Wi-Fi connection and Bluetooth file transfer problems ( Android-to-WindowsXP )

203 posts / novo 0
Último post
Para obter mais informações sobre otimizações de compiladores, consulte Aviso sobre otimizações.

>>>.Thanks. I also looked at MSDN and it clearly says that:

'...Bluetooth connectivity and data transfer are programmed through Windows Sockets function calls...'>>>
Probably as a wrapper functions or user-mode accessible API,but bear in mind that you won't be able to dissect lower levels of protocol stack where the most important information should be contained within.

The most interesting part is this driver BthPort.Sadly NetMon only can dissect NDIS drivers.

Hi Iliya,

>>...Can You use Linux?..

No.

>>...Probably as a wrapper functions or user-mode accessible API,but bear in mind that you won't be able to dissect lower levels of
>>protocol stack where the most important information should be contained within.

What I really would like to understand what side / component is responsible for that instability with file transfers? There are three components in my test environment:

- Windows XP ( with a generic Bluetooth driver, etc )
- Android device ( OS v4.0.4 ) with Bluetooth v3.0 ( a driver, etc )
- A generic Bluetooth device ( supports v2.0 or v2.1 ) which I connect to Windows XP

I also checked Windows Event Viewer logs and I found a message:

'TCP/IP has reached the security limit imposed on the number of concurrent TCP connect attempts'

Any comments on that?

>>>What I really would like to understand what side / component is responsible for that instability with file transfers? There are three components in my test environment:>>>
I wish I could give you an answer,but without performing first some kind of connection setup ,connection maintaining and connection tear-down monitoring we will not be able to eliminate failure of lowest level protocols.If and if lower Bluetooth protocol stack is not responsible in such a case
a culprit could be found in kernel-part of Bluetooth Windows or Android implementation and this invovlves invasive kernel debugging.

Earlier in my previous posts I asked you to perform simple elimination tests like Android -to-Android device,Android to your other Win machine and observe file transfer timing anomalies.It must be done in order to remove responsibilities either from Android device or from Windows pc and blame the oter side.

>>...Can You use Linux?..

No.

Sadly Linux port of Wireshark could have been very helpful in your case.You can install Linux as a vmware virtual machine.Wireshark has nice GUI interface and can sniff also USB packets.Sometimes Bluetooth is encapsulated inside USB packets.

>>>I also checked Windows Event Viewer logs and I found a message:

'TCP/IP has reached the security limit imposed on the number of concurrent TCP connect attempts'>>>

The best opotion is to fire up NetMon and scan for outbound connection attempts.You have to put a filter on SYN packets beign from your machine and observe validity of IP addresses.
Please consult also this article : http://www.microsoft.com/technet/support/ee/transform.aspx?ProdName=Wind...

Hi Iliya,

>>...The best opotion is to fire up NetMon and scan for outbound connection attempts.You have to put a filter on SYN packets beign from your
>>machine and observe validity of IP addresses.

I'll do it some time later and let you know.

>>Please consult also this article : http://www.microsoft.com/technet/support/ee/transform.aspx?ProdName=Wind...

Thanks. I will also try to upgrade Bluetooth devices to a newer which support version 4.0 and have Windows drivers. The current Bluetooth devices are very generic and Windows Hardware Wizard installed a generic Microsoft driver.

>>>I'll do it some time later and let you know.>>>
Try to search suspicious process which is generating those packets.

>>> I will also try to upgrade Bluetooth devices to a newer which support version 4.0 and have Windows drivers. The current Bluetooth devices are very generic and Windows Hardware Wizard installed a generic Microsoft driver.>>>

I hope that it will help you to solve your problems.
Please inform me about the results.

>>> Windows Hardware Wizard installed a generic Microsoft driver>>>
I think that Bluetooth miniport driver is written by Microsoft.

Hi everybody,

This is simply to inform that I have renamed the thread title to:

Wi-Fi connection and Bluetooth file transfer problems ( Android-to-WindowsXP )

>>>>> Windows Hardware Wizard installed a generic Microsoft driver>>>
>>I think that Bluetooth miniport driver is written by Microsoft.

