[TRACKING] USB port stopped working: over current condition

In the middle of otherwise normal use of the framework laptop with Ubuntu 22.04, the external USB-plugged mouse stopped working. Plugging the mouse into another computer showed it works just fine. The light of the mouse is off when plugged into the framework laptop.

Just what could have happened for the USB port to die like that?

Then I went to see the dmesg, and oops:

$ dmesg
dmesg: read kernel buffer failed: Operation not permitted

For some reason, one needs sudo rights. Trying again, now we see:

[  223.535740] usb 3-4: USB disconnect, device number 6
[  223.883543] usb 3-4: new full-speed USB device number 7 using xhci_hcd
[  224.015652] usb 3-4: device descriptor read/64, error -71
[  224.588922] usb 3-4: New USB device found, idVendor=0738, idProduct=1705, bcdDevice= 2.22
[  224.588932] usb 3-4: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[  224.588935] usb 3-4: Product: Mad Catz R.A.T.5 Mouse
[  224.588938] usb 3-4: Manufacturer: Mad Catz
[  224.593279] input: Mad Catz Mad Catz R.A.T.5 Mouse as /devices/pci0000:00/0000:00:14.0/usb3/3-4/3-4:1.0/0003:0738:1705.0007/input/input29
[  224.593893] saitek 0003:0738:1705.0007: input,hidraw1: USB HID v1.11 Mouse [Mad Catz Mad Catz R.A.T.5 Mouse] on usb-0000:00:14.0-4/input0
[  918.081423] usb usb3-port4: over-current condition
[  918.081439] usb 3-4: USB disconnect, device number 7
[  918.493418] usb usb3-port6: over-current condition

In other words: over-current condition. How can this happen? Is it because the laptop is charging via the adjacent USB-C port? Unplugging the charger didn’t solve the issue.

2 Likes

To further add: plugging other USB devices into the same USB-3 port works. Somehow the laptop decided it doesn’t like my mouse anymore. Is there a table somewhere of “blacklisted” USB devices to which my mouse may have been suddenly added? Worked well for long time.

This sounds eerily familiar to another thread:
USB cooling pad causes USB-A port to die, requiring an eject and re-insert.
I wonder if there is some weird behavior in certain USB products causing electrical issues in the USB expansion cards. It would be worthwhile to send a report to support on this so that they can let the hardware team know, they may want to try to recreate the issue.

1 Like

Thanks for the pointer and the recommendation.

I have just updated to the BIOS 3.10 and the USB 3 port is suddenly working again for the exact same mouse. So for now, problem solved.

Incidentally, the update to 3.10 also solved the deep sleep problem! I can now use “deep” instead of “s2idle”.

2 Likes

Spoke too soon: the USB 3 port is now not working, again, for the same mouse. Worked well while on battery power, and continued to work as the battery went from 60% to 85% or so, and suddenly stopped working.

Same issue: overcurrent. From the dmesg:

[15550.365064] usb 3-4: USB disconnect, device number 7
[15550.693068] usb 3-4: new full-speed USB device number 8 using xhci_hcd
[15550.821139] usb 3-4: device descriptor read/64, error -71
[15551.397351] usb 3-4: New USB device found, idVendor=0738, idProduct=1705, bcdDevice= 2.22
[15551.397359] usb 3-4: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[15551.397362] usb 3-4: Product: Mad Catz R.A.T.5 Mouse
[15551.397365] usb 3-4: Manufacturer: Mad Catz
[15551.401473] input: Mad Catz Mad Catz R.A.T.5 Mouse as /devices/pci0000:00/0000:00:14.0/usb3/3-4/3-4:1.0/0003:0738:1705.0007/input/input29
[15551.402136] saitek 0003:0738:1705.0007: input,hidraw3: USB HID v1.11 Mouse [Mad Catz Mad Catz R.A.T.5 Mouse] on usb-0000:00:14.0-4/input0
[15852.226063] audit: type=1326 audit(1661862733.192:249): auid=1000 uid=1000 gid=1000 ses=3 subj=? pid=143897 comm="chrome" exe="/snap/chromium/2064/usr/lib/chromium-browser/chrome" sig=0 arch=c000003e syscall=330 compat=0 ip=0x7f88f09f900b code=0x50000
[15874.365723] iwlwifi 0000:aa:00.0: reached 10 old SN frames from d4:20:b0:af:36:f4 on queue 7, stopping BA session on TID 0
[15874.725079] usb usb3-port4: over-current condition
[15874.725092] usb 3-4: USB disconnect, device number 8
[15875.125037] usb usb3-port6: over-current condition
[15895.856607] iwlwifi 0000:aa:00.0: reached 10 old SN frames from d4:20:b0:af:36:f4 on queue 7, stopping BA session on TID 0
[15919.198901] iwlwifi 0000:aa:00.0: reached 10 old SN frames from d4:20:b0:af:36:f4 on queue 7, stopping BA session on TID 0
1 Like

Hey @Albert_Cardona and welcome to the forum! :wave:

To find out if it’s a problem with the mainboard, internal usb-c header or the card: Try plugging the card you’re experiencing problems with into another Expansion bay and then plugging your adapter into there.

