TL;DR: Because of the two uneven units of guidelines for substitute, a complete of 5 transactions is enough to make a cycle of replacements that beat one another in flip. This enables an attacker to rebroadcast the identical transactions again and again whereas at most paying for one small transaction per block.
You’ve got two confirmed UTXOs C1
and C2
. Let’s say 20 s/vB is the underside of the primary block.
- You create a big low-feerate transaction
tx_LL
with 100,000 vB at 1 s/vB (charge: 100,000 s). It spends the confirmed outputC1
and has an outputtx_LL:0
. - You connect a small low-feerate transaction
tx_LS
as a baby with 100 vB at 1 s/vB (charge: 100 s) by spendingtx_LL:0
.
- You RBF
tx_LS
with a high-feerate transaction that spendsC2
andtx_LL:0
in a brand new transactiontx_HS
.tx_HS
has 5000 vB and pays 21 s/vB, however because it spends an output from a low-feerate mother or father, its mining rating is only one.95 s/vB.
- You RBF
tx_LL
andtx_HS
withtx_LM
that has 100,000 vB and pays 3.05 s/vB (charge: 305,000 s) by spending the outputsC1
andC2
. That is permitted, since solelytx_LL
is a direct battle, so the feerate oftx_HS
doesn’t must be beat immediately.
- You employ the brand new RBFr guidelines to switch
tx_LM
with a small excessive feerate transactiontx_RBFr
with 100 vB paying 20 s/vB (charge: 2000 s) that spendsC2
and makes it into the highest block of the mempool.tx_LM
was not going to be within the subsequent block, andtx_RBFr
pays greater than 1.25× the feerate oftx_LM
. So that is permitted beneath the brand new guidelines.
- You then rebroadcast
tx_LL
andtx_LS
as a result ofC1
is now not being spent.
- You instantly exchange each
tx_LS
andtx_RBFr
withtx_HS
.tx_HS
has a feerate of 21 s/vB which is increased thantx_RBFr
(20 s/vB) andtx_LS
(1 s/vB), and pays extra absolute charges than each (105,000 s vs 2000 s + 100 s). However because it’s a baby oftx_LL
it solely has a mining rating of 1.95 s/vB.
Repeat 4.–7. to make each node on the community cycle the identical 5 transactions advert nauseam. Roll the locktimes or sequences to make the transaction have a brand new TXIDs in every iteration, whereas spending the identical UTXOs. The one transaction that’s ever in any hazard of getting mined is tx_RBFr
which prices you 2000 s. If it it does get included in a block, simply begin over with a brand new confirmed UTXO as your c2'
.