Such crappy readers exist? Oo Sure it's not the media?

Usually I would just refuse adding code to support trash-hardware or software, esp. since it reduces security. But we already have the Generic options for vendor and model and it will be optional anyway... Soooooo....

Will be added SOMEWHEN, but feel free to submit a PR if you're in a hurry.

Too lazy to test, but in theory it's available in branch issue-261-support-generic-serial

thanks for the quick update, will hopefully have some time to test this week!

So I installed it like this:

git checkout issue-261-support-generic-serial
git diff origin #looks good
make && sudo make install
echo $? # is 0
reboot

Then changed everything but the vol id to Generic.

Does not work because it cant find the device...

only tried with FAT vol

Serial needs to be set. Everything else can be Generic

No worries it's not urgent. I am just grateful for you maintenance of this project ^_^

Maybe I am just to stupid to install the commit properly. Will try again when it's merged into master.

Would appreciate if you could at least quickly test if your device still gets recognized when you set it's serial to Generic.

Thank you very much. I also had to set the Model (which was previously "MassStorageClass") also to Generic (Otherwise it would get stuck on the first drive), but now a I can add as many drives via this reader as I want!

The only thing worth mentioning would be that the warning for Model/Vendor gets printed 3 times each, kinda pollutes the log.
Otherwise works perfectly fine!

Okay maybe that was too early. How is it when you have 2 Devices with everything set to Generic?

I have issues when both drives are all generic. The first ones get's skipped but the second one works without any issues. ( And only the first ones prints the warning 3 times each).

When I set the model of the first one to MassStorageClass it starts working again but the second one only works once (the first time you unlock after changing the conf).

IDK what's going on here.

I think you should test if it works on you Machine with 2 devices all Generic and if it does close this issue. I don't want to steal more of you time when almost nobody is benefiting from this feature. Maybe this SD reader is really beyond saving.

Rebooted and tried to add a third device. From what I can tell it always get's stuck on the first device where everything but the vol. id is generic. It falsely detects that the first drive is connected and tries to probe for that volume (does not check if the UUID of the partition is there).

Also I don't get the warning printed 3 times each anymore ( only shows up when Model is not generic)

Well good news is that it's now a lot more stable. Before setting everything to generic I had to deal with a lot of timeouts / connections issues, now works like a breeze.

I am satisfied with the current state.

I would merge the pull request and add a quick note that setting serial to generic is not fully supported, has issues and link to this thread? (If you see no real benefit in fixing this)

Anyway thanks for you time.

Rebooted and tried to add a third device. From what I can tell it always get's stuck on the first device where everything but the vol. id is generic. It falsely detects that the first drive is connected and tries to probe for that volume (does not check if the UUID of the partition is there).

That is kinda an expected sideeffect and unlikely to change. But again: your cardreader is utterly broken, you should just use proper devices. As soon as ANY field is set to generic, you're reducing the already low security of pam_usb.

I am satisfied with the current state.

Glad it workarounded your "wtf hardware" 😁

I would merge the pull request and add a quick note that setting serial to generic is not fully supported, has issues and link to this thread? (If you see no real benefit in fixing this)

I don't plan to merge this into master since it has too many caveats and is only required for very niche and very broken hardware.

You will have to either stick to your current build, or to be able to update: maintain your own branch forked from master but merging this and then do local builds.

You will have to either stick to your current build, or to be able to update: maintain your own branch forked from master but merging this and then do local builds.

Alright no problem, thanks