For unknown reason NULL PLCI queue is not properly processed while line interconnect is in process. The queue grows from line interconnect to line interconnect and causes long delay. This behavior to be investigated. Current solutionis to drom queue if exceeds practicelly acceptable limits

divastreaming/chan_capi_divastreaming_utils.c

@@ -84,6 +84,7 @@ static int divaStreamingMessageRx (void* user_context, dword message, dword leng
 								struct capi_pvt* bridgePeer = pE->i->bridgePeer;
 								if (bridgePeer->NCCI != 0 && bridgePeer->diva_stream_entry != 0 &&
 										bridgePeer->diva_stream_entry->diva_stream_state == DivaStreamActive &&
+										bridgePeer->diva_stream_entry->diva_stream->get_tx_in_use (bridgePeer->diva_stream_entry->diva_stream) < 512 &&
 										bridgePeer->diva_stream_entry->diva_stream->get_tx_free (bridgePeer->diva_stream_entry->diva_stream) >
 																																																2*CAPI_MAX_B3_BLOCK_SIZE+128) {
 									dword i = 0, k = 0, b3len;