[ Date Index ] [ Thread Index ] [ <= Previous by date / thread ] [ Next by date / thread => ]
On 07/04/2020 22:25, Julian Hall wrote:
I suspect the fly in the ointment is that last line ending 'added'.
Believe it or not I mostly agree with you - that's not quite right and it clashes with the correctly DKMS built 8812au modules listed above. I've seen that type of "added" message before when getting the DKMS status on systems where something has gone subtly wrong but am not immediately sure quite what. We do need to eliminate it though - this is exactly the sort of artefact I was worrying about when I asked you previously to clear up all previous experiments and messes so we could start clean. This one obviously snuck past you...
Actually I'm not so sure now I think about it - that line just means that it was added previously but is not currently in use(??). Let's just try and purge it anyway. Try:
sudo dkms remove -m rtl8814AU/4.3.21 --allThat might not be quite right so if it doesn't parse try tab completion (another sticky point on your system somehow...) for the "rtl..." part and the "4.3.21" part: both should autocomplete fine on a sane system.
Let's try and solve your weird bash completion issue as well: sudo apt install apt-file sudo apt-file updateReboot once you've DKMS purged that rogue rtl8814AU instance and updated the apt-file cache. Then check the usual suspects:
* sudo dkms status * lsmod | egrep -i 'rtl|881'See if you can autocomplete some random kernels "sudo apt install linux-image- TAB TAB" and some other stuff, hopefully that'll work by now as well.
Then if all looks good driver-wise and the correct module has been loaded, "dmesg -wT" and plug it in.
After this it gets tricky. I'd be inclined to report a bug to the github issues tracker for the project, it's pretty active. I've already had a browse through the bugs there and your system still doesn't look right compared to others - the "SerialNumber: 123456" and the device plumbing in but not getting a designation are both issues that have come up before for people and are a sure sign it's not working. Does the light come on by the way when you plug it in? Reading the bug reports it looks as if it should come on bright blue when fully activated.
If you're up to it then personally I'd do all the usual sysadmin tests: swap it between USB ports, sanity check it in a couple of different computers with different operating systems, etc. All time consuming but if you want to fix it properly then tough luck I'm afraid! I'd also switch it to a different Mint box (which I know you have to hand) and now you know what you're doing, redo the driver DKMS install from scratch on a clean system to see how that goes.
After _that_ it's file a bug report and wait for the dev to come back I'm afraid. Welcome to Linux!
Tip for the future - I expect you've had this older wifi adaptor lying around for a while so it's a sunk cost, no problem. When buying wifi anythings in the future that are destined for a Linux system don't bother looking up the usual driver support matrices, instead research what chipsets/models the evil hackers are using for their packet injection shenanigans. If a wifi adaptor is not just linux supported but also has solid recommendations online for use in airmon-ng etc and is known good for supporting packet injection then it's a keeper. Pay over and above the normal prices for a unit like that and you won't have to put up with this crap in the future. In the old days that meant _always_ buying certain Atheros chipset units but there are a lot of options now. Remember, no packet injection support = no buying for linux boxes.
-- The Mailing List for the Devon & Cornwall LUG https://mailman.dcglug.org.uk/listinfo/list FAQ: http://www.dcglug.org.uk/listfaq