chore: merge builder code #354

danyalprout · 2026-01-12T00:51:45Z

Description

Merge the latest op-rbuilder code into this repo base/op-rbuilder@23f42c8

This PR:

Minimal diff to integrate it into this repository (d0e7185, 2f7eac2, 8391c04)
Preserves the op-rbuilder history
Long term we'll remove logic we don't use from this and migrate to rblib for these crates.

Follow ups

In a couple of follow up PR's, we should:

reenable formatting
migrate to workspace lints
extract the binaries to bin/
delete the unused sections of the codebase
migrate to using the base/shared types instead of the tips types

## 📝 Summary Cleanup PR to remove unused variables. ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

## 📝 Summary Lovely reth changes. ## 💡 Motivation and Context Pectra ## ✅ I have completed the following steps: * [X] Run `make lint` * [X] Run `make test` * [ ] Added tests (if applicable) --------- Co-authored-by: Ferran Borreguero <[email protected]> Co-authored-by: shana <[email protected]> Co-authored-by: Solar Mithril <[email protected]> Co-authored-by: Vitaly Drogan <[email protected]>

## 📝 Summary  ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable)

## 📝 Summary This PR changes uses the incremental format for Flashblocks and adds support for it on tester. Generate a genesis file: ``` $ cargo run -p op-rbuilder --bin tester --features optimism -- genesis --output genesis.json ``` Run op-rbuilder with flashblocks ``` $ cargo run -p op-rbuilder --bin op-rbuilder --features optimism,flashblocks node --authrpc.port 4444 --authrpc.jwtsecret ./crates/op-rbuilder/src/tester/fixtures/test-jwt-secret.txt --chain ./genesis.json --datadir /tmp/builder2 --disable-discovery --http --http.port 8545 ``` Run the tester with flashblocks enabled: ``` $ cargo run -p op-rbuilder --bin tester --features optimism -- run --flashblocks-endpoint ws://localhost:1111 ``` Now, there are two streams available, one from the builder (websocat ws://localhost:1111) and another one from the tester which mimicks the Rollup-boost preconfirmations (websocat ws://localhost:1112). ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable) --------- Co-authored-by: Daniel Xifra <[email protected]>

## 📝 Summary Marks reverting transactions as invalid in the payload building process and removed dependents. Also removes reverting hashes from the tx pool. This would result in transactions that revert not be considered for future blocks once it's simulated once. ## 💡 Motivation and Context Remove the need to resimulate reverting transactions for revert protection. --- ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable)

## 📝 Summary Deposit command was not funding the address correctly ## 💡 Motivation and Context Tested deposits in integration tests --- ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable)

## 📝 Summary Add integration test to op-rbuilder that asserts that the block is built with priority fee ordering ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable)

## 📝 Summary Making flashblock websocket as flag so that it can be customizable. Also added receipts, account balance, block number into the metadata field, such that the RPC node can build the RPC response for `eth_transactionReceipt` and `eth_getBalance` for pending blocks. (#4) Added an integration test to test that the metadata field indeed has the new data. ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable)

## 📝 Summary Currently the block time is hard coded to assume the chain block time is 1s and flashblock time is 250ms. But for Base the chain block time is 2s. Thus fixing this by adding args to make it dynamic.  --- ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable)

## 📝 Summary The index for getting the receipt is incorrect as it's using an updated index, fix it. Added an additional assertion in integration test to test this.  ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

## 📝 Summary Moved ExecutionInfo and ExecutingPayload into primitives, combining them Now we could implement revert protection for Flashblocks too ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

## 📝 Summary  ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

## 📝 Summary This PR introduces a monitoring service that prints all the transactions that arrive to the transaction pool in op-rbuilder. ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

Implement SupervisorValidator to be used in rbuilder Add additional primitives crate, that is used for storing external ## 📝 Summary  ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable)

Corrected grammar: "transactions events" → "transaction events" for proper noun-adjective agreement. Preposition correction: "on" → "at" for referring to a website link properly. Corrected article usage: "any" → "a" for clarity.

