Skip to main content

· One min read
James Chapman

The team works on applied research and consulting in formal methods that is directly applicable to evidence based engineering in Core Tech and beyond.

High level summary​

The team is working on formalising mini protocols, the performance modelling prototype and also finishing off their ICE papers.

Details​

  • Formalization of the chain synchronization mini-protocol in the thorn calculus

  • Final pre-publication steps for ICE 2023 papers

  • Developing approach for specification and verification of mini-protocols

  • Started work on porting the DeltaQ framework to a new, more concrete backend based on piecewise-polynomials

· 2 min read
Carlos LopezDeLara

2023-07-05 - 2023-07-18

High level summary​

  • Integrate Conway certificates in cardano-api
  • Prepare cardano-cli and cardano-api for cardano-node 8.1.2
  • Tagged cardano-cli 8.3.0.0
    • Query for the hash of the constitution (Conway)
    • Ability to create votes and governance actions
  • Tagged cardano-cli 8.3.1.0
    • Make it build with ghc-9.6
  • Upgrade to optparse-applicative-fork-0.18.1.0 for parsing command line options

cardano-cli​

cardano-api​

cardano-node​

cardano-testnet​

docs​

CI & project maintenance​

· One min read
Franco Testagrossa
Sasha Bogicevic

High-level summary​

This week the team focused on exploring the event sourced persistence in order to improve hydra-node performance. Because of this work the team noticed we need to refactor the emit snapshot emission logic and update the spec in the light of new changes. They also took the time to revisit their goals and product plans for the next quarter as well as doing some security fixes related to multisignatures.

What did the team achieve this week​

  • Finished spike about performance improvements of event sourced persistence #963.
  • Refactor snapshot emission in protocol logic.
  • Revisited our roadmap and goals.
  • Prepared and conducated a learning session on lean-waste.
  • Improve security of multi-signature checks, see this Github security advisory.
  • Implemented a cache friendly way to version our binaries #962.

What are the goals of next week​

  • Implement Event sourced persistence #913.
  • Remove deprecated internal commit #954 and close #728.

· 2 min read
Jean-Philippe Raynaud

High level overview​

The Mithril team created a new 2327.0 distribution. They focused on preparing the beta launch on the mainnet: they tested the new production signer deployment model with the pioneer SPOs, they prepared an SPO on-boarding guide, and they kept working on the deployment and monitoring of the mainnet infrastructure. The team also worked on the implementation of a simple stress test tool for benchmarking the aggregator. Additionally, they completed the refactoring of the interface to the cryptographic library.

Finally, they fixed a bug that sporadically prevented the latest signer registration of a SPO to be used in the associated signing epoch, they fixed a bug in the epoch gap detection of the certificate chain in the aggregator, and worked on multiple other optimizations and bugs.

Low level overview​

  • Released the new distribution 2327.0
  • Worked on the epic that prepares the Mithril infrastructure for mainnet #767:
    • Completed the issue Add infrastructure monitoring #987
    • Worked on the issue Deploy 'mainnet' infrastructure #988
    • Worked on the issue Handle Secrets management #989
  • Worked on the epic Benchmark performances of Mithril Aggregator #904:
    • Worked on the issue Design & implement basic stress test tool for aggregator #991
  • Worked on optimizations:
    • Completed the issue Remove certificate hash from Artifact #932
    • Completed the issue Check vulnerabilities in CI #1037
    • Completed the issue Add 'created_at' in Mithril Stake Distribution messages #1030
    • Completed the issue Add a 'run-only' option in end to end test #1048
  • Worked on refactoring:
    • Completed the issue Factorize protocol crypto operations #669
    • Completed the issue Refactor aggregator dependency injection and services #1058
    • Completed the issue Build static binaries in CI #874
  • Worked on documentation:
    • Completed the issue Prepare SPO on-boarding guide #1049
    • Completed the issue Add instructions to set firewall using iptables #1040
    • Completed the issue Update ufw command to set firewall on Mithril Signer installation instructions #1041
  • Worked on bugs:
    • Completed the issue Aggregator does not detect certificate chain epoch gap #952
    • Completed the issue 'testing-preview' network does not create certificates #1015
    • Completed the issue SQLite compatibility in aggregator #837
    • Completed the issue Q&A followup fixes #1035
    • Completed the issue E2E tests are flaky in CI #1023

· One min read
Damian Nadales

High level summary​

This week the team working on UTxO-HD discovered a space leak in the peer metrics code. This was communicated to the Networking team who has a proposed fix. The ad-hoc benchmarks that the team ran using a local immutable DB server showed good memory and time performance. We still have to check the performance on a memory constrained machine.

The team working on the Genesis design started onboarding the team of engineers that will implement the new Genesis protocol. This team is also finalizing the statistical model for historical Genesis feasibility.

On the support front, the team drafted an information exchange requirement (IER) for the Networking team to safely and efficiently control peer load.