[vbox-dev] USB in SVN not working

Mon Jun 4 18:00:19 GMT 2012

On 04.06.2012 19:36, Perry Halbert wrote:
> I'll double check it here in a bit, however i am not sure this issue is
> related to your changes. Part of me wants to think that this has been
> around for some time.
>
> Where are the locks stored ( i can look for them but if you know it
> would make my search easier) I know of /tmp/.V** but they don't seem to
> be there.

The locks I'm talking about are thread synchronization primitives, not 
filesystem locks. So nothing which is visible outside VBoxSVC and the VM 
process.

Klaus

>
>
>
> On 06/04/2012 12:28 PM, Klaus Espenlaub wrote:
>> On 04.06.2012 18:19, Perry Halbert wrote:
>>> Here is a strange issue.
>>>
>>> The USB device works great and I don't seem to have any issues with the
>>> exception to one. I can use the devices tab and remove the device and it
>>> returns to the host as expected, but if you use the guests release it
>>> gets lost in limbo. Not available to the guest or the host.
>> What's the state of the device, as reported by "VBoxManage list usbhost"?
>>
>> It's unlikely, but of course still possible that the changes have this
>> weird side effect...
>>
>>> Unplugging the device and inserting again returns the use of the device.
>>>
>>> Note: I do not use a filter. Always select the USB device from the
>>> devices tab.
>> On my system (with a filter for the USB device) there's a quirk with
>> logically detaching a USB device the first time, too, but I already
>> verified it's definitely unrelated to the changes - it immediately
>> re-grabs the device as for some reason linux assigns a new device name
>> to it...
>>
>> Klaus
>>
>>>
>>> On 06/04/2012 06:17 AM, Klaus Espenlaub wrote:
>>>> On 01.06.2012 23:01, Perry Halbert wrote:
>>>>> It works well (Linux hosts 32-64).
>>>>>
>>>>> I will see about running my debug build tomorrow and see what I can
>>>>> break as I am a little tied up today.
>>>>>
>>>>> Are the subtle misbehaviors in Linux as well as Windows?
>>>> The main USB breakage (related to locking changes) in trunk was common
>>>> to all platforms. The vast majority of fixes are in the common code,
>>>> however it was also necessary to make changes to the platform dependent
>>>> USB code, so there is a (hopefully very small) chance that platform
>>>> specific issues are left. The changes to the platform dependent code
>>>> were extremely simple and repetitive across the platforms compared to
>>>> the changes in the common code.
>>>>
>>>> Klaus
>>>>
>>>>> On 06/01/2012 01:01 PM, Klaus Espenlaub wrote:
>>>>>> On 14.05.2012 13:57, Frank Mehnert wrote:
>>>>>>> Hi Perry,
>>>>>>>
>>>>>>> I've checked this issue again and now I can reproduce it. No idea
>>>>>>> what I've tested before, sorry for that!
>>>>>>>
>>>>>>> This is indeed a recent regression and will be fixed ASAP, I will
>>>>>>> let you know.
>>>>>> It should be fixed now (r41528 and r41529) - please report any
>>>>>> noticeably different behavior compared to VBox 4.1 (which wasn't
>>>>>> affected by this trunk only change/regression).
>>>>>>
>>>>>> It'd be great if people with experience could run debug builds (both
>>>>>> VBoxSVC and the VM frontend inside a debugger, otherwise triggering an
>>>>>> assertion will lead to a sudden death) and let us know if any assertions
