Hi all, after finishing my house (well I guess a house is never finished, but let’s say, I’ve reached the 1.0 version) and moving in (well at least my stuff is contained in the house … will need some defragmentation and unzipping of archives), I can finally concentrate on the important things again :)
So as I reported two weeks ago, I have finished a first working version of the PDU fragmentation. This was quite tricky as I had to also respect the Max AMQ Caller value which restricts the number of unacknowledged messages a PLC is willing to accept. If more messages are sent, the PLC simply declines them. Now we have some cool queueing mechanism that takes care of not sending more unacknowledged messages, than the PLC is willing to accept. My current solution simply splits every read up into one message per address range. This allows really big read/write requests, but drastically reduces the performance. What I’m working on now is to implement some naïve algorithm, that simply adds new items as long as the request or response size doesn’t exceed the maximum PDU size. This would probably be a state in which I think we should probably do a little testing but that could be part of a first release. In parallel we have done the wiring for our IoT network. At least Sebastian and I can now access all of our PLCs in the office … we are hoping to be able to provide VPN access to all PLC4X committers quite soon. I think this would make working on the protocols a lot easier for you guys. Sebastian is making good progress on the Beckoff side. On Friday he even showed me first results using a subscription … so we will probably be extending the PLC4X API to subscription types of operations … really looking forward to that :-) Hope you are all well and that we’ll be able to get the lights back on here as soon as possible :-) Chris