## 📝 Summary  Bump deps to reth `v1.3.3` ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [ ] Added tests (if applicable) --------- Co-authored-by: Daniel Xifra <[email protected]>

## 📝 Summary I will remove previous interop version in favor of newly written reth implementation. After this PR i'll make a follow-up PR with dependency bump and additional logic for tx filtering. ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

## 📝 Summary When batcher's max channel duration is big enough (e.g. 10m), the batcher would be pushing its updates at specified intervals. This causes the sequencer to send an avalanche of FCUs (and getBlockByNumber) that push safe head step-by-step. As a consequence it can happen that the time b/w FCU and ensuing getPayload would be on the scale of ~2.5s. This means that we should "remember" the payloads long enough to accommodate that corner-case. --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

## 📝 Summary Fix deposit and flashblocks integration tests. Resolves flashbots/rbuilder#533 and flashbots/rbuilder#521 ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable)

Enhanced the OpRBuilderMetrics struct to include new metrics for tracking flashblocks, messages sent, and invalid blocks. Updated the OpPayloadBuilder to utilize these metrics during block building and transaction simulation processes. - Added `flashblock_count`, `messages_sent_count`, and `invalid_blocks_count` to OpRBuilderMetrics. - Integrated metrics tracking in OpPayloadBuilder for message sending, block building success, and transaction simulation. - Recorded durations for various operations to improve performance monitoring. ## 📝 Summary  ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

## 📝 Summary Bumps reth dependencies to `1.3.8`. ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [ ] Added tests (if applicable)

## 📝 Summary Adding reth 1.3.11 for the upcoming op chain sepolia Isthmus hardfork for op-rbuilder flashblock builder Also fixing request hash for Isthmus validity rules https://specs.optimism.io/protocol/isthmus/exec-engine.html#header-validity-rules. Tested deploying on Base devnet, it's able to sync again ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

## 📝 Summary Update the op-rbuilder devnet documentation: * Update it to use the builder-playground * Various fixes (e.g. remove optimism feature, use fully qualified docker image so it pulls automatically) ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable)

## 📝 Summary  ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

## 📝 Summary Add isthmus withdrawal hash fix on the vanilla builder. Previously was added to the flashblocks builder. ## 💡 Motivation and Context Fix errors around invalid blocks due to incorrect withdrawals hash --- ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable)

## 📝 Summary Fixes unbounded mem growth in monitoring task ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

## 📝 Summary Currently it's possible for the builder to make more Flashblocks than intended during a block (likely due to delays). If more flashblocks are made than desired (e.g. 11 flashblocks for a 2s block w/ 200ms Flashblocks), the 11th flashblock will have a gas limit that is over the blocks gas limit and will cause invalid blocks on the sequencer. ## 💡 Motivation and Context If the builder produces a full block that uses more gas than the gas limit, the local EL client will treat it as invalid. --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable)

## 📝 Summary 1. Update the version of op-reth used by CI 2. Make the CI script work on macOS as well as linux 3. Update the instructions for running/testing op-rbuilder ## 💡 Motivation and Context 1. The CI script was handy for downloading Reth 2. The instructions didn't work for me (due to the optimism feature being removed etc). --- ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable) --------- Co-authored-by: shana <[email protected]>

