Fast Confirmation Rule (FCR) #9, June 23, 2026

Agenda

  • Client Updates
  • Spec Updates
  • Testing Updates
  • API and Metrics Updates
  • Research Updates

Meeting Time: Tuesday, June 23, 2026 at 12:00 UTC (60 minutes)

GitHub Issue

Meeting Summary:

The team held a client update meeting focused on fast confirmation (FCR) implementation progress across different clients. Nazar reported successful deployment of past information to test and RC fleets, with plans to include metrics in the upcoming release, though the team decided not to delay the release for metrics inclusion. Boma discussed completed metrics implementation awaiting maintainer reviews, particularly from Samas and others who have competing priorities with EPBS work. Mikhail provided updates on Lighthouse regression investigations involving reversions to finality and mentioned ongoing work on edge case tests. The team discussed API and metrics updates, including a recently merged PR by Nazar that adds current slots to the event stream, and Roberto shared that his research paper was released this week after a four-week pause due to other projects. The conversation ended with discussions about monitoring different client implementations for edge cases and questions about the usage of SAFe blocks in APIs, which would require significant coordination with other teams before Glamsterdam.

Click to expand detailed summary

Nazar reported that their past information deployment in test fleet and RC is working well and will be included in the upcoming public release with a related blog post. When asked about including metrics in the release, Nazar indicated the metrics would be included in the next release rather than rushing them into the current one, as Mikhail confirmed was acceptable. The team discussed that users could alternatively build from the master branch if they needed the metrics immediately.

The team discussed node behavior observations, with Nico reporting delayed first block confirmation but eventual catch-up after two epochs. Boma updated on metrics implementation changes awaiting review from maintainers, particularly Samas, who has priority on other EPBS work. Mikhail mentioned investigating a regression in Lighthouse causing reversion to finality issues and noted that spec tests for gloss and haze are passing since alpha.11. The team also discussed recent PRs for beacon-APIs and beacon-metrics, with one metrics PR already merged and another awaiting merge approval.

The team discussed API metrics and research updates, with Roberto reporting that a paper was released this week after focusing on functionality and other projects for the past four weeks. Nico identified that the current reset metric tracks 3 total resets over the last 7 days, though there may be issues with how the metric is calculated, particularly regarding initial node startups. The team also discussed investigating edge cases with different FCR client implementations, with Mikhail suggesting they would need to consult with PandaOps about running comparison tests between clients.

The team discussed the decision to use “Safe” tags for the EL side, but Mikhail confirmed they had not yet reached out to RPC providers to verify if anyone relies on the current Safe implementation. Roberto and Mikhail identified that implementing a new tag would require significant coordination across multiple components including CL and DL, and would likely need to wait until after the Glamsterdam work is completed. Mikhail agreed to investigate the current usage of Safe block before making any final decisions.

Next Steps:

  • Nazar: Integrate the cost information metrics into the charts during this week.
  • Nazar: Investigate the observed behavior of the first confirmed block delay and the resets, and provide more debug information.
  • Mikhail: Reach out to PandaOps (possibly Perry Barnabas) to get the metrics PR merged.
  • Mikhail: Talk to PandaOps about running a comparison test between different client implementations of FCR.
  • Mikhail: Investigate the usage of the “safe” block tag by reaching out to infrastructure and RPC providers.
  • Boma: Seek help in getting the maintainers (Samas or another person) to review and merge the metrics PR for Grendine.
  • Mikhail: Open a PR for edge case tests related to fast confirmation, possibly by tomorrow.

Recording Access:

YouTube recording available: https://youtu.be/GxroFGw6cck