0x3536a9f03536…3536a9f3
Core Lightning Needed Two Releases to Close Channel-Crash Flaw
Core Lightning shipped two releases, in July and September, to close a flaw that could crash Lightning nodes repeatedly; researcher Nishant Bansal disclosed it on Nov. 13.

Outputs
Flaw affected Core Lightning v0.12.0 through v26.06.7; disclosed Nov. 13 by researcher Nishant Bansal.
First attack route needed zero on-chain fees; patched in v26.06.6 on July 22.
Second route, via closed-channel records, caused persistent crash loops; patched in v26.06.8 on Sept. 22.
Offline nodes risk missing revoked commitments and time-sensitive claims across other channels.
Disclosure included no count of affected nodes or confirmed losses.
Core Lightning, one of the three major implementations of Bitcoin's Lightning Network, needed two separate software releases — on July 22 and September 22 — to fully close a flaw that could crash a node and keep it offline indefinitely. Researcher Nishant Bansal disclosed the details in an advisory dated November 13, following a public forum post on November 14.
The vulnerability affected Core Lightning versions v0.12.0 through v26.06.7. That range spans more than four years of releases, which suggests a large share of deployments could have carried the flaw at some point. The disclosure offered no count of affected nodes and no confirmed losses, a gap that deserves emphasis.
How did the flaw work?
The problem sat in how Core Lightning derives an identifier for a channel. The identifier comes from the channel's funding outpoint — the specific transaction output that anchors the channel on the Bitcoin blockchain. If an attacker reused one outpoint across two channels, both channels would receive the same identifier inside the node's records.
That duplication opened two distinct attack routes. The first required no on-chain transaction fees at all, meaning the cost to an attacker was effectively zero per attempt.
An attacker could create duplicate pending channels and crash the target node at the moment of reconnection. The node would restart, reconnect, and crash again. Core Lightning shipped v26.06.6 on July 22 at 12:16 p.m. ET (16:16 UTC) to close this path. The update blocked a peer from reusing an outpoint among its live channels and changed reconnect handling so duplicate records already stored on the node would not trigger the same failure.
What did the September release fix?
A second route survived the July patch. It involved an outpoint tied to a channel that had already closed. If a node held that stale record, closing another channel with the matching identifier could trigger a crash loop that persisted across restarts — with no further action needed from the attacker.
This route did require on-chain fees, since the attacker had to close channels on the Bitcoin base layer. Core Lightning released v26.06.8 on September 22 at 9:38 a.m. ET (13:38 UTC), expanding identifier checks to cover both live and closed channels and allowing existing duplicate closed-channel records to load and close cleanly.
In plain terms: the July fix guarded the front door, but a side entrance through archived channel records stayed open for two more months.
Why does node downtime matter for funds?
A crashed Lightning node is not merely an inconvenience. A node that remains offline can miss blockchain events that protect money held in its other, unrelated channels — for example, a revoked channel commitment from a misbehaving peer, or a time-sensitive payment claim that expires if unwatched.
The flaw therefore created indirect financial exposure even though it was a crash bug rather than a direct theft vector. The disclosure did not state whether any operator lost funds this way, and absence of reported losses is not evidence of safety.
Bansal's recommendation was unambiguous. "Users should upgrade to the latest release to protect their nodes," he wrote in the advisory.
Operators still running any version before v26.06.8 remain exposed to the crash-loop route through closed-channel records. With the advisory now public, the details are available to anyone, which compresses the window for laggard deployments to patch before the technique is tested in the wild.
via tokenpost.com (Source)
More from Elena Vasquez
Show full bio
Staff writer covering regulation, policy and their impact on crypto and web3 industry at Mempool Brief.
1,525 articles