[Active] AE Maintenance (weeks 21-22, 2024)

Application Status

Status: Active

Hello Community Members,

We hope you’re all doing great. We’re thrilled to share the latest updates and enhancements from the past two weeks. Let’s get started!

Weeks 21-22 (20 May - 31 May)

The past two weeks have been packed with Hyperchain discussions and both short-term and long-term planning meetings.

Hans Svensson

The current default configuration is very defensive, allowing the node to switch to a longer/heavier fork of up to 100 generations. This is now changed to 20 generations in a first step (of several) to improve chain finality.

We spent some time on HyperChains meetings, planning, creating short-term plans, and also adding topics to the agenda for an upcoming dev meetup.

Finally got around to “productify” the work done on safe access to DB data. Added two features; (1) a configurable safe-access DB-mode, where every read of a value from a Merkle Patricia tree (i.e. all state tree values) is checked (hash of binary data matching its key), and (2) a one-shot (background) check of the whole trees.

This is useful when a node is started from an untrusted DB snapshot.

Justin Mitchell

Adding a WebSocket API to publish blockchain events that are required by state channels, will allow state channels to be moved out of the node.

Some more details you can find here.

Thomas Arts

We cleaned up and removed old swagger dependencies by PR4354, PR4355.

Valentin Atanasov

A general dex swaps route has been added and for aex9 we returning now the correct number of holders. Check the details here.

The aexn open api schemas has been fixed as well.

Dincho Todorov

We optimized, reorganized, and cleaned up many things:

  • Optimised (configuration tuning) of the applications run in the applications cluster to save on resources.
  • Optimised database restore playbooks and snapshots to increase bootstrap time
  • Reorganised the bootstrap process to have a more robust order with monitoring tools
  • Node maintenance/fixes
  • Alerts tuning
  • Increased the pod limits per node of the application cluster to increase utilization and reduce costs.

We can also share with you the next release and deployment of 7.1.0!

Moreover, we started working on an alternative to the Datadog logging solution using Grafana Loki as a basis.

Mihail Dobrev

Docs for aex141 and names in the middleware have been updated. Debug syncing issues with the /wealth endpoint in the middleware mainnet and set up a local middleware for the mainnet from a backup to test the /wealth endpoint.

Write a migration that updates all account balances in the middleware with the values from the node, to sync the data for the /wealth endpoint.

The dex_swaps migration has been fixed.

We made the skipped migrations run again and the wealth task in the middleware worked with the database instead of AsyncStore and ets.

Richard Carlsson

I participated in Hyperchains meetings and conducted some research on the topic. Additionally, I took part in planning discussions and studied the Aeternity architecture.

Radosław Rowicki

Some news from Sophia compiler. Finalized tree-sitter and made it parse all valid Sophia files, prepared tree-sitter-aesophia release and we started working on CST->AST conversion in the Rust compiler.

Your input and feedback are essential. Let’s keep building this together!