Thx @Heinz !
It was not planned to release it. I do this work to see the dynamic of mash functionality. To make it public the interest must be very high, I had to invest a lot of time, and I should not become in conflict with Myscontroller (best tool, I lerned a lot at the use). In general I hope it improves the quality of mysensors, because I have to deside to use it or not.

@Heinz The GW is connected online to my webside and a php script collect all data and add it to a mysql database. From this DB the structure is rendered by an js library. The structure and the sensor values are updated frequently (not shown in this screenshot - take a look to ) ... in parallel I can connect the server with a controller or with MYSContoller via TCP.

If I power up the GW as first ...

Multi messages like

0;3;0;9;1;foo\n0;3;0;9;2;bar\n0;3;0;9;3;foo\n0;3;0;9;2;bar\n

must be understand also. I got it from the tcp stream of the actual ether-gw.

Here is a screenshot of the visualizing of all node types (not all sensor types)

Hi, I build up DB entrys from the informations about the notes, repaters and the GW and create a live visualization in a browser. The goal was to see the structure of the mash and see the routing. If I power up the GW as first and then the nodes it will be worked very fine. In the other order there are missing things (i.e things the bootloader send). But the worst thing is that i'm not able to get a feedback from the node/mash about there presens (health) at a later time.

So it would be great to have:

-a function to scan the network (without to reset all nodes)(except sleeping nodes).

• get information about battery powered (/sometime sleeping node) and full powered (every time responding) node.
• In the case of a battery powered node - the info about the minimal notification time would be very interesting. So I would be able to say: "The minimal notification time is 1day, I didn't hear something from this node in the last 3 days, so I change the health state from green to yellow (warning) and again 3 days later to red (error).
• Or, and this is my request in general for powered devices. Set the time of minimal keep_alive messages.

A littel bit other topic:

• OTA BL: To reduce the risk of a wrong firmware the configuration of the hardware is interesting. Means mainly running at 8 or 16Mhz. (( Maybe also the µC (future), the connection's schema of the nrf24 or in general the connected communication module - if you also plan to connect things like the rf69).

Tell me if I overlook such features.
-Pit

As I get my fakes I open a dispute and add this info - it war clear enough to get my money back ;-):

So, the chips differ in the register number 09 (CD/RPD).

@mainali The chips on your first picture I got also one time. IMO it is the nRF24L01 without the plus '+' - I detect it by comparing the register resonses with there datsheets.

@dzairo Late, but to answer your question: It's also XTEA with a compiled master-key and variations over node id and rf freq. And you are right, on repeated date without random input it's a bad solution... Thx for hint... -Pit

Yes - noooo Right - you are able to detect the connection loss (good to know) but why do you not trigger the NE555, but to re-init the ENC28. Have you try if it response via SPI ?