Followup from #19969 #20032

In #19969 @kallewoof mentioned:

I find it weird that even though I used "1 sat/b" it complains about failing to do fee estimation. It shouldn't need fee estimation at all. Anyway, that's outside the scope of this nit-fix PR.

Thanks for creating an issue. ~I believe this problem does not only relate to the send RPC, but to all of them, including sendtoaddress, sendmany.~ (It only applies to send, I was mistaken.)

Edit: Verified to be incorrect, the bug only applies to send.

It is probably best to fix this as part of #19543

@kallewoof any particular steps to reproduce? I've been using the explicit feerate frequently (not yet with the new send rpc though) and have not seen anything out of the ordinary.

Actually, I was wrong. It only applies to send:

$ ./bitcoind -regtest -daemon
Bitcoin Core starting
$ ./bitcoin-cli -regtest getnewaddress
bcrt1qkfveed8g8vnktsc3490aarzh302l2qc2ymhlge
$ ./bitcoin-cli -regtest sendtoaddress bcrt1qkfveed8g8vnktsc3490aarzh302l2qc2ymhlge 1 '' '' false true 1 sat/b
a93d1ccee5bdca288908a97f5a6801d91cc8a174a5ed3ad10c8cf22e84e9f6e8
./bitcoin-cli -regtest sendmany "" '{"bcrt1qkfveed8g8vnktsc3490aarzh302l2qc2ymhlge":1}' 0 '' null true 1 sat/b
c26f6c2d404ae1228dd7d53a37e70b87e546f7138fc88efc800c208f8733a60a
$ ./bitcoin-cli -regtest send '[{"bcrt1qkfveed8g8vnktsc3490aarzh302l2qc2ymhlge":1}]' 1 sat/b
error code: -4
error message:
Fee estimation failed. Fallbackfee is disabled. Wait a few blocks or enable -fallbackfee.

Thanks @kallewoof

Is this still an issue?