Absolutely correct.

>>>Absolutely correct.>>>

I would also recommend you to consult WDK documentation please follow this link:http://msdn.microsoft.com/en-us/library/windows/hardware/gg487458.aspx
Bluetooth specification here:http://www.bluetooth.org/Technical/Specifications/adopted.htm

Hi Iliya,

I discovered that two Bluetooth devices I use for testing on Windows XP have identical network addresses. How is it possible? When I tried to use both on the same computer with Windows XP the 2nd Bluetooth device can not be initialized. I'll post some screenshoots later.

Best regards,
Sergey

>>>I discovered that two Bluetooth devices I use for testing on Windows XP have identical network addresses. How is it possible? >>>

Sometime it is possible.It could stem from DHCP server internal adress pool error or on bigger network when addresses are assigned staticalyy by admins also in some hacking attacks IP address spoofing is used.
In order to solve this problem assign manually static ip address to both devices.
I would also recommned to monitor with NetMon both devices connection setup stage.Pay attention to DHCP protocol frames containing ip address field

Take a look at enclosed pictures.

Anexos: 

>>...it is possible.It could stem from DHCP server internal adress pool error or on bigger network when addresses are assigned staticalyy by
>>admins also in some hacking attacks IP address spoofing is used.

It doesn't look like an IP address ( on attached screenshot 'BtProblemScreenB.jpg' ):
...
Address: ae:2d:22:00:ff:00
...

>>...This is a test with Embeded image in the post.

I hope that will work...


That works!

PS: After a couple of days the image is not displayed any longer!

>>>It doesn't look like an IP address ( on attached screenshot 'BtProblemScreenB.jpg' ):
...
Address: ae:2d:22:00:ff:00>>>
Sorry it is hardware MAC address.This address is asigned by the manufacturer of the device.Afaik many NIC's contain special status hardware register which is responsible for changing a MAC address.Your situation is very strange because MAC address range is 2^48 and encountering
two unmodified devices with the same address tells us about the some abnormal situation.The problem lies in ability to reprogram one of the MAC
addresses on Android device.I need to do some research on this issue.
Please read this article :http://androidforums.com/t-mobile-g1/3506-changing-spoofing-mac-address....

>>...This is a test with Embeded image in the post.

I hope that will work...>>>
Does that inactive Bluetooth icon device have the same MAC address as active one?
How are your Bluetooth receivers connected to the pc?
If this is an USB you can view the hubs with the help of usbview tool.Please try to run this tool and post the results.
Please connect only the faulting device and post the results.

Hi Iliya,

>>...Do you know some very technical numerical methods and/or scientific calculation forums?

Please take a look at:
.
www.mersenneforum.org/showthread.php?t=11765
.
www.mersenneforum.org
.
PS: You really need to look at it!

>>>PS: You really need to look at it!>>>
Thank you very much.You really made my day.Sadly they are coding in AT&T assembly notation.

What about your Android Bluetooth problems?

>>...What about your Android Bluetooth problems?

There is nothing new and a file transfer is not stable. After a couple of transfers I need to reboot the tablet and computer.

Some time ago you've asked me about ad-hoc connections settings for Wi-Fi. I verified it and it is as follows:

'Network to Access' - a setting is 'Any available network'

>>>There is nothing new and a file transfer is not stable. After a couple of transfers I need to reboot the tablet and computer.>>>
Did you connect one of your malfunctioning Bluetooth devices to pc?I would like to know if the same MAC (on both devices) address could be a reason for the inactive bluetooth device exactly as you have described in one of your posts.

>>>'Network to Access' - a setting is 'Any available network'>>>
Is this Android setting?

>>>>There is nothing new and a file transfer is not stable. After a couple of transfers I need to reboot the tablet and computer.>>>
>>
>>Did you connect one of your malfunctioning Bluetooth devices to pc?

Yes.

I'd like to note again that both Bluetooth devices are working. The problem is that when both devices are connected (!) only one is working ( a device that was connected first ).

>>I would like to know if the same MAC (on both devices) address could be a reason for the inactive bluetooth device..

