Part of the team is on holidays this week. The rest of us focused on:
-
Continue with the process of extracting changes from our integration branch into PRs targeting
main. This week we merged:- Peras 43: Move Test.Consensus.Committee.Utils to Test.Util.Committee
- Peras 44: Introduce Peras epoch context resolution machinery
- Peras 45: Introduce Peras type family wrappers for the HFC
- Peras 46: Introduce Void Peras types and helpers
- Peras 47: Introduce OneEraPeras type wrappers
- Peras 48: Do not hardcode PerasEnabled/NoPerasEnabled in era parameters
- Peras 49: Use injective type families for associated BlockSupportsPeras types
- Peras 50: Use SeatIndex-based PerasVoteId
-
Investigate how Peras should interact with the existing Consensus tools used for chain analysis and disaster recovery. While in principle these can mostly work without knowing anything about Peras, we have identified a handful of minor adjustments that some of these tools will need before the mainnet release.
-
Start working on the logic relating certificate size to blocks:
- If a cert happens to be larger than the maximum block size, reject it (because it may not be included in a block), and
- When forging a block that must contain a certificate, reserve the space needed for it before filling the rest of the block body with transactions.
-
Continue iterating on the Peras testnet. This week we successfully started a non-Peras testnet and made it transition from Conway into Dijkstra via governance actions. From there, we are currently porting these changes back to our Peras-enabled testnet setup.