[lug] Gparted doesn't recognize new USB drive

[D. Stimits]

...
> > Can we see the output of 
> > 
> > $ lsblk
> 
> It doesn't show up there at all, nor in the output of fdisk -l.
> 
> In particular, "sudo lsblk -O /dev/sde" doesn't show any output at all.


> 
> It is present:
> $ ls -l /dev/sde
> brw-rw---- 1 root disk 8, 64 Mar 12 12:01 /dev/sde

FYI, "/dev/sde" is the result of the disk driver loading and is unrelated to USB. If USB were not functioning, then it would be impossible for that file to show up (sde is a virtual file which does not exist on the hard drive...it is entirely in RAM and part of the running driver).

USB can indeed interfere with low power or suspend modes, but the mass storage driver, despite being "standardized" and distributed with USB, is not technically a USB driver. USB is hot plug and announces a plug-in event and describes the device to drivers; if a driver can handle the device described, then it takes ownership. Ownership and loading of the mass storage driver is what the sde file represents.

> 
> All my /dev/sd block devices are in group "disk".
> 
> So, I plugged in a known working stick. What's now interesting is the lack,
> in the kernel messages for the new one, of something like this:
> 
> Product: USB 3.0 FD
> usb 2-4.3: Manufacturer: PNY Technologies
> usb 2-4.3: SerialNumber: AA67192670013655
>

What shows up in "dmesg --follow" as you plug in the other device? What shows up in "lsusb -t" before and after changing USB devices? The tree view does not give many details, but it does show the chain of HUBs and speeds without regard to any other drivers (it is purely a USB tool and won't care about mass storage).
 
> IOW, the USB subsystem apparently doesn't know enough about it to operate
> on it. That's my hypothesis. Maybe, in addition to seeing it as USB mass
> storage, it needs to know something about the chip, which it can't figure
> out. This might in fact point to a defective unit. Bad vendor ID or
> something like that.

It might be defective, it might be a power mode quirk. On the other hand I'd still be interested in knowing the output from "sudo gdisk -l /dev/sde". If gdisk can operate on the device, even if it says there is no readable partition or filesystem, then I'd try zeroing it out first and partitioning again with gdisk (don't use fdisk on any solid state drive).

> 
> Looking up the Vendor/Product, I find a Kingston "SP Mobile C31 (64GB)".
> Phillips might repackage Kingston, but there's a size mismatch.
> -- 
> All operating systems suck, but Linux just sucks less
>  - Linus Torvalds
> _______________________________________________
> Web Page:  http://lug.boulder.co.us
> Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
> Join us on IRC: irc.hackingsociety.org port=6667 channel=#hackingsociety