Yes. They have identical MAC addresses.

>>>>'Network to Access' - a setting is 'Any available network'>>>
>>
>>Is this Android setting?

No. It is the Windows XP setting.

>>>Yes.

I'd like to note again that both Bluetooth devices are working. The problem is that when both devices are connected (!) only one is working ( a device that was connected first ).>>>
By simple logical reasoning yor problem boils down to two identical MAC addresses beign active on your local network.On such a network two or more devices cannot have the same address.One of the option to solve this problem is to reprogram your Bluetooth device MAC address register I posted two links.Other option is to change somehow Win Bluetooth driver stack which is responsible for the device connection and identification.
Maybe you could call SAmsung tech service and talk to them.

Regarding wi-fi connecting have you proceeded like explained in this article :http://www.labnol.org/software/ad-hoc-wireless-network-in-windows-xp/11486/

Slightly off topic question
Have you tried to write inline SSE assembly elementary functions class or program.My problem is that every function grows to more than 100 lines of code because of structures used to pass vectors to XMM registers.How such a issue could be implemented more efficiently.

Hi Iliya,

>>Have you tried to write inline SSE assembly elementary functions class...

No. My exposure to SSE is very limited due to portability constraints of the project.

Would you be able to post source codes for one of your not-optimized function? Is it in C/C++ or Java?

Best regards,
Sergey

By the way, I'd like to make a comment on a battery life of the Samsung Galaxy Tab 2 (7").

It worked for 9 hours 31 minute in one of my test without re-charging. So, it was pretty impressive and I strongly recommend the Samsung Galaxy Tab 2 tablet ( especially 10" over 7" ).

>>...
>>Regarding wi-fi connecting have you proceeded like explained in this article:
>>
>>http://www.labnol.org/software/ad-hoc-wireless-network-in-windows-xp/11486

I'll check it tomorrow. Thanks, Iliya!

>>>No. My exposure to SSE is very limited due to portability constraints of the project.

Would you be able to post source codes for one of your not-optimized function? Is it in C/C++ or Java?>>>

Yes of course I will post source code in my sine function thread.My intention is to write static library with elementary and special function designed as a vectors.Every vector is build from double and float primitives and can contain between 2 to 4 scalars so I can effectively do SIMD - style calculations.
My library will be written in C++ and will contain only static functions.I will also perform an extensive testing of the execution speed and post the results.

@Sergey
I posted source code of vectorised fastsin4D() function which returns a 4D component vector of sine float values.

>>>It worked for 9 hours 31 minute in one of my test without re-charging. So, it was pretty impressive and I strongly recommend the Samsung Galaxy Tab 2 tablet ( especially 10" over 7" ).>>>
Very impressive indeed, but one thing bothers me how Samsung engineers could overlook Mac address issue?
I strongly advise you to call Samsung tech support and talk to them.
Was your Galaxy somehow modified befory you purchased it?

>>>I'll check it tomorrow. Thanks, Iliya!>>>
Please read this article very interesting here is link : http://www.pcadvisor.co.uk/how-to/google-android/3372947/connect-android...
Please read also this article it explains how to enable wi-fi connection on Android device link here :http://androidtablethacks.com/android-tablet-hacks/how-to-enable-adhoc-w...

>>...Was your Galaxy somehow modified befory you purchased it?

No. It is a regular US/Canada version and it was purchased for R&D.

>>>No. It is a regular US/Canada version and it was purchased for R&D.>>>
What surprises me how those two units got the sam MAC address.Vendor ID is first 2 or 3 octets and Device ID is remaining 4 or 3 octets so the
adresses range is very large.Maybe someone programming mistake.

>>I also checked Windows Event Viewer logs and I found a message:
>>
>>'TCP/IP has reached the security limit imposed on the number of concurrent TCP connect attempts'

It happens every time when I open too many copies of Internet Explorer when browsing the web. A Windows utility 'Netstat' with option '-no' helped to clear that issue.

>>What surprises me how those two units got the sam MAC address.Vendor ID is first 2 or 3 octets and Device ID is
>>remaining 4 or 3 octets so the adresses range is very large. Maybe someone programming mistake.

I can't find another explanation. Also, this is what somebody posted on:
http://androidforums.com/t-mobile-g1/3506-changing-spoofing-mac-address....

...Part of the MAC address is the manafacturer ID which is normally hard coded, the remainder of the address is left to be incremented when the card is rolled out, so supposedly no two addresses are ever the same...

>>>It happens every time when I open too many copies of Internet Explorer when browsing the web. A Windows utility 'Netstat' with option '-no' helped to clear that issue.>>>
Did you use NetMon to find misbehaving process?

>>>I can't find another explanation. Also, this is what somebody posted on:>>>
Strange that QC did not catch and fixed that issue.

>>>I can't find another explanation. Also, this is what somebody posted on:
http://androidforums.com/t-mobile-g1/3506-changing-spoofing-mac-address....>>>

Will you try to do this , as kind of last resort action?

>>Will you try to do this , as kind of last resort action?

No. Two new Bluetooth devices version 4 will be used.

>>>Two new Bluetooth devices version 4 will be used.>>>
Are you going to use Bluetooth devices version 4 instead of Galaxy?

>>>>Two new Bluetooth devices version 4 will be used.
>>
>>Are you going to use Bluetooth devices version 4 instead of Galaxy?

No and the tablet will be used anyway. Here is a list of configurations:

- connected to
- connected to
- connected to
- on the same computer ( to simplify testing )

[Updated]

>>> Desktop with Windows7Pro ( as soon as I have one ) and BTv4 device connected to Samsung Galaxy Tab 2.0 tablet>>>
What is BTv4 device?
Is this bluetooth receiver?

>>...Is this bluetooth receiver?

Yes. Sorry if I didn't express it clearly.

>>>Yes. Sorry if I didn't express it clearly.>>>
its ok.
Please post the update of your testing.

Hi Iliya. I've updated a post with a list of configurations: I've added square brackets to identify systems, like connected to .

connected to
- connected to

Sergey in post from Tue, 11/06/2012 - 07:11 you included two pics of Bluetooth devices.Are those Bluetooth radio receivers or Galaxy tablets?

>>... in post from Tue, 11/06/2012 - 07:11 you included two pics of Bluetooth devices. Are those Bluetooth radio receivers ...

Yes, and as I've already described, both have the same MAC address.

>>>Yes, and as I've already described, both have the same MAC address.>>>
Now I understand it more clearly.I have been thinking that Galaxy tablets have the same MAC address.
Were those receivers supplied with tablets?

>>...Were those receivers supplied with tablets?

No. They were purchased online at some computer store about 6 months ago. Unfortunately, it is too late to ask for an exchange and I will simply buy another two that support Bluetooth version 4.

>>>No. They were purchased online at some computer store about 6 months ago>>>
When you will be installing your new devices I would recommend to check and monitor devices status with the help of usbview tool.Please consider also this tool link :http://www.pcausa.com/Utilities/UsbSnoop/default.htm
Can you also check receivers MAC address before buying them?

>>...Can you also check receivers MAC address before buying them?

Of course No but I will exchange a device(s) in case of MAC-address problems.

>>>Of course No but I will exchange a device(s) in case of MAC-address problems.>>>
Can not you buy those receivers personally?Maybe you will be given an opportunity to connect them to your laptop and check devices MAC addresses.

>>>>...In order to eliminate this try to send files by Bluetooth to Win7 machine or send files between two Android devices...
>>
>>As soon as I have a new computer with Windows 7...

I checked how these Bluetooth adapters are working on a 64-bit Windows 7 Professional and the problem is reproduced as well. Here are two warning messages from the Event Log:

Warning Message 1: Only one active Bluetooth adapter is supported at a time.

Warning Message 2: Windows cannot store Bluetooth authentication codes (link keys) on the local adapter. Bluetooth keyboards might not work in the system BIOS during startup.

Páginas

Deixar um comentário

Faça login para adicionar um comentário. Não é membro? Inscreva-se hoje mesmo!