Problematic fee estimation in coordinator

[batmanbytes]

https://mempool.space/testnet4/tx/c8f8961bf0cc51e8f8607eaf9e129537e8d5d96a01f4db6376b6189e82c5b9af

At the time of broadcasting this coinjoin, my coordinator's mempool looked like this:

root@coordinator-test:~# bitcoin-cli --testnet4 getmempoolinfo
{
  "loaded": true,
  "size": 1821,
  "bytes": 334664,
  "usage": 2608512,
  "total_fee": 0.00346967,
  "maxmempool": 300000000,
  "mempoolminfee": 0.00001000,
  "minrelaytxfee": 0.00001000,
  "incrementalrelayfee": 0.00001000,
  "unbroadcastcount": 0,
  "fullrbf": true,
  "permitbaremultisig": true,
  "maxdatacarriersize": 100000
}

As you can see, it's less than a block long (334kb). The network is not congested at all. Therefore, the 13 sat/vb that the coordinator selected is clearly a waste of money.

In my coordinator/Config.json, I have "ConfirmationTarget": 3. What could be the issue here?

[Kruwed]

Wasabi listens to whatever feerate your node says, and Bitcoin Core's fee estimation is complete shit. There is an additional sanity check (at least on mainnet) that queries the current fee rate from mempool.space and sets that as the ceiling.

[batmanbytes]

I see, thanks. I remember Bitcoin Core being complete garbage on estimating fee rate. I wonder how they haven't resolved this yet.