Debian freezes when connecting witch usb device

Hi,

I have a very serious problem. When I try to connect with a USB device, debian freezes and I have to reboot. My client is debian machine in a VirtualBox VM and the server runs on a Raspberry Pi.
Does anybody know how I can fix that problem.

Greetings Adrian

#2

Some old versions of 64-bit based debain e.g Ubuntu 64bit 12.04 have this issue. What Debian are you using?

#4

Ok, looks like that device is not compatible with the Linux kernel usbip drivers. What device is it?

#6

Thats odd, that should work ok, have you tried with ubuntu etc? in the virtualbox, that should work. Im away for a few days, when i get back to the office I will install debian 7.8 in a vm and give it a test and let you know how it goes..

I haven't tried Ubuntu. I will try it in a different vm, maybe is something weird with my installation. Thank you that you are going to try it as well.

#8

OK i installed debian 7.8 and i see the issue...its still using the old 3.2 kernel . I remember that was the same kernel that didnt work in ubuntu 12.04 so definately you need to use a newer kernel.

Debian 7.8 as a server will work, debian 7.8 as a client will not...

#9

adi961,

If the old kernel is the problem here, you can add the Debian Wheezy backports repository to your sources.list file on Debian 7.8 and grab the much newer Linux kernel 3.16 from there.

#10

Thank you for your help. I decided to switch to Ubuntu and now the client works fine.
Greetings Adrian

#12

FYI, it seems that 3.2.54 does not work, but 3.2.65 does.

In reply to by caseyboardman

#13

Temporarily, it seems. Now getting same error again:
Client is Debian 7, kernel 3.2.65-1. 64-bit. Virtual Machine.

tail -f /var/log/syslog:
Aug 12 15:46:19 vcboardma-deb7 VirtualHere: Received IPC verb="USE", arg="4294967710"
Aug 12 15:46:19 vcboardma-deb7 kernel: [ 324.493749] vhci_hcd vhci_hcd: rhport(0) sockfd(21) devid(4) speed(2)
Aug 12 15:46:19 vcboardma-deb7 kernel: [ 324.493912] vhci_hcd: changed 1

Interestingly, though those 3 lines are the last we see from tailing that log, they are *not* included in the log after resetting this machine. So that's weird.

Going to try updating and upgrading the works to see if that solves this.

Any and all insight appreciated - figured some of these messages could look familiar.

-Casey

It seems like the trick is to try to connect to a non-existent device:

"USE,666" <- not a thing
"LIST" <- works OK
"USE,4294967710" <- a real device

Machine hangs as above.

Server log:
Wed Aug 12 16:33:45 2015 LOG_INFO Device 414 BOUND to connection 6
Wed Aug 12 16:33:55 2015 LOG_DEBUG Removing connection 6 (timeout)...
Wed Aug 12 16:33:55 2015 LOG_INFO Device 414 UNBOUND from connection 6
Wed Aug 12 16:33:57 2015 LOG_INFO Connection 6 successfully removed

Thoughts? Is there more logging I can turn on?

#15

If the machine is hanging it is definately a kernel bug, so you will need to use a more recent kernel than 3.2.65 perhap jessie with 3.16