No. The VOLUNTARY signaling period failed to reach lock-in; the proposal itself did not get cancelled. The mandatory window (blocks 961,632 to 963,647) is a separate mechanism with its own count, and it's already open. Activation itself, if it ever came, would fall later still, at block 965,664.

Could 2.53% miner support trigger a fork?

Not a hard fork: BIP-110 is a soft fork, so a block valid under its rules is also valid under unmodified Bitcoin Core. What could happen, and did, is the minority running the activation client isolating itself by rejecting an ordinary non-signaling block: that happened on August 8 at block 961,632, and the resulting chain has been stuck at 2 blocks ever since. Full update below.

Does this have anything to do with the Bitcoin Core vs. Knots debate?

No, they're unrelated. Core and Knots share the exact same consensus rules and differ only in which unconfirmed transactions they relay by default, zero fork risk there. BIP-110 does introduce a new (temporary) consensus rule, and its activation client happens to be a Knots fork, but that doesn't make Knots itself part of this risk.

On August 8, at block 961,632, the minority running the activation client rejected the first non-signaling block and split off from the main network. That minority chain has been stuck at just 2 blocks ever since. Full story, including what matters if you self-custody, in a dedicated follow-up.

A golden 3D Bitcoin coin at the center, surrounded by a mostly-dim segmented ring in blue-gray, with only a small stretch lit in amber near the ring's closing point, over a dark candlestick-chart background — Original illustration generated with AI (Adobe Firefly) — not representative of real data.

BIP-110's Voluntary Window Closed Without Lock-In: What Changes, What Doesn't

NodeWitness Team August 8, 2026 4 min read

BIP-110's voluntary signaling period closed on August 8, 2026 with 2.53% miner support, far short of the 55% required. This isn't the end of the story — it's the end of one specific chapter, and here's the difference.

Update, August 13, 2026 — the isolation risk we called "close to nil" happened anyway. That same block 961,632 was rejected by the minority running the activation client, and that chain genuinely split off. We had the mechanism right (soft fork, no hard-fork path, the minority isolates itself, not the other way around) and underweighted that "close to nil" isn't zero: it only took one large miner staying on the activation client. The good news is that chain has been stuck at 2 blocks ever since, effectively dead. Full story, including the part that actually matters if you self-custody, in a dedicated new piece.

If you've been following BIP-110 with us, you already know the individual pieces: the delayed activation, the arithmetic that made this outcome certain as of August 1, the operator playbook for anyone running their own node. What was missing was pulling them together now that the period has actually closed, and answering the question that matters: what now? If you need the full mechanics first, our node operator playbook covers what BIP-110 does and what it means for your keys.

The final number, in one line

What this does NOT mean

This is where coverage tends to flatten the story, so it's worth being precise: BIP-110 is not cancelled. What failed is one specific signaling period, not the proposal itself. The mandatory window (blocks 961,632 to 963,647) is a distinct mechanism with its own count, and it's already open. Activation itself, should it ever happen, sits further out still, at block 965,664.

Three separate clocks are running in sequence, and conflating them produces wrong headlines in both directions: neither "BIP-110 is dead" (it isn't, formally) nor "BIP-110 is proceeding as planned" (not with this level of support, and not in the way its proponents were hoping for).

Is there fork risk here?

This is the question we've gotten the most this week, and the short answer is: not the kind most people picture.

BIP-110 is a soft fork by construction — it tightens rules, it doesn't loosen them. A block that satisfies its restrictions is also valid under unmodified Bitcoin Core, so the overwhelming majority of the network (anyone staying on plain Core) never rejects anything over this. There's no scenario where BIP-110 produces a hard fork; that would require a block valid under the new rules to be invalid under the old ones, and here it's exactly the reverse.

The real risk was different: the small number of nodes running the activation client (a fork of Bitcoin Knots, not of Bitcoin Core) would start rejecting perfectly valid, non-signaling blocks from block 961,632 onward. It would be them, the enforcing minority, isolating themselves onto their own chain, not the other way around. With 2.53% miner support and an explicit instruction from the proposal's own backers to revert to a non-BIP-110 node before that height, that risk was already close to nil before the block was even mined.

And this has nothing to do with Core vs. Knots

Worth clarifying because the two get mentioned in the same breath a lot: Bitcoin Knots' 2025 growth (from roughly 2% to over 20% of public nodes) has no consensus-level relationship to any of this. Core and Knots validate the exact same blocks as valid — they differ only in which unconfirmed transactions each relays by default: a local preference with zero chain-split risk of its own. That BIP-110's activation client happens to be technically a fork of Knots is an infrastructure coincidence, not a sign that Knots itself is part of this risk.

What's next

That's what we said on August 8, and the material change we set as our bar for writing again arrived within hours: the activation-client minority rejected the first non-signaling block and isolated itself, exactly as described above as a real possibility. The full account — who was behind it, why the resulting chain is already effectively dead, and above all what it means if you hold your own keys — is in BIP-110's Fork Was Real: What Happened, and the Replay-Attack Risk You Should Know About.

Would rather check this yourself than take our word for it? Network and mining publishes our own node's live status, with the current block height so you can cross-check it against any explorer.

Last updated: August 13, 2026