I ALSO Cannot recognize DTMF tones

[stipus]

I had an idea that "could" help with finding what's happening.

- Create a small IVR script that records the call to a wav file. I can help on this point if you need.

- Place several calls, and hit a few DTMF keys each time.

When you succesfully get one wav record with DTMF sounds but no DTMF recognition, send this it to me, together with a working wav file.

This way I could compare both wav records, and try to determine why there is no recognition in some cases. I could also replay the wav file to the DTMF recognizer at will, and see if changing some parameters would help.

[bignick]

I think this might help clear up the problem. HS finally automatically produced an error, which I fwd'd to their HelpDesk. They responded that it's something you need to look at. Also, they recommended that I upgrade to HS 2.3.0.19 because, as they said, version 2.3.0.0 contained a "lot of errors". So I did. I am running another set of tests right now with the new HS version.
The original error's diagnostics are shown below. Perhaps you can get a clue from that whether it would be HS2 that's faulty or whether it is the plugin. Personally, I'd like to complete my tests with the new HS before asking you to go on a wild-goose chase.

To put the error below in context: I have an event that fires every 5 seconds which runs a script that opens a SQL server to check for a status change. If there is a status change, depending on the status, HS makes a call to the telephone number on record and the person receiving the call is expected to enter certain information with the telephone keypad to complete the purpose of the call. That's when the DTMF recognition sometimes fails. Sometimes the person is required to call into the system and enter information; that's when the DTMF works perfectly and I couldn't get it to fail yet, even from the very same telephone.

At first I suspected a stack overflow or perhaps objects that didn't close properly because with 5-second intervals one could quickly exhaust the server's resources. But I checked them carefully and the code is cleaned up properly. It ran for about 36 hours before HS reported the error.

I hope this helps.

Thanks

Nico

ERROR TEXT:

Object reference not set to an instance of an object.
HomeSeer version: 2.3.0.0
Source: hspi_skype
Target: Void m_outputSocketReceivedDataReadyHandler(System.IAsyncResult)

Stack trace: at HSPI_SKYPE.SkypeCall.m_outputSocketReceivedDataReadyHandler( IAsyncResult ar) at System.Net.LazyAsyncResult.Complete(IntPtr userToken) at System.Net.ContextAwareResult.CompleteCallback(Object state) at System.Threading.ExecutionContext.runTryCode(Object userData) at System.Runtime.CompilerServices.RuntimeHelpers.ExecuteCodeWi thGuaranteedCleanup(TryCode code, CleanupCode backoutCode, Object userData) at System.Threading.ExecutionContext.RunInternal(ExecutionConte xt executionContext, ContextCallback callback, Object state) at System.Threading.ExecutionContext.Run(ExecutionContext executionContext, ContextCallback callback, Object state) at System.Net.ContextAwareResult.Complete(IntPtr userToken) at System.Net.LazyAsyncResult.ProtectedInvokeCallback(Object result, IntPtr userToken) at System.Net.Sockets.BaseOverlappedAsyncResult.CompletionPortC allback(UInt32 errorCode, UInt32 numBytes, NativeOverlapped* nativeOverlapped) at System.Threading._IOCompletionCallback.PerformIOCompletionCa llback(UInt32 errorCode, UInt32 numBytes, NativeOverlapped* pOVERLAP)

[bignick]

Now I have another problem (after the HS upgrade; I am not sure it's related).
When the database status changes, the call is initiated, but Skype returns a FAILUREREASON 7 and it doesn't make the call.
Can you quickly eye-ball the log entries below and see if it's something obvious? I replaced the last four digits of the telephone number with ****
Thanks



1/15/2009 4:12:09 PM - Skype Debug - Skype Connector : Placing Skype Call to +1818917****
1/15/2009 4:12:09 PM - Skype Debug - SkypeCommandEvent - #78 CALL +1818917****
1/15/2009 4:12:09 PM - Skype Debug - SkypeResponseEvent - #78 CALL 2049 STATUS UNPLACED
1/15/2009 4:12:09 PM - Skype Debug - Creating new call '2049'
1/15/2009 4:12:09 PM - Skype Debug - SkypeCommandEvent - GET CALL 2049 PARTNER_HANDLE
1/15/2009 4:12:09 PM - Info - Speak ():call customer
1/15/2009 4:12:09 PM - Skype Debug - SkypeResponseEvent - CALL 2049 PARTNER_HANDLE +1818917****
1/15/2009 4:12:09 PM - BEL - +1818917****
1/15/2009 4:12:09 PM - C4A3 - Authorization=bel;+1818917****
1/15/2009 4:12:10 PM - Skype Debug - SkypeResponseEvent - CONTACTS FOCUSED
1/15/2009 4:12:10 PM - Skype Debug - SkypeResponseEvent - CALL 2049 STATUS UNPLACED
1/15/2009 4:12:10 PM - Skype Debug - SkypeResponseEvent - GROUP 9 NROFUSERS 2
1/15/2009 4:12:10 PM - Skype Debug - SkypeResponseEvent - GROUP 15 NROFUSERS 3
1/15/2009 4:12:10 PM - Skype Debug - SkypeResponseEvent - CALL 2049 STATUS ROUTING
1/15/2009 4:12:10 PM - Skype Debug - SkypeResponseEvent - CALL 2049 FAILUREREASON 7
1/15/2009 4:12:10 PM - Skype Debug - SkypeResponseEvent - CALL 2049 STATUS FAILED
1/15/2009 4:12:10 PM - Skype Debug - SkypeResponseEvent - CALL 2049 VAA_INPUT_STATUS FALSE
1/15/2009 4:12:14 PM - Info - Event Trigger "checkAuth"

[stipus]

Skype doc states FailureReason 7 is "Sound IO error".

Can you check that there is no problem with your soundcard. I think it must be available at the beginning of the call. What's happening if you try to place a manual call from the Skype Client ?

I'll double check my code for the other error you experienced.

[bignick]

OK, my bad. Thanks, I was running my application through terminal services, which couldn't properly connect to Skype. Running it on the console works fine.
Hopefully the HS error will be resolved through the upgrade. I will check and let you know.
Thanks for your help, Stipus.

[stipus]

I have also noticed the same (as I'm accessing my developement computer with RDP): The Skype Client doesn't work through a RDP terminal session

To work around this problem, I did install VNC. It's about the same as RDP, but the Skype Client do work.

I did also fix any potential "Object reference not set to an instance of an object" error in the function m_outputSocketReceivedDataReadyHandler. I'll release an updated version soon.