Devnet Tracking Issue

[eyusufatik]

We are very close to making our testnet targets.

Following are what features and enhancements to be done urgently:

• Proving soft confirmation batches. And prover node related implementations. (will add details)
  • Provable citrea-stf
  • Connnecting our prover process with risc0 (ParallelProverService, StateTransitionRuner)
  • Diff format. Probably will just use working_set diff.
  • Sending proof and state diffs to DA layer.
  • Tests
• DevOps -- @Chinacolt is working on this, plan is ready.
• L1 fee rate related tasks
  • Estimations so that txs don't fail Add l1 fee to gas estimation #417 and Diff size estimation #420
  • Sequencer can't know a tx fails by not enough balance for L1 fees Handle not enough L1 fee transactions #409 -- maybe we can modify mempool to account for L1 fees, but this will be a little resource heavy. Or Remove failed transactions from mempool as well #477 ensuring we run them only once. (enough for devnet)
  • Deviding L1 fees between transactions. EVM: Figure out a way to divide L1 fees between transactions #267 -- This can be a hard fork post testnet launch.
• Full node to sequencer transition -- in case of failure a full node we run can take over as sequencer, but it should know any pending sequencer commitments, everything else should work without block loss. Offchain sequencer commitment records #443
• Activating the bridge Activate Deposit transactions #403
  • Write tests for deposit transactions
  • E2E test for deposit transactions (might not work with mock-da we'll see)
• Ethereum foundation tests. @rakanalh is working on this but there are a lot of cases reth doesn't run, we'll need to implement those cases on our own.
• Writing a seperate pool for deposit transactions in mempool.
• Handling multiple commitments per DA block Add support for multiple commitments in a single block #189 -- this is really easy.
• Making blockscout ready for deployment. Things work right now, should set genesis and set branding.
• Support reorgs in full nodes. This shouldn't happen, but we must implement this nevertheless. Data Availability re-orgs handling tracking issue Sovereign-Labs/sovereign-sdk#1023
• Sequencer main loop enhancements Sequencer Main Loop Enhancement #133 -- Nothing fancy, we should just make our sequencer not miss the 2 second block time.
  • This also means we should be more careful with what we push to our state transition function. Sequencer: Account gas usage of system transactions #412
• Better error handling all around -- we are ok imo, but always can do better :)
• Bridge frontend @selinisik 's working on this but to accelerate things we can help.
• Bridge backend @ekrembal and @gwn making really fast progress on this but we should sync regularly esp. for the seperate deposit pool.
• Mempool Dump Store mempool transactions in shared backup db #478

[okkothejawa]

Sorry, misclicked.