Santiago Palladino
Santiago Palladino
Cloudwatch is smart enough to parse JSON logs, and we can use the different fields extracted for querying in Insights. **Builds on #8095**
See the message from the first commit for context. I've managed to get past the`Proof length doesn't match expected length (16 != 409)` error [here](https://github.com/AztecProtocol/aztec-packages/actions/runs/10375644353/job/28727687668?pr=7952#step:5:543) by making some changes blindly...
We want to split up sequencers and provers, as they are two different entities in the system. With the changes to block building, sequencers no longer need to upload a...
In https://github.com/AztecProtocol/aztec-packages/pull/7844 we added an epochNumber for sorting jobs. However, this makes it difficult to implement with a fifo queue with no support for priorities. Instead, we could use a...
Tweaks the archiver so it can detect when a prune _will_ happen and can start unwinding blocks then. This is then leveraged by the sequencer, so it builds the next...
We only build tsc on build, and use the generate step for moving build artifacts and generate code.
In complex executions it's difficult for the app to foresee the exact authwit requests that need to be added to the pxe before simulating execution. It's particularly difficult if the...
A call to `getL2ToL1MessageMembershipWitness` causes the node to load the entire L2 block, create N trees backed by ephemeral cpp stores (where N is the number of txs in the...
In addition to what we return today, include: - Index of the log in the tx - Index of the tx in the block - Unique note hashes from the...
Today PXE usually calls `findLeafIndex` followed by `getSiblingPath` given the result. We should add a method to batch both together and eliminate a roundtrip. Consider `getSiblingPathByValue(blockNumber, treeId, value)`.