[RESPONDED] Issue with booting USB from port 6

I found an unusual issue with my Framework 16 where port 6 doesn’t allow me to boot live USB drives the same way as port 3 when using a USB-A adapter in each.

I documented more of the issue on Stack Exchange (linux - Grub "failure reading sector" from Live USB - Unix & Linux Stack Exchange), but I had two live USBs on two different drives:

  • Ubuntu 22.04 LTS on a 16 GB USB 3.0 drive
  • Ubuntu Sway Remix (based on 22.04 LTS) on a 4 GB USB 2.0 drive

The Ubuntu Sway live USB was able to boot in either port 3 or port 6 (both have a USB-A adapter). The Ubuntu 22.04 live USB was only able to boot in port 3. On port 6, I continually got an error:

error: failure reading sector 0x5b88c0 from 'hd0'

I’m hoping someone else can help test this to see if they have the same issue. More details are in the Stack Exchange link, including a picture of the error.

I have the Ryzen 7 7840HS and Radeon RX 7700S dGPU. Ubuntu 24.04 LTS was installed, but my issues are during boot, so I don’t think the installed OS matters.

1 Like

Just tried booting into a live install of Fedora, worked fine with no issues on my slot 6

I might have had the same issue. I spent two days trying to get various OSs to boot off a USB-A drive from slot 6. None totally worked, then I got a new USB-C drive and plugged into slot 1 and was able to install. Though I also disabled secure boot at that point, which may have been hindering the installs. I also think it’s possible I fried my USB-A drive by too many flashing of OSs…

What size drive and what USB generation? I don’t have many drives on me, so I am not sure if it’s the fact that my smaller drive is 4 GB vs 16 GB or USB 2.0 vs 3.0, but that one works on port 6 while the larger one doesn’t.

I checked secure boot several times, but I had it disabled in all my tests. I actually don’t think it matters too much for me since I can boot into Ubuntu Sway Remix and Ubuntu 22.04 just fine with it enabled, as long as I use port 3 for the drive with Ubuntu 22.04 on it.

Could you try moving your USB-C to A adapter to port 3 (or any other port) and see if the USB-A drive boots in that? I tried swapping my adapters between ports 3 and 6 and it didn’t affect anything, so I don’t think it’s related to the adapter but the port itself.

Alternatively, if that’s the case, you should have trouble booting the USB-C drive on port 6 as well (unless the generation difference matters).

I’ve tried two different sticks both are 64GB and USB 3.0

1 Like

I recently tried booting from ports 3 and 6… ended up with an nvme drive hitting 80c at idle and no visible boot option. I think there’s some other things going on… but I don’t think it should’ve been possible to get the drive that hot. It wasn’t the usb-adapter doing it either as I tried a couple from different brands.

I just kept banging my head against the wall and eventually it broke and I got what I wanted… but I never figured out what happened with the temperatures.

Adding my 2 cents so I can get a notification if anything weird is found to be going on with ports 3/6.


In this photo, the drive refuses to mount and is performing no I/O operations.

1 Like

Which processor option does everyone have? Maybe it also depends on that.

Tested, was not able to reproduce, tested all ports. No issues. Tested again with another flash drive. This feels like bad sector behavior (which can work fine until it doesn’t - hard to repro).

I can try again this weekend, but it was 100% reproducible previously, even after reformatting my 16 GB drive and reinstalling a live USB. The GRUB error did prompt me to scan the disk in Windows as well as use badblocks, but nothing showed issues with the disk (and the write test was hugely destructive – lesson learned).

If it is a bad sector, why would it work on one port but not on another?

@Silntknight
If a particular sector is the problem, you would only see the error message if a program accesses that sector.
As it is in grub and not in Linux kernel, there are not any logs to view, so difficult to diagnose further.

Sure, I can understand that, but what I don’t understand is why the behavior is different between two different ports. Again, booting from port 3 works fine, but port 6 presented the issue.