>>>>>> were triggered in relationship with USB operations. Release builds
>>>>>> should still show subtle misbehavior, but debug builds will shout ;)
>>>>>>
>>>>>> Thanks,
>>>>>> Klaus
>>>>>>
>>>>>>> Thanks for your heads-up,
>>>>>>>
>>>>>>> Frank
>>>>>>>
>>>>>>> On Friday 11 May 2012 16:38:24 Perry Halbert wrote:
>>>>>>>> This started happening sometime after ExtPack VNC was enabled in the
>>>>>>>> build but not right away.  It worked as it should for a while with VNC
>>>>>>>> and with USB.  See pipermail link below.  Could this be a reason for the
>>>>>>>> issue now?
>>>>>>>>
>>>>>>>> https://www.virtualbox.org/pipermail/vbox-dev/2012-March/005001.html
>>>>>>>>
>>>>>>>> It should be noted that there are no build errors as such except the
>>>>>>>> normal errors about hard links.
>>>>>>>>
>>>>>>>> On 05/10/2012 06:36 PM, Perry Halbert wrote:
>>>>>>>>> It has to be something in the base code.
>>>>>>>>>
>>>>>>>>> I installed the latest .deb and it works.
>>>>>>>>> Compiled packed and install the latest tarball (4.1.14) and ExtPack.
>>>>>>>>> and it works.
>>>>>>>>> Compile the latest from SVN.  Fails.  Says the guest is not running.
>>>>>>>>> (with the latest ExtPacks) from Frank.
>>>>>>>>>
>>>>>>>>> Completely purged the build environment and started fresh. So it could
>>>>>>>>> not be a corrupt file. Still failed.
>>>>>>>>>
>>>>>>>>> Sorry to say but i don't check USB too often and the first time I
>>>>>>>>> noticed it failed was r41127 but I tried all the way back to r40671
>>>>>>>>> (oldest backup I have) and it fails there as well.
>>>>>>>>>
>>>>>>>>> On 05/10/2012 11:40 AM, Perry Halbert wrote:
>>>>>>>>>> Oh and to be clear this happens on two different machines so
>>>>>>>>>> something is afoot.
>>>>>>>>>>
>>>>>>>>>> On 05/10/2012 11:35 AM, Perry Halbert wrote:
>>>>>>>>>>> Same thing.  See additional information..
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> Version = 4.1.51_OSEr41221
>>>>>>>>>>>
>>>>>>>>>>> id = 1005(vboxusers)
>>>>>>>>>>> I removed VirtualBox&     the vboxusers group and logged out.
>>>>>>>>>>> Installed VirtualBox and it created the vboxusers group again (it
>>>>>>>>>>> was 1001 and now 1005)&     added my login to the group and rebooted.
>>>>>>>>>>>
>>>>>>>>>>> list extpacks = Extension Packs: 1
>>>>>>>>>>> Pack no. 0:   Oracle VM VirtualBox Extension Pack
>>>>>>>>>>> Version:      4.1.51
>>>>>>>>>>> Revision:     77890
>>>>>>>>>>> Edition:
>>>>>>>>>>> Description:  USB 2.0 Host Controller, VirtualBox RDP, PXE ROM with
>>>>>>>>>>> E1000 support.
>>>>>>>>>>> VRDE Module:  VBoxVRDP
>>>>>>>>>>> Usable:       true
>>>>>>>>>>> Why unusable:
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> Try to enable inside the working guest:
>>>>>>>>>>>
>>>>>>>>>>> Failed to attach the USB device Kingston DataTraveler 2.0 [0200] to
>>>>>>>>>>> the virtual machine Debian-Buildbox.
>>>>>>>>>>>
>>>>>>>>>>> The virtual machine is not powered up.
>>>>>>>>>>>
>>>>>>>>>>> Result Code: E_ACCESSDENIED (0x80070005)
>>>>>>>>>>> Component: Console
>>>>>>>>>>> Interface: IConsole {1968b7d3-e3bf-4ceb-99e0-cb7c913317bb}
>>>>>>>>>>>
>>>>>>>>>>> I will try to go back a few builds and see where it start to work
>>>>>>>>>>> again, but it is not working now and the funny thing is the report
>>>>>>>>>>> of the machine not powered up when trying to enable the device in
