fast startup begins with the shutdown process and includes writing data to disk similar to the
    hibernate process. A key difference is that all user sessions (Session
    1) are logged off and the remaining information is written to the hiberfile."

    When user sessions are logged off, all user apps die and so aren't
    written to the hiberfil.sys, which, as I stated, makes this "fast
    startup" fairly useless.

    Depends on whether or not you chose to sleep (whether manually
    instigated, or by idle timeout) or shutdown while Fast Startup (aka
    FastBoot) was enabled.

    [Fast Startup doesn't affect 'sleep'. See more below.]

    On sleep, Fast Startup saves an image of memory into the hiberfil.sys
    file. That's in case power is lost during sleep which means a reboot is needed, so the hiberfil.sys file is used to write back the memory image
    on the boot. If power is not lost during sleep mode, well, you just
    resume out of sleep mode.

    With "On sleep", you probably mean on Shutdown, because when Fast
    Startup is enabled, it only affects, what happens during shutdown
    (does a hibernate of the OS, not the user session, etc.) and what
    happens during boot/power-up (resumes from the hibernated system copy),
    it does not affect what happens during sleep.

    Also the "That's in case power is lost during sleep" is misleading,
    because that's not the/a reason for Fast Startup, I even doubt that Fast Startup will use the hibernated system copy when rebooting/powering-on
    after a power loss, because the system might have been changed since
    the time of the *previous* shutdown, so the hibernated system copy might
    be stale/invalid.

    [Much more of the same, IMO also not fully correct, deleted.]

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: NOYB (2:250/1@fidonet)
  • From Carlos E.R.@2:250/1 to All on Monday, November 17, 2025 18:24:39
    On 2025-11-17 13:19, VanguardLH wrote:
    Fast Startup may speed up booting, but it slows shutdown. When the
    computer goes into sleep mode, you don't notice the writes to
    hiberfil.sys. Users are more sensitive to how long it takes them to
    resume using Windows, because they're waiting there staring at the
    monitor. They are less sensitive to what happens going to sleep, or
    during shutdown, because typically they leave, so they are not waiting. Faster startup, slower shutdown.

    But sometimes it fails to power down.

    Once, in the airport queue for security, I noticed a lady with the
    laptop lid closed, but still powered up. I told her. She then tried to
    power it off, failed, and in despair, took again her place in the queue.

    Long ago, I had that situation, laptop that would not finally power down
    and the battery ran out. I do not remember the cause.

    --
    Cheers, Carlos.
    ES🇪🇸, EU🇪🇺;

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Air Applewood, The Linux Gateway to the UK & Eire (2:250/1@fidonet)
  • From Carlos E.R.@2:250/1 to All on Monday, November 17, 2025 18:58:56

    Restoring all groups. Or I will miss some messages to reply to.

    On 2025-11-17 18:38, VanguardLH wrote:
    MikeS <mikes@is.invalid> wrote:
    On 17/11/2025 12:26, VanguardLH wrote:
    "Carlos E.R." <robin_listas@es.invalid> wrote:
    On 2025-11-16 03:18, VanguardLH wrote:
    Anssi Saari <anssi.saari@usenet.mail.kapsi.fi> wrote:
    Dan Purgert <dan@djph.net> writes:

    Win10, 11 have that "FastBoot" thing that mucks with hardware releases >>>>>>> on "reboot" (win basically goes into hibernate).

    No. Reboot is always reboot, Windows would be completely useless without >>>>>> that. "FastBoot" aka fast startup happens when shutting down if not >>>>>> disabled. And it's hibernate without hibernating apps so fairly useless. >>>>>
    Actually Fast Starup is a full hibernate (all memory copied into the >>>>> hyberfil.sys file). The computer then goes into sleep mode. When
    brought out of sleep, the computer resumes from sleep. If, however, the >>>>> computer ever lost power during sleep, the computer resumes using the >>>>> hibernate file. Because a memory image is reinstated or resumed from a >>>>> Fast Startup mode, there is no re-initialization of hardware. This is >>>>> the same as a Windows restart which is a warm boot.

    No, hardware has to be reinitialized "somehow". Hardware has been
    powered off, they have to be put back in the same status as they were
    when the machine hibernated. The driver needs adequate entries to
    restore status.

    It is the lack of initialization of hardware on a FastBoot startup why
    hardware that was hung, or in an inoperable state, remains so. FastBoot >>> copies a memory image of the kernel and drivers in their state at that
    time, and restore those states on startup. Even if the hardware gets
    the CPU reset on a cold boot, FastBoot is going to reinstate its memory
    image of the kernel and drivers. The kernel and drivers are NOT loaded
    on a FastBoot. They are resumed from a saved state. That the memory
    image of states doesn't match hardware state is why FastBoot causes
    problems. When rebooting to attempt troubleshooting, FastBoot can make
    the boot-time menu disappear so fast the user has no chance of hitting a >>> key to get it recognized to go into troubleshooting mode, or make a
    selection from a boot menu.

