On Friday and Saturday, a series of events unfolded as a denial-of-service attack struck Litecoin, leveraging a vulnerability within its Mimblewimble Extension Block protocol to push through invalid transactions. The attack ultimately led to a 13-block chain reorganization, effectively rewinding approximately 32 minutes of network activity.
This reorganization occurred after attackers exploited a bug, allowing them to bypass nodes that had not been updated, before the network's longest valid chain intervened to correct the issue. In response, Litecoin Core version 0.21.5.4 was released, containing crucial security updates and advising all users to upgrade. According to the Litecoin Foundation, the bug was fully patched by Sunday morning, and the network has resumed normal operations.
However, security researchers have pointed out discrepancies in the timeline of events, as indicated by the litecoin-project GitHub repository. The repository shows that the consensus vulnerability was privately patched between March 19 and 26, more than four weeks before the attack occurred.
Furthermore, a separate denial-of-service vulnerability was addressed on the morning of April 25, with both fixes being incorporated into the release 0.21.5.4 on the same afternoon, after the attack had commenced. The term 'zero-day' refers to a vulnerability that is unknown to defenders at the time of an attack. In this case, Litecoin's commit history reveals that the consensus vulnerability was known and patched privately a month prior to the exploit, but the fix had not been publicly disclosed or mandated for all mining pools. This created a window of opportunity where some miners were running the patched code, while others were still running the vulnerable version, and the attackers seemed to be aware of this distinction.
Alex Shevchenko, CTO of the NEAR Foundation's Aurora project, has expressed similar concerns. Blockchain data indicates that the attacker had pre-funded a wallet 38 hours before the exploit, using a Binance withdrawal, with the destination address already set up to swap LTC for ETH on a decentralized exchange. Shevchenko argues that the denial-of-service attack and the MWEB bug were separate components, with the DoS designed to take patched mining nodes offline, allowing the unpatched ones to form a chain that included the invalid transactions.
The fact that the network automatically handled the 13-block reorganization once the DoS stopped suggests that enough hashrate was running updated code to eventually overpower the attack, but only after the unpatched fork had run for 32 minutes. This incident highlights the differences in how attacks on various networks are handled, depending on the reaction of code maintainers and developers to exploits. Newer chains with smaller, more centralized validator sets can coordinate upgrades through chat groups and push patches network-wide in a matter of hours. In contrast, older proof-of-work networks like Litecoin and Bitcoin rely on independent mining pools choosing when to upgrade, which works for non-urgent changes but creates a window of vulnerability when a security patch needs to reach everyone before an attacker can exploit the gap.
As of Sunday morning, the Litecoin Foundation has not publicly addressed the GitHub timeline. The amount of LTC pegged out during the invalid block window and the value of any swaps completed before the reorganization reversed them have not been disclosed.