You are not logged in.
I have a wireless antenna which i use both with windows and arch linux (dual boot on separate drives).
If on windows 10 i "turn off wifi" (clicking the blue wifi button or enabling airplane mode to be clear) and reboot into linux, i can't see any Access Point until i boot windows again and turn it back on.
How is this possible? I always tought a reboot would clear/reset all peripherals or settings a driver could have set, seems like i was wrong.
Fast boot on windows is completely disabled trough powershell with
powercfg -h off It's not really a problem for me but im really interested on why does it happen, any clarification / link where i can lean more about this is very welcome
Last edited by Skunky (2020-07-07 10:42:01)
Offline
Did you check rfkill? It's entirely normal for state changes like these to be persisted in the BIOS/firmware to not unnecessarily reenable wifi. You should however be able to unblock it just the same in linux. If it's hard blocked then that's likely a firmware quirk, which you might be able to fix by an update of said firmware.
Offline
Also ensure to disable windows fastboot...
Should have read to the end first.
Last edited by seth (2020-07-03 08:13:37)
Online
Yep rfkill says it's "unblocked"
It's entirely normal for state changes like these to be persisted in the BIOS/firmware
So it is possible that the uefi save the driver settings / devices state or something like that? Is there is any difference between having a legacy or uefi OS in this regard?
Anyway systemctl status NetworkManager tell me the interface is managed(?) when it doesn't work
#Non working
lug 03 10:03:38 arch NetworkManager[698]: <info> [1593763418.1603] manager: (eth): new Ethernet device (/org/freedesktop/NetworkManager/Devices/2)
lug 03 10:03:38 arch NetworkManager[698]: <info> [1593763418.1609] device (eth): state change: unmanaged -> unavailable (reason 'managed', sys-iface-state: 'exte
nal>
lug 03 10:03:38 arch NetworkManager[698]: <info> [1593763418.1622] device (wifi): driver supports Access Point (AP) mode
lug 03 10:03:38 arch NetworkManager[698]: <info> [1593763418.1626] manager: (wifi): new 802.11 Wi-Fi device (/org/freedesktop/NetworkManager/Devices/3)
lug 03 10:03:38 arch NetworkManager[698]: <info> [1593763418.1631] device (wifi): state change: unmanaged -> unavailable (reason 'managed', sys-iface-state: 'ext
rna>
lug 03 10:03:38 arch NetworkManager[698]: <info> [1593763418.1762] device (wifi): set-hw-addr: set MAC address to 12:34:56:78:90:00 (scanning)
lug 03 10:03:38 arch NetworkManager[698]: <info> [1593763418.6076] ovsdb: Could not connect: No such file or directory
lug 03 10:03:38 arch NetworkManager[698]: <info> [1593763418.6307] device (wifi): supplicant interface state: internal-starting -> disconnected
lug 03 10:03:38 arch NetworkManager[698]: <info> [1593763418.6307] device (wifi): state change: unavailable -> disconnected (reason 'supplicant-available', sys-i
ace>
lug 03 10:03:44 arch NetworkManager[698]: <info> [1593763424.1624] manager: startup completesystemctl status NetworkManager when it works
#Working
lug 03 10:07:25 arch NetworkManager[284]: <info> [1593763645.4313] dhcp4 (wifi): state changed unknown -> bound
lug 03 10:07:25 arch NetworkManager[284]: <info> [1593763645.4326] device (wifi): state change: ip-config -> ip-check (reason 'none', sys-iface-state: 'managed')
lug 03 10:07:25 arch NetworkManager[284]: <info> [1593763645.5343] device (wifi): state change: ip-check -> secondaries (reason 'none', sys-iface-state: 'managed'
)
lug 03 10:07:25 arch NetworkManager[284]: <info> [1593763645.5347] device (wifi): state change: secondaries -> activated (reason 'none', sys-iface-state: 'managed
')
lug 03 10:07:25 arch NetworkManager[284]: <info> [1593763645.5353] manager: NetworkManager state is now CONNECTED_LOCAL
lug 03 10:07:25 arch NetworkManager[284]: <info> [1593763645.5373] manager: NetworkManager state is now CONNECTED_SITE
lug 03 10:07:25 arch NetworkManager[284]: <info> [1593763645.5374] policy: set '<SSID>' (wifi) as default for IPv4 routing and DNS
lug 03 10:07:25 arch NetworkManager[284]: <info> [1593763645.6281] device (wifi): Activation: successful, device activated.
lug 03 10:07:25 arch NetworkManager[284]: <info> [1593763645.6711] manager: startup complete
lug 03 10:07:26 arch NetworkManager[284]: <info> [1593763646.6835] manager: NetworkManager state is now CONNECTED_GLOBALAlso lsmod says that the rt2800usb module is loaded
Thank you.