    A little knowledge is a dangerous thing. Here is an accurate description:
    https://learn.microsoft.com/en-us/windows-hardware/drivers/kernel/distinguishing-fast-startup-from-wake-from-hibernation

    To distinguish fast startups from wake-from-hibernation,
    kernel-mode device drivers can examine system power IRPs.

    ....

    If the driver for a device configures the device
    differently depending on whether a cold startup or
    a wake-from-hibernation occurred, this driver should,
    after a fast startup, configure the device as though
    a cold startup occurred. For example, the system-supplied
    NDIS driver disables miniport wake capabilities on a
    fast startup but not on a wake-from-hibernation.


    "2. Next, the power manager sends system power IRPs to device drivers to
    tell them to prepare their devices to enter hibernation."

    You've never encountered drivers that did not restore state correctly on resuming from hybrid hibernate?

    "If the driver for a device configures the device differently depending
    on whether a cold startup or a wake-from-hibernation occurred, this
    driver should, after a fast startup, configure the device as though a
    cold startup occurred."

    Those wonderful "shoulds".

    If there is a power outage during sleep, or during a Windows session,
    how is all this graceful recovery information going to be captured on a computer that is instantly dead?

    Which is why sleep and hibernation are complicated things. It is not
    just the memory contents that have to be restored, but also all
    hardware, like for example, the video mode. The driver must know how to restore, what to do if warm or cold, etc.

    The video card has its own registers to program the video mode, which
    are not part of the main memory, and have to be restored in a certain sequence.

    And the video card memory has to be restored, or recreated.

    And the same for all chips and cards.

    And of course there can be bugs when multibooting.

    --
    Cheers, Carlos.
    ES🇪🇸, EU🇪🇺;

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Air Applewood, The Linux Gateway to the UK & Eire (2:250/1@fidonet)
  • From Carlos E.R.@2:250/1 to All on Tuesday, November 18, 2025 00:23:25
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24


    Restoring all groups. Or I will miss some messages to reply to.

    On 2025-11-18 00:41, Windows Elf wrote:
    On 17/11/2025 10:14, Anssi Saari wrote:
    VanguardLH <V@nguard.LH> writes:

    Anssi Saari <anssi.saari@usenet.mail.kapsi.fi> wrote:

    Dan Purgert <dan@djph.net> writes:

    Win10, 11 have that "FastBoot" thing that mucks with hardware releases >>>>> on "reboot" (win basically goes into hibernate).

    No. Reboot is always reboot, Windows would be completely useless without >>>> that. "FastBoot" aka fast startup happens when shutting down if not
    disabled. And it's hibernate without hibernating apps so fairly useless. >>>
    Actually Fast Starup is a full hibernate (all memory copied into the
    hyberfil.sys file).

    Source? For example here:
    https://learn.microsoft.com/en-us/windows-hardware/test/weg/delivering-a-great-startup-and-shutdown-experience
    "Starting with Windows 8.x, the default shutdown and restart scenario
    has been updated and named fast startup. Fast startup begins with the
    shutdown process and includes writing data to disk similar to the
    hibernate process. A key difference is that all user sessions (Session
    1) are logged off and the remaining information is written to the
    hiberfile."

    When user sessions are logged off, all user apps die and so aren't
    written to the hiberfil.sys, which, as I stated, makes this "fast
    startup" fairly useless.

    If you think hiberfil.sys is useless, why not disable it? Use this
    command as an administrator:

    powercfg.exe /hibernate off.

    Restart the machine and see if it makes any difference.

    Personally, I find fast startup useful on an old machine, but some
    people prefer to start afresh while they make a cup of coffee!

    Fast boot is not compatible with double booting.

    --
    Cheers, Carlos.
    ES🇪🇸, EU🇪🇺;

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Air Applewood, The Linux Gateway to the UK & Eire (2:250/1@fidonet)
  • From Lawrence =?iso-8859-13?q?D=FFOlivei@2:250/1 to All on Tuesday, November 18, 2025 04:44:43
    On Mon, 17 Nov 2025 19:24:39 +0100, Carlos E.R. wrote:

    Long ago, I had that situation, laptop that would not finally power down
    and the battery ran out. I do not remember the cause.

    You were running Windows.

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Lawrence =?iso-8859-13?q?D=FFOlivei@2:250/1 to All on Tuesday, November 18, 2025 04:47:25
    Subject: Re: Disable FastBoot (was: Strange cabled network fault when
    rebooting from Windows (any version) into Ubuntu 24)

    On Mon, 17 Nov 2025 06:26:20 -0600, VanguardLH wrote:

    It is the lack of initialization of hardware on a FastBoot startup why hardware that was hung, or in an inoperable state, remains so.

