[OpenAFS] Windows client stops working with Bluetooth adapter attached to client workstation

[John Perkins]

My oversight for Windows version: XP SP3.

Yes, there is a roaming profile being sync'ed (or attempted to sync) to 
AFS, thus the delay.  I strongly suspect the locked state is an attempt 
to sync the profile back to AFS when AFS is not answering for whatever 
reason.

I have no way to tell whether the Bluetooth adapter survives logoff.  
Removing the Bluetooth adapter seems to allow a successful logout, though.

The only eventlog eventlog entry that strikes as even remotely related 
are these:

 MRxSmb  Unknown User    N/A     Warning ROADRUNNER      None    0000: 
00 00 00 00 04 00 4e 00   ....... 0008: 00 00 00 00 cb 0b 00 80   
........  0010: 00 00 00 00 84 01 00 c0   ........  0018: 00 00 00 00 00 
00 00 00   ........  0020: 00 00 00 00 00 00 00 00   ........         
The redirector failed to determine the connection type.
 AFS Client      Unknown User    N/A     Warning ROADRUNNER      
None            Unable to Send SMB Packet: NRC_SNUMOUT session number 
out of range.     30

(The  SMB redirector error pops occasionally on a number of our systems, 
working or not, so I didn't think much of that error here.)

John Perkins



Jeffrey Altman wrote:
> Important missing detail: version of Windows?
>
> Are there any profiles or directories that are redirected to AFS
> that would need to be written to during logoff?
>
> Does the bluetooth adapter survive logoff?
>
> Are there any messages in the Windows Event Log from services other than
> AFS during logoff?
>
> On 3/17/2010 12:49 PM, John Perkins wrote:
>   
>> We've run across one user who is using a computer with a USB Bluetooth
>> adapter attached and running the OpenAFS client for Windows (1.5.68
>> right now, to be updated to 1.5.72 this evening).
>>
>> The symptom the user has reported is that the computer appears to hang
>> when logging out.  The afs_init.log indicates something funky going on
>> with the network interfaces:
>>
>> 3/15/2010 11:05:04 AM: Mountpoint[0] = .cs%cs.wisc.edu:root.cell.
>> 3/15/2010 11:05:04 AM: Mountpoint[1] = .cs.wisc.edu%cs.wisc.edu:root.cell.
>> 3/15/2010 11:05:04 AM: Mountpoint[2] = cs.wisc.edu#cs.wisc.edu:root.cell.
>> 3/15/2010 11:05:04 AM: Mountpoint[3] = @cell#cs.wisc.edu:root.cell.
>> 3/15/2010 11:05:04 AM: Mountpoint[4] = cs#cs.wisc.edu:root.cell.
>> 3/15/2010 11:05:04 AM: Mountpoint[5] = .@cell%cs.wisc.edu:root.cell.
>> 3/15/2010 11:05:06 AM: Windows Firewall Configuration succeeded
>> 3/15/2010 11:05:36 AM: SERVICE_CONTROL_INTERROGATE
>> 3/15/2010 11:06:02 AM: SERVICE_CONTROL_INTERROGATE
>> 3/15/2010 11:06:02 AM: SERVICE_CONTROL_INTERROGATE
>> 3/15/2010 11:09:55 AM: smb_LanAdapterChange
>> 3/15/2010 11:09:55 AM: NCBLISTEN lana=5 failed with NRC_BRIDGE, retrying
>> ...
>> 3/15/2010 11:09:55 AM: NCBLISTEN lana=5 failed with NRC_NOWILD, retrying
>> ...
>> 3/15/2010 11:10:17 AM: smb_LanAdapterChange
>> 3/15/2010 11:10:17 AM: NCBLISTEN lana=5 failed with NRC_BRIDGE, retrying
>> ...
>> 3/15/2010 11:10:17 AM: NCBLISTEN lana=5 failed with NRC_NOWILD, retrying
>> ...
>>
>> ...and the computer will sit, apparently frozen, for hours at this point.
>> This appears to be similar to the behavior reported and commented on in
>> this post
>> http://www.openafs.org/pipermail/openafs-info/2009-February/030926.html
>>
>> The computer in question has 2 wired interfaces (only one in use) as
>> well as the MS Loopback Adapter and the Bluetooth adapter installed. 
>> The wired adapters are Realtek PCIe GBE Family Controller adapters.
>> The user in question was using a different computer that had only one
>> wired adapter with an NVIDIA nForce Networking Controller installed.
>>
>> Is this likely due to the unused wired network adapter or the Bluetooth
>> adapter?  Any other suggestions for how to avoid such problems?
>>
>> afsd_init.log available upon request (or I could send it on to
>> openafs-bugs if that is more appropriate).
>>
>> -- 
>> =========================================================================
>>   John Perkins                   |   University of Wisconsin-Madison
>>   Researcher                     |   Department of Computer Science
>>   john@cs.wisc.edu               |   1210 W. Dayton St.
>>   608-262-0438/608-262-6626 FAX  |   Madison, WI  53706-1685
>> =========================================================================
>>
>>
>>
>> _______________________________________________
>> OpenAFS-info mailing list
>> OpenAFS-info@openafs.org
>> https://lists.openafs.org/mailman/listinfo/openafs-info
>>
>>     
>
>