feat: modify txmgr to send EIP-1559 txns#2086
feat: modify txmgr to send EIP-1559 txns#2086mslipper merged 6 commits intoethereum-optimism:developfrom
Conversation
🦋 Changeset detectedLatest commit: 3a7e709 The changes in this PR will be included in the next version bump. This PR includes changesets to release 1 package
Not sure what this means? Click here to learn what changesets are. Click here if you're a maintainer who wants to add another changeset to this PR |
6c860d7 to
e7d25e2
Compare
Codecov Report
@@ Coverage Diff @@
## develop #2086 +/- ##
========================================
Coverage 75.37% 75.37%
========================================
Files 81 81
Lines 2705 2705
Branches 436 436
========================================
Hits 2039 2039
Misses 666 666
Flags with carried forward coverage won't be shown. Click here to find out more. Continue to review full report at Codecov.
|
e7d25e2 to
c812161
Compare
c999a67 to
52c5bd7
Compare
There was a problem hiding this comment.
@tynes is the best we can do in terms of a fallback? i.e. using a constant, or is there a better way you can think of?
This commit adds a fallback in case the L1 provider does not support queries for eth_maxPriorityFeePerGas. Of our infrastructure providers, i.e. Alchemy, Infura, Quicknode, and Hardhat (in test envs), only Alchemy suports this call. In production, we will be routing this call directly to Alchemy, which should work in the general case. However, if they were to be unavailable the batch submitter would be stalled until service was restored. This modification also allows the new EIP-1559 batch submitter to pass our integration tests, since the default Hardhat backend can't support the query either.
Removes the MAX_L1_GAS_PRICE_IN_GWEI and GAS_RETRY_INCREMENT env vars, as well as the MinL1GasPrice, MaxL1GasPrice, and GasRetryIncrement values from the txmgr config struct.
6169fed to
5cdc950
Compare
…`derivationOriginUpdate` (#2331) Closes #2086 The PR: - Introduces new table `BlockTraversal` to store all L1 blocks, even if there are no L2 blocks derived from it. - Add respective trait methods. - Modified `latest_derived_block_pair` to send the latest derived block from `DerivedBlocks` and latest source block from `BlockTraversal`. - Exclusively dealing with `derivationUpdate` and `derivationOriginUpdate` i.e. updating latest derived on `derivationUpdate` and updating latest source on `derivationOriginUpdate`. - The e2e test `TestFinalizedHeadAdvancing` passes now. TODO: (in another PR) - Remove `current_l1` and its usage. No longer required. - ~~For now, `check_node_consistency()` is commented in this PR. #2330 takes care of that.~~ ~~Currently in draft: Tests are missing.~~ --------- Co-authored-by: Arun Dhyani <dhyaniarun7@gmail.com>
…`derivationOriginUpdate` (op-rs/kona#2331) Closes #2086 The PR: - Introduces new table `BlockTraversal` to store all L1 blocks, even if there are no L2 blocks derived from it. - Add respective trait methods. - Modified `latest_derived_block_pair` to send the latest derived block from `DerivedBlocks` and latest source block from `BlockTraversal`. - Exclusively dealing with `derivationUpdate` and `derivationOriginUpdate` i.e. updating latest derived on `derivationUpdate` and updating latest source on `derivationOriginUpdate`. - The e2e test `TestFinalizedHeadAdvancing` passes now. TODO: (in another PR) - Remove `current_l1` and its usage. No longer required. - ~~For now, `check_node_consistency()` is commented in this PR. #2330 takes care of that.~~ ~~Currently in draft: Tests are missing.~~ --------- Co-authored-by: Arun Dhyani <dhyaniarun7@gmail.com>
Description
Modifies the
txmgrto use EIP-1559 dynamic fee txns, rather the currentmethod of linearly increasing the gas price between a configured minimum
and maximum.
In addition, we add generic fallback in case the L1 provider does not support
queries for
eth_maxPriorityFeePerGas. Of our infrastructure providers,i.e. Alchemy, Infura, Quicknode, and Hardhat (in test envs), only
Alchemy supports this call. In production, we will be routing this call
directly to Alchemy, which should work in the general case. However, if
they were to be unavailable the batch submitter would be stalled until
service was restored. This modification also allows the new EIP-1559
batch submitter to pass our integration tests, since the default Hardhat
backend doesn't support the query either.
Compare to the existing, approach, we should see quicker confirmation times,
as the prior iteration would start ratcheting from well below the market rate. This
should have small, net positive effect on the overall TPS of the system.
Metadata