    None of that is relevant to booting another OS, though. If Windows expects
    the hardware to be in a certain state, another OS will not and will reinitialize it from power up.

    Unless that second OS is trying to do its own equivalent “fast boot”, which in its turn assumes that the hardware will be in the same start it
    last left it. That is likely to lead to sadness.

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Carlos E.R.@2:250/1 to All on Tuesday, November 18, 2025 06:08:18
    On 2025-11-18 05:44, Lawrence D’Oliveiro wrote:
    On Mon, 17 Nov 2025 19:24:39 +0100, Carlos E.R. wrote:

    Long ago, I had that situation, laptop that would not finally power down
    and the battery ran out. I do not remember the cause.

    You were running Windows.

    At that moment, yes.

    Linux has also been known to do something similar, except that it goes
    through the motions of saving the hibernation image, only that at the
    last instant the power off fails.

    --
    Cheers, Carlos.
    ES🇪🇸, EU🇪🇺;

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Air Applewood, The Linux Gateway to the UK & Eire (2:250/1@fidonet)
  • From Carlos E.R.@2:250/1 to All on Tuesday, November 18, 2025 06:32:23
    On 2025-11-18 05:47, Lawrence D’Oliveiro wrote:
    On Mon, 17 Nov 2025 06:26:20 -0600, VanguardLH wrote:

    It is the lack of initialization of hardware on a FastBoot startup why
    hardware that was hung, or in an inoperable state, remains so.

    None of that is relevant to booting another OS, though. If Windows expects the hardware to be in a certain state, another OS will not and will reinitialize it from power up.

    Unless that second OS is trying to do its own equivalent “fast boot”, which in its turn assumes that the hardware will be in the same start it
    last left it. That is likely to lead to sadness.

    I remember cases of people telling about some hardware that would fail
    on a cold boot of Linux, that would work in a cold boot of Windows, and
    would work if booting Linux warm after closing Windows.

    Obviously after the developers were told about this the situation was corrected.

    --
    Cheers, Carlos.
    ES🇪🇸, EU🇪🇺;

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: Air Applewood, The Linux Gateway to the UK & Eire (2:250/1@fidonet)
  • From Lawrence =?iso-8859-13?q?D=FFOlivei@2:250/1 to All on Tuesday, November 18, 2025 07:31:56
    On Tue, 18 Nov 2025 07:32:23 +0100, Carlos E.R. wrote:

    Obviously after the developers were told about this the situation was corrected.

    I imagine the same would apply here: there is some aspect of the hardware setup that the Linux driver is neglecting to ensure is properly
    initialized, and no doubt the kernel developers would fix it if it was reported.

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Anssi Saari@2:250/1 to All on Wednesday, November 19, 2025 09:11:23
    Frank Slootweg <this@ddress.is.invalid> writes:

    [Much more of the same, IMO also not fully correct, deleted.]

    Thanks. I thought about responding again to VanguardLH but sheesh.

    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: An impatient and LOUD arachnid (2:250/1@fidonet)
  • From Java Jive@2:250/1 to All on Friday, November 21, 2025 20:41:56
    Subject: Re: Strange cabled network fault when rebooting from Windows (any
    version) into Ubuntu 24

    On 2025-11-15 19:43, Carlos E.R. wrote:

    On 2025-11-15 15:39, Java Jive wrote:

    In case it's helpful to anyone else ...

    I have some Dell Precision M6800s & M6700s dual-booting via Grub
    between various versions of Windows and Ubuntu 24, all cabled to the
    LAN.  If I choose to move between any version of Windows and Ubuntu by
    rebooting, Ubuntu 24 cannot make a reliable cabled LAN connection, it
    keeps trying to connect but never succeeds.  However, if I choose to
    move between any version of Windows and Ubuntu by shutting down the PC
    in between, Ubuntu 24 'just works', no networking problem at all.
    Reboot from U24 and go back into U24, the situation is unchanged  -
    if it was working before the reboot, it's working now, if it wasn't
    working before the reboot, it isn't working now.  Reboot from U24 and
    go back into Windows, no network problems in Windows at all.  Reboot
    from Windows and go back into any version of Windows, no network
    problems at all.

    In summary, when rebooting ...
         Ubuntu 24 -> Ubuntu 24    No change
         Ubuntu 24 -> Windows      No problem
         Windows -> Ubuntu 24      U24 cannot make cabled LAN connection
         Windows -> Windows        No problem

    It seems to me from this that the cabled network hardware is not being
    properly initialised when booting Ubuntu 24.  The cure is simply to
    actually power down the PC before launching Ubuntu 24

    [snip]

    Open a bug with the Ubuntu people, somehow.

    Done:

    https://bugs.launchpad.net/ubuntu/+bug/2132151

    --

    Fake news kills!

    I may be contacted via the contact address given on my website: www.macfh.co.uk


    --- MBSE BBS v1.1.1 (Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)