On September 10, 2023, it was reported that Linea experienced a failure in the sequencer of its L2 network with a noticeable performance degradation.
Engineers released a quick patch in about 25 minutes, and the mainnet was fully operational again in less than an hour, under close monitoring to ensure stability.
For a technical overview on rollups and the role of the sequencer in L2s, see the official overview on Ethereum.org.
Essential Timeline
- 05:52 UTC — An anomaly was detected in the component responsible for sorting and grouping transactions.
- 06:15 UTC — fix distributed; gradual resumption of batching pace initiated.
- 06:32 UTC — activated enhanced monitoring on telemetry and logging.
- Status — incident closed; monitoring ongoing to confirm stability.
Impact on Users
The partial halt of the sequencer caused slowdowns in block inclusions and temporary transaction queues. There are no effects on L1 nor any loss of funds.
- Funds are safe and L1 consensus unchanged.
- Delays in confirmations and increase in backlog during the peak.
- Normal operations restored after the application of the patch.
Technical Details Known So Far
Linea has not disclosed the specific cause. The fix involved the sorting/batching module of the sequencer, without any interventions on the level 1 chain. After the deployment, the team intensified telemetry to verify stability and absence of regressions.
- Control of throughput (batching and inclusion rate).
- Measurement of transaction insertion latency.
- Verification of the error rate and system logs.
Why it matters: the single sequencer node
The episode reignites the debate on the single sequencer as a single point of failure in L2s. Many rollups are exploring models with distributed sequencers, shared sequencing, or failover mechanisms to improve resilience and operational continuity.
Uptime remains crucial for dApps and operators that depend on fast and predictable confirmations. To delve deeper into the topic of decentralization in Ethereum’s layer 2, see Layer-2 of Ethereum: Vitalik Buterin calls for more decentralization.
Announced Actions and Prevention
- Periodic reviews of the sequencer code.
- Faster rollback procedures in case of regression.
- Test automation to detect performance degradations.
- Proactive alerts and continuous telemetry monitoring.
What to Watch in the Coming Hours
- Latency of median inclusion and 95th percentile.
- Throughput (tps) and batch publication rate.
- Network error rate and any residual time-outs.
- Disposal of any tails remaining from the peak.
Data and Quotes
According to the data collected by our editorial team during the monitoring of public telemetry feeds and aggregated logs, the median inclusion latency has increased significantly compared to pre-incident values, with the 95th percentile exceeding 10 seconds in some public snapshots.
Industry analysts note that, in similar cases, the transaction backlog can grow from a few tens of percent to over 100% during peaks, depending on the sequencer configuration and load.
The editorial team contacted Linea for additional clarifications; no response was received at the time of publication. Update: this article was updated on September 10, 2025, to contextualize the case two years after the original event.
Source: https://en.cryptonomist.ch/2025/09/10/linea-l2-sequencer-in-tilt-at-dawn-patch-in-25-minutes-network-stable-in-less-than-an-hour/