>>>>>>>>>>> the running guest.
>>>>>>>>>>>
>>>>>>>>>>> On 05/10/2012 10:19 AM, Frank Mehnert wrote:
>>>>>>>>>>>> Hi Perry,
>>>>>>>>>>>>
>>>>>>>>>>>> On Friday 04 May 2012 02:40:36 Perry Halbert wrote:
>>>>>>>>>>>>> SVN build = 4.1.51_OSEr41127
>>>>>>>>>>>>> ExtPak = 4.1.51r77491
>>>>>>>>>>>>>
>>>>>>>>>>>>> Trying to mount a USB device (after the guest is already running)
>>>>>>>>>>>>> error says the virtual machine is not powered up.
>>>>>>>>>>>>>
>>>>>>>>>>>>> E_ACCESSDENIED (0x80070005)
>>>>>>>>>>>>>
>>>>>>>>>>>>> [...]
>>>>>>>>>>>>>
>>>>>>>>>>>>> User is in vboxusers group (verfied)
>>>>>>>>>>>> I have no idea. This works here and we didn't change any access
>>>>>>>>>>>> mechanism thing in the code.
>>>>>>>>>>>>
>>>>>>>>>>>> Could you check again if something changes if you use a more recent
>>>>>>>>>>>> ExtPack?
>>>>>>>>>>>>
>>>>>>>>>>>> http://www.virtualbox.org/download/testcase/Oracle_VM_VirtualBox_Exten
>>>>>>>>>>>> sion_Pack-4.1.51-77890.vbox-extpack
>>>>>>>>>>>>
>>>>>>>>>>>> Kind regards,
>>>>>>>>>>>>
>>>>>>>>>>>> Frank
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>> vbox-dev mailing list
>>>>>>>>>>>> vbox-dev at virtualbox.org
>>>>>>>>>>>> https://www.virtualbox.org/mailman/listinfo/vbox-dev
>>>>>>>>>>> _______________________________________________
>>>>>>>>>>> vbox-dev mailing list
>>>>>>>>>>> vbox-dev at virtualbox.org
>>>>>>>>>>> https://www.virtualbox.org/mailman/listinfo/vbox-dev
>>>>>>>>>> _______________________________________________
>>>>>>>>>> vbox-dev mailing list
>>>>>>>>>> vbox-dev at virtualbox.org
>>>>>>>>>> https://www.virtualbox.org/mailman/listinfo/vbox-dev
>>>>>>>>> _______________________________________________
>>>>>>>>> vbox-dev mailing list
>>>>>>>>> vbox-dev at virtualbox.org
>>>>>>>>> https://www.virtualbox.org/mailman/listinfo/vbox-dev
>>>>>>> _______________________________________________
>>>>>>> vbox-dev mailing list
>>>>>>> vbox-dev at virtualbox.org
>>>>>>> https://www.virtualbox.org/mailman/listinfo/vbox-dev
>>>>> _______________________________________________
>>>>> vbox-dev mailing list
>>>>> vbox-dev at virtualbox.org
>>>>> https://www.virtualbox.org/mailman/listinfo/vbox-dev
>>>
>>> _______________________________________________
>>> vbox-dev mailing list
>>> vbox-dev at virtualbox.org
>>> https://www.virtualbox.org/mailman/listinfo/vbox-dev
>>
>
>
> _______________________________________________
> vbox-dev mailing list
> vbox-dev at virtualbox.org
> https://www.virtualbox.org/mailman/listinfo/vbox-dev

-- 
Oracle <http://www.oracle.com>
Dr. Klaus Espenlaub | Snr. Manager Software Development Desktop
Virtualization
Phone: +49 7151 60405 205 <tel:+49715160405205>
Oracle VM VirtualBox

ORACLE Deutschland B.V. & Co. KG | Werkstr. 24 | 71384 Weinstadt

ORACLE Deutschland B.V. & Co. KG
Hauptverwaltung: Riesstr. 25, D-80992 München
Registergericht: Amtsgericht München, HRA 95603
Geschäftsführer: Jürgen Kunz

Komplementärin: ORACLE Deutschland Verwaltung B.V.
Hertogswetering 163/167, 3543 AS Utrecht, Niederlande
Handelsregister der Handelskammer Midden-Niederlande, Nr. 30143697
Geschäftsführer: Alexander van der Ven, Astrid Kepper, Val Maher

Green Oracle <http://www.oracle.com/commitment> 	Oracle is committed to
developing practices and products that help protect the environment