Delays in transmission process

[sopgreg]

I'm experiencing long delays between the transmission statuses of up to a minute (without retries), see screenshot. Is this normal behaviour which can be configured, or is it a bug?

[sopgreg]

Some further observations:

• the 10 second gap between READY_TO_PUSH and PROCESSING seems to be the regular interval of the SenderWorker. Reducing the interval to like two seconds will also have an effect. It would be useful if the few lines of code pro processing a message in SenderWorker could be executed right away when a message is set to READY_FOR_PUSH, without any interval inbetween
• the 5 second gap between SENDING and TRANSPORT_FAILURE is okay and expected, this is a regular 5000ms HTTP timeout in my case (other MSH not reachable)
• the 40 seconds delay between TRANSPORT_FAILURE and FAILURE seems to be again a result of the interval of the RetransmissionWorker plus some additional delay. This should be eliminated completely, if no retransmission is required per Leg definition

[RenateS]

This is normal behaviour. Such a long delay can indeed happen in a situation where the maximum intervals apply. As most users do use retries, immediately checking retry configuration after a transmission failure would not have a major impact on performance. We may look into this for a future version but it has lower priority. To reduce the 40 second delay you could decrease the worker interval of the RetransmissionWorker.