Can you say which ports it manages to boot from?
I.e. test with Port 1, 2, 3, 4, 5, 6 and not just report on 3,6.
I am having some problems with both 3 and 6 for certain USB devices but other ports are OK.
Also, have you tried (on a port that works):
smartctl -a /dev/xxxx where xxx is the USB device, e.g. /dev/sdb
and post the output here. It might mention some errors.

I am guessing that this might be one of two things:

  1. a faulty device, with a bad sector 0x5b88c0. a “dd” command to scan entire disk, something like this might reproduce the problem on a working port:
    dd if=/dev/sdb of=/dev/null bs=4k status=progress
    Look for pauses, and any sector failure messages in the syslog/dmesg/journelctl -b.
  2. some sort of USB port / usb-pd problem not sending enough power to the USB device.

I can test additional ports and report back. I only tested those two because those are the two in which I use my USB A expansion cards. The others are typically doing something else (like providing power), so I didn’t test them before.

I didn’t try smartctl, but would that even work on a USB drive? I tried a SMART check on a Windows machine, but I don’t think the drive supports SMART diagnostics since it’s not an SSD/HDD. I can try that specifically, though.

I did run dd and it was very destructive (will need to re-setup live USB), but didn’t show any errors.

As for USB-PD, this is a thumb drive. I could understand that for an external HDD or SSD, but I don’t think that’s likely the cause here. Also, if power were an issue, I would expect the “bad” sector to constantly change as the power limitation would not likely result in a low voltage or similar issue at the exact same block each time.

I do appreciate the suggestions for what to try next, and once I get the live USB set up again, I will attempt to replicate the issue and perform the additional tests you requested

EDIT: I realized I ran badblocks, not dd, but I still didn’t get errors and I still ended up wiping the drive

I’ve been lurking around the forums for a while now, but I realized that I’m having a similar issue so… I guess it’s time I speak up.

I have an M.2 NVME enclosure with a 512GB SSD inside that acts pretty much like a beefy flash drive. It has a USB A port and a USB C port built-in. The first time I tried accessing files on it with my FW16 I used a USB A card in port 6, and it didn’t work. I then switched to another USB A card on port 5, and it started working immediately. I didn’t really think too much of it at the time.

I’ve done some investigating now and here’s what I’ve found on my end: The enclosure doesn’t work on only ports 3 and 6, using either USB A or USB C expansion cards, on both Windows and Linux, for both accessing files and booting off of. I’ve tried multiple other flash drives and a 512GB SATA SSD with a USB A adapter, and everything else seems to work fine.

Here is the exact enclosure that I’m having issues with. And, even though it probably doesn’t matter, the SSD is an XPG SX8200 Pro salvaged from an old laptop.

1 Like

Hi,

I have a similar RTL9210B enclosure.
It does not work in slots 3, 6.
By you also seeing problems, at least this confirms the I am not the only one.
I am now looking for a way to capture the CC1, CC2 pin usb messages, to hopefully eventually understand why it works in some slots and not others.

@Silntknight
The purpose of the dd command is that it would have been non-destructive and would simply have tried to read every sector on the usb stick to see if any were a problem.
I suspect it is just a broken usb stick, and you probably need to buy a new one.

@Scrubz
When I plug my RTL9210B enclosure into port 3,6. Its light flashes, so it looks like the FW16 is applying power, removing it, applying power, removing it etc.
Do you see similar?

Yep, that’s what happens to mine too.

The enclosure I’m using here appears to use the RTL9210B IC as well where the drive reached 80c before I unplugged it. Applying and removing power would definitely result in at least some of the issues I was seeing.

https://www.amazon.com/gp/product/B0C572RFLQ/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

I also had disconnect issues with this one (RTL9210, they don’t specify a B on the description but that could be a typo afaik):
https://www.amazon.com/gp/product/B09T97Z7DM/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

That drive reaching 80c was one of the very few times I felt mildly unsafe around my pc.