Cannot create provider in linux (sometimes)

Q. I'm developing first JTAPI application and I'm totally stuck.

I have written small JTAPI application to forward calls. In Windows environment (Windows 10, JRE 1.8.0_66) everything works absolutely fine.

But if I try to run the application in Linux environment the application works only sometimes -  let's say 1 of 4 runs are OK.

If the application is not working I'm getting this exception when trying to create a JTAPI connection to CUCM:

Exception in thread "main" com.cisco.jtapi.PlatformExceptionImpl: Unable to create provider -- null
at com.cisco.jtapi.CiscoJtapiPeerImpl.getProvider(CTQF)
at test.Redirect.<init>(Redirect.java:66)
at test.Test.main(Test.java:22)

I'm using Red Hat x64 and JDK/JRE 1.7.0.40 - I tried many other version of Java, but without any success. I also tried other distributions without success. The absolutely same code is running without any problems in Windows.

I also tried to compile it in Windows and transfer to linux, compile it in Linux using different versions of JDK, running with different JRE but nothing helped me.

The example application from Cisco (makecall) has the same symptoms as my application.

The application fails at this code:

provider = (CiscoProvider) peer.getProvider(providerString);

I'm attaching the JTAPI log. This is the most problematic part of the log, I suppose:

49: Aug 19 04:14:19.645 EDT %JTAPI-CTIIMPL-7-UNK:(P1-10.128.100.130) ReceiveThread: caught java.net.SocketException: Connection reset
50: Aug 19 04:14:19.645 EDT %JTAPI-MISC-7-UNK:(P1-10.128.100.130) ReceiveThread: exiting
51: Aug 19 04:14:19.645 EDT %JTAPI-MISC-7-UNK:(P1-10.128.100.130) ReceiveThread: notifying handler of demise
52: Aug 19 04:14:19.647 EDT %JTAPI-PROTOCOL-7-UNK:(P1-10.128.100.130) received Event: com.cisco.cti.protocol.ProviderOutOfServiceEvent {
eventSequence = 1
PROVIDER_OUT_OF_SERVICE_EVENT = 200
}
53: Aug 19 04:14:19.648 EDT %JTAPI-MISC-7-UNK:(P1-10.128.100.130) EventThread: queuing com.cisco.cti.protocol.ProviderOutOfServiceEvent
54: Aug 19 04:14:19.648 EDT %JTAPI-CTIIMPL-7-UNK:ResponseTable: received unexpected response sequenceNumber #-1914896383
55: Aug 19 04:14:19.648 EDT %JTAPI-MISC-7-UNK:(P1-10.128.100.130) ReceiveThread: shutting down
56: Aug 19 04:14:19.648 EDT %JTAPI-MISC-7-UNK:(P1-10.128.100.130) ReceiveThread:java.lang.NullPointerException
57: Aug 19 04:14:19.648 EDT %JTAPI-MISC-7-UNK:(P1-10.128.100.130) ReceiveThread: finished shutting down

I'm using CUCM 10.5.2.14072-1, Cisco JTAPI version 10.5(2.10000)-3 Release (downloaded from the CUCM), JRE/JDK 1.7.0_40 64bit.

Did anyone saw this situation before? Any help is greatly appreciated

A. The socket used by Cisco JTAPI to communicate with CTIManager is running into error. You may want to start the investigation on the network side by collecting the wire shark traces. You may want to pull CTIManager logs as well to see if they have any additional info.

Check if there is any firewall/anti virus s/w interfering with the port (0xABC) used by Cisco JTAPI.