feat: added dropping all attester certificates when doing hard fork #2529
Conversation
There was a problem hiding this comment.
When I first saw that miniblocks_consensus is deleted, I thought that's okay because L1 doesn't care about L2 blocks, but I thought the QCs of the existing L1 batches can be kept around.
I see that a specific number of miniblocks are kept by the last_l2_block_to_keep number of miniblocks are retained, so that the API can keep serving transaction data to users. The corresponding L1 batches have already been submitted, and they are the ones that aren't rolled back, so the QC over the constituent L2 blocks are not relevant any more and the validator set could have changed as well during regenesis.
On the same account we can remove the L1 batch QCs for those past L1 batches since they are not going to be re-submitted, and because attesters might change in regenesis at the moment. In the future when attesters will be based on a smart contract, the historical quorums will stay amenable to validation by looking up the corresponding attesters, at which point we won't have the same argument for getting rid of these records as we do now.
|
@aakoshh my guess would be that even once the signature verification happens on L1, the signature itself won't be persisted in the contract state, so the qc will be available only in the calldata of the corresponding transactions. |
🤖 I have created a release *beep* *boop* --- ## [24.13.0](core-v24.12.0...core-v24.13.0) (2024-07-31) ### Features * Add recovery tests to zk_supervisor ([#2444](#2444)) ([0c0d10a](0c0d10a)) * Added a JSON RPC to simulating L1 for consensus attestation ([#2480](#2480)) ([c6b3adf](c6b3adf)) * added dropping all attester certificates when doing hard fork ([#2529](#2529)) ([5acd686](5acd686)) * **configs:** Do not panic if config is only partially filled ([#2545](#2545)) ([db13fe3](db13fe3)) * **eth-sender:** Make eth-sender tests use blob txs + refactor of eth-sender tests ([#2316](#2316)) ([c8c8334](c8c8334)) * Introduce more tracing instrumentation ([#2523](#2523)) ([79d407a](79d407a)) * Remove unused VKs, add docs for BWG ([#2468](#2468)) ([2fa6bf0](2fa6bf0)) * Revisit base config values ([#2532](#2532)) ([3fac8ac](3fac8ac)) * Server 10k gwei limit on gas price and 1M limit on pubdata price ([#2460](#2460)) ([be238cc](be238cc)) * **vlog:** Implement otlp guard with force flush on drop ([#2536](#2536)) ([c9f76e5](c9f76e5)) * **vlog:** New vlog interface + opentelemtry improvements ([#2472](#2472)) ([c0815cd](c0815cd)) * **zk_toolbox:** add test upgrade subcommand to zk_toolbox ([#2515](#2515)) ([1a12f5f](1a12f5f)) * **zk_toolbox:** use configs from the main repo ([#2470](#2470)) ([4222d13](4222d13)) ### Bug Fixes * **contract verifier:** Fix config values ([#2510](#2510)) ([3729468](3729468)) * fixed panic propagation ([#2525](#2525)) ([e0fc58b](e0fc58b)) * **proof_data_handler:** Unlock jobs on transient errors ([#2486](#2486)) ([7c336b1](7c336b1)) * **prover:** Parallelize circuit metadata uploading for BWG ([#2520](#2520)) ([f49720f](f49720f)) * VM performance diff: don't show 0 as N/A ([#2276](#2276)) ([2fa2249](2fa2249)) --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please). --------- Co-authored-by: zksync-era-bot <[email protected]>
Clean state at every fork is closer to what will happen during hard fork on L1.