static withdrawal - min relay fee not met

[RocketNodeLN]

Tried to withdraw 2 utxos from static address at 1 s/vb

loop s w --utxo A --utxo B --sat_per_vbyte 1

Expected behavior

Withdraw 2 utxos to lnd's wallet

Also, would like the raw tx to be in the logs.

Actual behavior

[loop] rpc error: code = Unknown desc = undefined: min relay fee not met

Restarting loopd shows this in the logs:

[DBG] SADDR: Recovering deposit ABCD
[DBG] SADDR: Deposit UTXO:1: NextState: Deposited, PreviousState: Withdrawing, Event: OnRecover
[DBG] SADDR: Recovering deposit EFGH
[DBG] SADDR: Deposit UTXO:2: NextState: Deposited, PreviousState: Withdrawing, Event: OnRecover
[DBG] SADDR: Deposit UTXO:1: NextState: Withdrawing, PreviousState: Deposited, Event: OnWithdrawInitiated
[ERR] SADDR: deposit-withdrawal-TXID: rpc error: code = Unknown desc = undefined: min relay fee not met
[DBG] SADDR: published deposit withdrawal with txid: TXID
[DBG] SADDR: Deposit UTXO:2: NextState: Withdrawing, PreviousState: Deposited, Event: OnWithdrawInitiated
[ERR] SADDR: deposit-withdrawal-TXID: rpc error: code = Unknown desc = undefined: min relay fee not met
[DBG] SADDR: published deposit withdrawal with txid: TXID

However, the utxos are still shown in loop s l

To reproduce

Unable to reproduce with same utxo

[loop] rpc error: code = Unknown desc = deposits to be withdrawn are unknown or inactive

System information

[INF] LNDC: lnd version: v0.19.1-beta, build tags 'autopilotrpc,signrpc,walletrpc,chainrpc,invoicesrpc,watchtowerrpc,neutrinorpc,monitoring,peersrpc,kvdb_postgres,kvdb_etcd,kvdb_sqlite'

$ loop -v
loop version 0.30.0-beta commit=```

[RocketNodeLN]

Update: restarted bitcoind with lower min relay fee and the tx was published. The tx created by loopd was 168.5 vB, but fee was 168 sats

[alexbosworth]

this was resolved by updating loopd to the latest version?

[RocketNodeLN]

Updated to 0.31.2-beta and was able to withdraw at a higher fee rate. However, the size estimation is still off as it did not use 168.5 vB as the tx size.

[alexbosworth]

Thanks for the update, OK so there seems to be a weight to vb rounding error in this case