Offline
https://wiki.archlinux.org/index.php/Ne … domization
If that's not it,
i can't see any Access Point
is a bit vague. Please stop NM and scan manuall, follow https://wiki.archlinux.org/index.php/Ne … ireless#iw until and including the "Discover access points" paragraph.
Online
At
# iw dev interface scan | lessI get no output, if i re-enable wifi on windows i can see all networks properly, seems like i can put the interface up and down anyway
Offline
Check dmesg for errors from the driver or the firmware.
To be sure, you did not literally issue "iw dev interface scan", did you?
Edit, no you didn't - it worked w/o windows… I need to pay more attention ;-)
Last edited by seth (2020-07-03 21:45:18)
Online
My bad i pasted the command from the wiki instead the one i actually typed.
i tried all dmesg log flags (err,warn,debug.. following the output with -w) and it doesn't output anything when i scan for networks, maybe the windows driver have the capability to set some kind of "state" that the linux driver hasn't?
EDIT: i just tried
- Disabling wifi from windows
- Booting linux, no wifi
- Replug the antenna in the usb port (power cycling it basically) it works!
Why doesn't it get power cycled automatically on reboot? I always tought all usb devices would have been reinitialized at reboot
Last edited by Skunky (2020-07-04 04:58:37)
Offline
The errors would likely occur before the scan (some initializationproblems)
If you boot "good", rfkill the NIC, then reboot linux (warm): do you get the same issue?
Have you tried to rfkill cycle the NIC (rfkill block, rfkill unblock) and whether that
a) causes kernel errors
b) effectively unblocks the device
Online
If i rfkill the NIC and reboot it's unblocked on the next warm reboot
Here is the dmesg log from when i started playing around with rfkill, no errors
[ 299.100911] audit: type=1130 audit(1593883411.256:101): pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=systemd-rfkill comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 304.107633] audit: type=1131 audit(1593883416.263:102): pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=systemd-rfkill comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 309.500290] audit: type=1130 audit(1593883421.656:103): pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=systemd-rfkill comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[ 314.506597] audit: type=1131 audit(1593883426.662:104): pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=systemd-rfkill comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'rfkill
root@arch ~ # rfkill
ID TYPE DEVICE SOFT HARD
0 wlan phy0 unblocked unblocked
root@arch ~ # rfkill block wifi
root@arch ~ # rfkill
ID TYPE DEVICE SOFT HARD
0 wlan phy0 blocked unblocked
root@arch ~ # rfkill unblock wifi
root@arch ~ # iw dev wifi scan
command failed: Network is down (-100)
root@arch ~ # ip link set wifi up
root@arch ~ # iw dev wifi scan
root@arch ~ #I can block and unblock the NIC with rfkill but it doesn't work anyway, thanks for your help
Offline
However as i said in my first post this is not really a problem for me, i just would like to know how is the uefi able to save such settings (even on legacy mode), i read alot about power states but i can't find any information about this, i would really like to know more about this what should i look for?
Offline
If it's an external stick I doubt that the UEFI has much relevance anymore. I'd say this is a flip within the sticks own firmware that does not get overriden unless power cycled, which I don't think you can inherently enforce without physically disconnecting the charge.
FWIW I incidentally had a similar incident yesterdayish that was only solvable (or rather I didn't look further into it and just did that) with complete disconnect from any source of power. I was working on some stuff and wanted to suspend to not have to manually reopen everything, this somehow lead the firmware to be in a state where, the suspend failed and on attempting to just shutdown instead it would continuously reboot the system even with a shutdown command.
Offline
That's interesting, a broken firmware in the device would also explain why rebooting doesn't help, i guess it also depends on how the motherboard firmware handle usb ports at reboot.
I also had another issue with a GTX 1060, like all gpu's it sometimes make fans go at full speed during POST and if i had to boot another OS then the fans would be stuck at full speed until another reboot.
I'm gonna mark this as solved as it seems to be very hardware dependent and it can vary from one setup to another
Offline