We have an application that uses JTAPI SDK to monitor devices. The main problem is provider shutdowns many times in a day with errors like ?AUSE_SNAPSHOT or Provider Transmission Ended.

Also sometimes application didn't get call events despite that addCallListener added to the address.
Can you help to find the cause of this problem and how we can solve it?

A common cause of provider shutdowns is that the AE Service has sent several messages to the application but not received an acknowledgement. The two main reasons for this are:

1. Network issues. The connection between AE Services and the computer running the application fails.

2. Poor handling of threads by the application. When a listener method is called, it should do no processing on the event/response in the callback. It should never make any API calls. Instead, all processing of the event should be done using a thread owned by the application. The listener thread must be released back to the JTAPI client as quickly as possible so it can be used to receive and acknowledge the next event.

If you are seeing this problem in your lab and need further help, you can open a Technical Support ticket with Devconnect (there may be a fee for this service). If your application has been compliance tested by Devconnect and the problem is in production, you should ask your customer to open a support ticket with Avaya Services who should be able to help you to investigate.

Martin

to troubleshoot #1 use a packet sniffer.
to check for #2 use the log files on AES, there should be an indication that the application is not responding and the stream is being closed.