❗ Sepolia Superchain Granite activation Next Monday 8/12
This release contains the activation of the Granite hardfork on the Sepolia Superchain on Mon Aug 12 16:00:00 UTC 2024. Node operators of any Sepolia OP Chains (OP, Base, Mode, Zora, and other chains in the Superchain Registry) must update before this activation timestamp, or their nodes will halt when the chains activate Granite.
It is a recommended update for all other node operators as it contains various improvements and bug fixes.
🤝 The corresponding Granite Sepolia op-geth release is v1.101315.3.
Granite
This release contains all features of the upcoming Granite protocol upgrade (specs pending). The Granite activation can be manually enabled via overrides for chains which don't activate Granite via the Superchain (currently only the Sepolia Superchain).
Batcher
Batcher Fixes
- The batcher correctly applies the new
maxRLPBytesPerChannellimit post-Fjord (#11169) - The batcher detects transactions of incompatible type that are blocking the mempool, and sends appropriate cancellation transactions (#10941)
Dynamic calldata/blob switching (#11219)
The batcher can be instructed to automatically switch between calldata and blobs, depending on the most economic choice at the time of starting a new channel. This feature can be activated by setting the --data-availability-type flag to auto (or the respective env var).
⚠️ This feature currently only works well for high-throughput chains that fill multi-blob transactions every few minutes. For low-throughput chains with channel durations of up to multiple hours, due to the high latency between choosing the channel configuration and submitting the channel, the current implementation doesn't work well. This will be improved in a future release.
Update of op-geth dependency
The op-geth dependency is updated to v1.101315.3.
Partial Changelog
- op-node: read BlockTime, SeqWindowSize from superchain-registry configs by @bitwiseguy in #11172
- have batcher submit appropriate cancellation transactions when mempool is blocked by @roberto-bayardo in #10941
- fix(op-batcher): support new fjord maxRLPBytesPerChannelFjord via rollup chain spec by @emilianobonassi in #11169
- op-node: batch-decoder: Fjord Frame Timestamp fix by @pcw109550 in #11143
- deps: delete unused eslint deps by @tynes in #11170
- opservice: rpc server into a cliapp service by @hamdiallam in #11197
- Add buffered caching of L1 block refs to the confDepth-aware fetcher by @BrianBland in #11142
- op-conductor: Adds connection test to cluster membership mutation apis by @zhwrd in #11210
- op-node: set genesis l1 start block to value from SystemConfig contract by @bitwiseguy in #11181
- op-node: Extract contract binding for SystemConfig by @ajsutton in #11227
- op-challenger: avoid passing FDG-factory address by @coderHarrii in #11236
- op-service: use binary search instead of walkback for checkRecentTxs by @bitwiseguy in #11232
- Delete chain-mon in favor of monitorism by @zchn in #11239
- op-node: genesis generation by @tynes in #11249
- [op-dispute-mon] Track the latest proposed L2 block number by @cbfyi in #11238
- ops-bedrock: Beacon-chain devnet with Dencun + Ecotone upgrade by @sebastianst in #9117
- feat(opsm): scaffold OP Stack Manager and tests by @mds1 in #11274
- cannon: Fix onchain clone flags by @mbaxter in #11293
- op-node/derive: linter fix by @KailMillor in #11265
- op-challenger: Add subcommand to continuously run trace providers by @ajsutton in #11294
- reduce scope of
expandedOwnersby @zhiqiangxu in #11301 - op-batcher: Add dynamic blob/calldata selection by @sebastianst in #11219
- contracts-bedrock: fix
SentMessageevent by @tynes in #11305 - feat: introduce SuperchainWETH and ETHLiquidity by @smartcontracts in #11042
- Cannon: optimize the example by @tanvirman in #11302
- op-node: optimize case to handle proposer requests by @protolambda in #11262
- Add Granite and Holocene fork references by @sebastianst in #11339
- Reduce ChannelTimeout to 50 in Granite by @Inphi in #11343
- fix: SuperchainWETH fuzz testing chain ID error by @smartcontracts in #11352
- op-challenger: Modify run_trace to run a block prior to the safe head by @ajsutton in #11349
- op-challenger: Add metric to report memory used in the cannon VM by @ajsutton in #11350
- op-proposer: retry failed output proposals by @bitwiseguy in #11291
- feat(challenger):
AsteriscKonatrace type by @clabby in #11140 - update sepolia-devnet-0 deploy-config by @Inphi in #11371
- op-node: abort block publishing if conductor commit fails by @protolambda in #11380
New Contributors
- @nidz-the-fact made their first contribution in #11177
- @DenseDenise made their first contribution in #11138
- @nnsW3 made their first contribution in #11190
- @rickck11 made their first contribution in #11191
- @yuhangcangqian made their first contribution in #11184
- @bravesasha made their first contribution in #11222
- @grubbyTimes made their first contribution in #11224
- @coderHarrii made their first contribution in #11233
- @0xtrait made their first contribution in #11241
- @caseylove made their first contribution in #11221
- @cbfyi made their first contribution in #11238
- @AntonCutthroat made their first contribution in #11264
- @DeeADouble made their first contribution in #11179
- @PinelliaC made their first contribution in #11296
- @KailMillor made their first contribution in #11265
- @tanvirman made their first contribution in #11302
- @blmalone made their first contribution in #11325
- @tooshiNoko made their first contribution in #11329
Full Changelog: v1.8.0...v1.9.0
🚢 Docker Images: