Are you sure that you really excluded the folder? Can you double-check? I'm asking because white-listing my "miners" folder appears to be working fine for me (Windows 10 Fall Creators Update). Did you try white-listing the folder and re-downloading the miner directly into that folder?

Found the remaining puzzle piece.

I did not purposefully exclude anyone. Could I have the Verge address you've been mining with to take a look?

The verge-blockchain.info block explorer is unfortunatly quite limited. It's supposed to work with Block numbers when it in fact only works using block hash links. What works is entering the Block Number 1890557 into its search box.

Have to track this for a while before a making a decision: https://www.reddit.com/r/Monero/comments/7xmfax/i_have_just_been_banned_from_rmonerov_guys_please/

Added.

If there's demand it can be added. Hopefully Sia has a working Testnet. :)

You are correct. Share counts reset after a round but that only happens if it was a round with effort >= 100%. The round for the first Verge-Lyra block was extremely short (< 10%) that's why shares appear as mostly unmodified. It's also the reason why payments for that block went out in two batches. The first wave was quite low because there were not enough shares to distribute the full block reward to miners. That's why I've manually adjusted miner balances to distribute the remaining block reward which triggered the second wave of payments.

Regarding your initial post: Blocks normally appear right after they've been found. The situation with both Verg Pools is currently the following: Blocks are found but they pool is unable to confirm that they've been actually accepted by the Node after submission due to one remaining problem I'm currently working on which is a difference between the block hash the pool computes to confirm the block vs. the block hash that's recorded on the block chain. The result is that I'm currently manually inserting found blocks into the database. Logging is in place that should reveal the correct block hash solution when the next block is found.

Checked it and server side is working as usual. No increased latency.

Yes that's what I'm saying. Other pool ops might have decided to cover it up. I didn't. 

When you have to work with a crypto-currency that has no working Testnet like Verge, things can go wrong in production. I spent the entire sunday working on fixing the problem and I'm confident I did. I understand that miners get pissed when something like this happens. Believe me so did I. If you have a Dash, LTC or ETH wallet, send your address to contact@coinfoundry.org and you'll receive some compensation for your wasted time.