You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, the relayer submits a block commit as soon as it detects either:
1. A new burn block.
2. A new highest tenure.
However, committing immediately upon seeing a new burn block—before confirming whether a tenure change block follows—is risky. If a tenure change block arrives afterward, the miner’s commit will confirm the wrong tenure, making it impossible to mine a valid Stacks block.
This risk can sometimes be rewarded:
• The miner’s commit might be the only one in a flash block, making it more likely to win, though the ATC-C rules reduce this probability
• If the previous miner failed to produce a tenure change block before the next Bitcoin block, the commit could still be valid.
• Even if a tenure change block is mined, the commit could remain valid due to the bad timing reorg allowance.
However, in the common case, the miner must quickly replace-by-fee (RBF) the early commit with a new one referencing the correct tenure. If this RBF fails, the miner’s PoX payments are wasted.
Proposed Change
Modify the default behavior to be more patient by waiting a configurable delay (e.g., ~30s) after detecting a new burn block before submitting a block commit. This delay allows time for a tenure change block to arrive, reducing the chance of committing to the wrong tenure.
With a well-functioning network, if no tenure change block is seen within ~30s, it’s unlikely to be mined at all, so submitting the block commit at that point should be safe.
The text was updated successfully, but these errors were encountered:
Currently, the relayer submits a block commit as soon as it detects either:
1. A new burn block.
2. A new highest tenure.
However, committing immediately upon seeing a new burn block—before confirming whether a tenure change block follows—is risky. If a tenure change block arrives afterward, the miner’s commit will confirm the wrong tenure, making it impossible to mine a valid Stacks block.
This risk can sometimes be rewarded:
• The miner’s commit might be the only one in a flash block, making it more likely to win, though the ATC-C rules reduce this probability
• If the previous miner failed to produce a tenure change block before the next Bitcoin block, the commit could still be valid.
• Even if a tenure change block is mined, the commit could remain valid due to the bad timing reorg allowance.
However, in the common case, the miner must quickly replace-by-fee (RBF) the early commit with a new one referencing the correct tenure. If this RBF fails, the miner’s PoX payments are wasted.
Proposed Change
Modify the default behavior to be more patient by waiting a configurable delay (e.g., ~30s) after detecting a new burn block before submitting a block commit. This delay allows time for a tenure change block to arrive, reducing the chance of committing to the wrong tenure.
With a well-functioning network, if no tenure change block is seen within ~30s, it’s unlikely to be mined at all, so submitting the block commit at that point should be safe.
The text was updated successfully, but these errors were encountered: