amsmota

Hi @siladu and @matthew1001 can youse please review this one? Cheers.

> Left a few more comments. Was there any follow up to my earlier comment about what the behaviour should be if a new TX arrives during an empty block...

It is not the behaviour I see in my tests. Did you saw the comment I mentioned above? https://github.com/hyperledger/besu/issues/3810#issuecomment-2053663437 ``` 2024-04-13 **15:10:57**.174+01:00 | QbftBesuControllerBuilder | Imported empty block #49 /...

> Sorry I couldn't find your previous comment, but that's great if that's the current behaviour. I think I pulled and re-built with your latest changes but I'll check. Here's...

Just tested again the emnpty/non-empty block times, it seems good to me. It's based on > "blockperiodseconds": 10, > "emptyblockperiodseconds": 30, The block that contains 1 tx is the line...

One thing I notice in those logs above, in the previous logs I copied from tests all(most) blocks, profuced or imported, appeared on the QbftBesuControllerBuilder output, which is not the...

> > Anyway, to be sure is better if somebody else confirms this data. > > I'll pull your latest commits today and try with a 1 node and 4...

Hi @matthew1001 , that seems different from what I saw in my one-node test, but it's hard for me to test correctly with only remix and metamask. I'm not very...

Here's a test I did today with a one-node network based on a 5/10 seconds intervals. Block 3 (the first non-empty) took `10 seconds when it should take 5. However,...

Here's my latest test with 5/10 seconds, one-node network. I logged some messages to try and understand the issue. ``` 2024-05-02 18:37:46.387+01:00 | BftProcessorExecutor-QBFT-0 | INFO | QbftBesuControllerBuilder | Produced...