(As I’ve seen on the forum it might actually be a problem with the mainboard as the overcurrent protection seems to be indeed a bit finicky but won’t hurt to try)

Hope this helps! :blush:

Thank you for the suggestion. Can the expansion bays be hot-swapped or does one need to shutdown? I assume being USB-C internally that they can be swapped live?

1 Like

Expansion bays can be hot swapped as they’re basically just internal dongles using usb-c yeah

OK just tried it live, and yes the USB bay is now working. And the mouse works. Thanks!

So what’s the underlying issue? I did swap bays, so now there’s a card reader where the “faulty” USB 3 was.

1 Like

Give it a bit of time to see if the issue comes back. If it doesn’t contact support about this and link this thread :blush:

If it actually is just this specific port something went wrong with the OC protection on that specific port either because it shorted out on something that was plugged in or you got a faulty mainboard.

Oh i’d also suggest to not plug anything ‘mission critical’ into that SD card reader as it might get borked if my hypothesis is true! :sweat_smile:

1 Like

Thanks. Do you know which bays are these (from dmesg):

[    2.065361] usb usb3-port1: over-current condition
[    2.281360] usb usb3-port3: over-current condition

This issue with overcurrent on USB ports has been cloaked for some months by this other issue Kworker stuck at near 100% CPU usage with Ubuntu 22.04 - #12 by Albert_Cardona which renders USB ports entirely unusable except for charging the battery. The USB ports and the camera do work when modprobe’ing back the xhci-pci module, but it’s quite the chore.

Please create and boot to a Live Ubuntu USB. We need to know if this is a bad installation or a hardware issue. This means we need to work with a brand new, vanilla Live USB to see if this issue happens there as well.

Regarding your other thread, I echo the steps there as well. Thanks :slight_smile:

encountered this issue as well, the usb devices on the left side just stopped working until a eject and replugged in, and became unavailable after next reboot, however the ports on the right side is normal, i also have usb overcurrent in my dmesg outputs

1 Like

Definitely something to keep an eye on. Noting:

  • What was connected to said port; expansion card with USB or direct connect to USB-C.
  • TLP installed and active, split testing with it temporarily disabled to see if the issue repeats. Should have nothing to do with overcurrent obviously, but good to keep thing vanilla for tracking.

If it’s an expansion card, could be a bad card.

1 Like

The problem happens even without TLP active, on Windows, I recieve warnings about port power surge, I didn’t tested connecting USB-C directly since I didn’t have dual-USB-C cables at my hand that time, AC-in worked fine though.
but I’ve just resolved this by resetting CMOS, strange thing is, after a reset, ports(with expansion cards) weren’t available until AC plugged in on that side, now all ports worked fine.

I think I spoke too soon, the ports went overcurrent again overnight, not sure what happened, dmesg didn’t output anything about overcurrent thing, but when I reboot, the message in boot sequence showed usb overcurrent condition just as before.

Few things we can try.

  • When the over-current is happening, running journalctl -f to see things unfold in real time.

  • If we can spot what is happening, great, please post it back here. If nothing and considering this is happening in Windows, the issue is likely the expansion card itself and needs to be brought up with support.

The issue continues present, and makes this framework laptop feel crippled–because it is.

I was one happy customer when I bought it, this issue didn’t exist then, even though this laptop has only ever run Ubuntu 22.04. If it is happening across Fedora and Ubuntu, it could either be the kernel, or somehow the mere process of aging a bit is behind this. Either way, it shows poorly on Framework, the company, that all I got so far is “try this, try that”, and when returning all info asked for, I got silence for a while, then more questions, never ending, stretching for months. In the US, I would have simply returned it for a refund. Here in Europe one feels screwed over, as the only apparent solution is to buy a new motherboard + CPU for about £1000, which is out of question.

So here we are, using a crippled laptop that can’t have any ports open without one CPU being pegged. Works well even, with that CPU pegged, but it isn’t good for noise (the fan shoots up) or battery life, obviously.

Do you want to spend under £1000 in good PR? Send me a replacement motherboard. At this point it’s likely much cheaper than that, as I have an old version 11, with an 11th Gen Intel(R) Core™ i7-1165G7 @ 2.80GHz.

I’m not exactly sure what you mean here. If you purchased the laptop in Europe you should have a 2 year warranty (1 year better than the US). Unless you purchased it before it was available in your country (accepting all the risk of doing so) you should hopefully be able to reach out to support to ask for a replacement.

1 Like