Starknet pins brief mainnet outage on state discrepancy between execution and proving layers

The team behind Starknet, an Ethereum layer‑2 scaling network, released a post‑mortem report on Monday outlining the root cause of a temporary mainnet downtime.

The report said the outage stemmed from a discrepancy in network state between the blockifier execution layer and the proving layer that verifies execution. "In one specific combination of cross‑function calls, variable writes, reverts, and catching them, the blockifier remembered a state‑writing that happened within a function that was reverted, causing an incorrect transaction execution," the team wrote. The proving layer flagged the error and the incorrect execution never saw L1 finality, the report added.

The incident forced a block reorganization that rolled back 18 minutes of network activity. The Starknet team said mainnet functionality is back to normal and that it will pursue testing and code audits to help prevent similar issues. The report noted the episode highlights the challenge of coding for multi‑layered blockchain technology stacks.

The outage was the second major Starknet disruption in 2025 that required a block reorganization. In September, following the Grinta protocol upgrade, a sequencer bug caused an outage lasting over five hours, prompted two chain reorganizations and forced about one hour of network activity to be reverted, according to the earlier post‑mortem cited in the report.

Key Topics

Crypto, Starknet, Ethereum, Blockifier, Proving Layer, Grinta