Almost there now

I don't think counterparty is ready for us

Or you're not ready for counterparty?

I updated my node, deleted all DB files, and did a full parse.

My txlist hash does not match other nodes. Neither does tokenscan match horizon.

The ledger hash, however,  is consistent across nodes.

Check a specific block on the explorers.

tokenscan has corrupted balances for some addresses.
For instance this one is empty but tokenscan shows 163,084 XCP

https://www.tokenscan.io/address/1ML2b9tY5V8S9qQw6jNUs5uxkm6nKayk6x

Also it has had this maintenance banner for many weeks and nobody cares lol

Provide an example. I checked block 928694 and it's identical.

Update:

I deleted the DB files and started parsing from the beginning. My wifi internet dropped after about 12 hours. Unable to reconnect. I suspect bitcoin core is to blame here.

Counterparty entually crashed tho. Block hashes at this height, 782070, are identical to horizon and tokenscan.

I will restart the VM and try to contine from where it crashed

What's that xcp virtual machine and is anyone running a node without virtual servers??

There is no xcp virtual machine.

My laptop runs Windows but I use an Ubuntu VM for Counterparty.

It is impossible to run Counterparty directly in Windows, or at least there's no easy way to do it.

My node finished parsing blocks. Unfortunately both txlist and messages hashes diverge.

I will pinpoint the exact blocks (where divergence first starts) and file a github issue.

@jp_janssen thanks for finding this issue! If you're actually able to share your whole DB, that might make it even easier to identify the point of divergence. The problem is very likely some subtle caching logic

JP J thank you for assisting will try ubuntu.

It's almost certainly not an OS problem. It's probably a race condition.

Ok, i will try to share the db tomorrow.