## 📝 Summary The builder block building times always slowly increase to a very high number on Base Sepolia, suspecting revert protection is causing failed txs to get backed up, which causes slow down in op-rbuilder block building. ``` # HELP reth_op_rbuilder_payload_num_tx_simulated_fail Number of transactions in the payload that failed simulation # TYPE reth_op_rbuilder_payload_num_tx_simulated_fail summary reth_op_rbuilder_payload_num_tx_simulated_fail{quantile="0"} 77 reth_op_rbuilder_payload_num_tx_simulated_fail{quantile="0.5"} 3198.061173550786 reth_op_rbuilder_payload_num_tx_simulated_fail{quantile="0.9"} 3223.1035892442424 reth_op_rbuilder_payload_num_tx_simulated_fail{quantile="0.95"} 3223.748274430612 reth_op_rbuilder_payload_num_tx_simulated_fail{quantile="0.99"} 3226.973635127358 reth_op_rbuilder_payload_num_tx_simulated_fail{quantile="0.999"} 3226.973635127358 reth_op_rbuilder_payload_num_tx_simulated_fail{quantile="1"} 3228 reth_op_rbuilder_payload_num_tx_simulated_fail_sum 64118848 reth_op_rbuilder_payload_num_tx_simulated_fail_count 21069 # HELP reth_op_rbuilder_payload_tx_simulation_duration Duration of payload simulation of all transactions # TYPE reth_op_rbuilder_payload_tx_simulation_duration summary reth_op_rbuilder_payload_tx_simulation_duration{quantile="0"} 0.018329514 reth_op_rbuilder_payload_tx_simulation_duration{quantile="0.5"} 0.49002444205448686 reth_op_rbuilder_payload_tx_simulation_duration{quantile="0.9"} 0.5026304818952316 reth_op_rbuilder_payload_tx_simulation_duration{quantile="0.95"} 0.5079867028401951 reth_op_rbuilder_payload_tx_simulation_duration{quantile="0.99"} 0.553164415015488 reth_op_rbuilder_payload_tx_simulation_duration{quantile="0.999"} 0.553164415015488 reth_op_rbuilder_payload_tx_simulation_duration{quantile="1"} 0.555461782 reth_op_rbuilder_payload_tx_simulation_duration_sum 9654.159540928 reth_op_rbuilder_payload_tx_simulation_duration_count 25870 # HELP reth_op_rbuilder_flashblock_build_duration Flashblock build duration # TYPE reth_op_rbuilder_flashblock_build_duration summary reth_op_rbuilder_flashblock_build_duration{quantile="0"} 0.055144161 reth_op_rbuilder_flashblock_build_duration{quantile="0.5"} 0.5166962986266059 reth_op_rbuilder_flashblock_build_duration{quantile="0.9"} 0.5348868728374531 reth_op_rbuilder_flashblock_build_duration{quantile="0.95"} 0.5672812838109652 reth_op_rbuilder_flashblock_build_duration{quantile="0.99"} 0.583272957920482 reth_op_rbuilder_flashblock_build_duration{quantile="0.999"} 0.583272957920482 reth_op_rbuilder_flashblock_build_duration{quantile="1"} 0.60984729 reth_op_rbuilder_flashblock_build_duration_sum 9081.926654393968 reth_op_rbuilder_flashblock_build_duration_count 21065 # HELP reth_op_rbuilder_payload_num_tx_simulated_success Number of transactions in the payload that were successfully simulated # TYPE reth_op_rbuilder_payload_num_tx_simulated_success summary reth_op_rbuilder_payload_num_tx_simulated_success{quantile="0"} 1 reth_op_rbuilder_payload_num_tx_simulated_success{quantile="0.5"} 11.000052032263987 reth_op_rbuilder_payload_num_tx_simulated_success{quantile="0.9"} 20.999528921206892 reth_op_rbuilder_payload_num_tx_simulated_success{quantile="0.95"} 24.001108231483542 reth_op_rbuilder_payload_num_tx_simulated_success{quantile="0.99"} 29.00012111115082 reth_op_rbuilder_payload_num_tx_simulated_success{quantile="0.999"} 29.00012111115082 reth_op_rbuilder_payload_num_tx_simulated_success{quantile="1"} 60 reth_op_rbuilder_payload_num_tx_simulated_success_sum 136029 reth_op_rbuilder_payload_num_tx_simulated_success_count 21069 ```  ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

## 📝 Summary This PR introduces three changes: - It refactors some of the common utilities shared among the tests into a single `TestHarness` struct which spins the test framework with all the components. Note that it does not update the tests to use this new utility. - It fixes an issue with the block builder that would stop the block building process and not return any block if a cancel request was found. This happens when an FCU and a getPayload request are called to close to each other, the getPayload cancels the block building process, and getPayload waits forever for a block that will never be built. Now, the block building finishes. - It adds an integration test to cover this use case with the new utility. ## 💡 Motivation and Context  --- ## ✅ I have completed the following steps: * [ ] Run `make lint` * [ ] Run `make test` * [ ] Added tests (if applicable)

Move config extraction directly into extension constructor calls to reduce intermediate variables in main.rs.

Migrates the following crates from op-rbuilder repository: - op-rbuilder: Main builder application - p2p: P2P networking layer - tdx-quote-provider: TDX attestation service All crates are placed under crates/builder/ with full git history preserved.

The tss-esapi-sys crate requires libtss2-dev to build, which was missing from the udeps job causing CI failures.

cb-heimdall · 2026-01-12T04:35:12Z

✅ Heimdall Review Status

Requirement Status More Info

Reviews

✅ 1/1

Denominator calculation

Show calculation

1 if user is bot 0

1 if user is external 0

2 if repo is sensitive 0

From .codeflow.yml 1

Additional review requirements

Show calculation

Max	0

0

From CODEOWNERS 0

Global minimum 0

Max 1

1

1 if commit is unverified 1

Sum 2

haardikk21

Nice, looks good. I can take care of the follow up PRs here as I'm going to be working on the builder now anyway

…ilder

refcell · 2026-01-12T13:35:15Z

Cargo.toml

+
+# base
+concurrent-queue = "2.5.0"
+tips-core = { git = "https://github.com/base/tips", rev = "c08eaa4fe10c26de8911609b41ddab4918698325", default-features = false }


non-blocking; we should be able to use base-bundles instead of tips-core. Can be done in a follow-up.

Will do in a follow up 👍

ferranbt and others added 30 commits February 12, 2025 19:47

Fix integration tests (#536)

d7c0c23

## 📝 Summary Fix deposit and flashblocks integration tests. Resolves flashbots/rbuilder#533 and flashbots/rbuilder#521 ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [x] Added tests (if applicable)

deps: reth v1.3.8 (#553)

5230230

## 📝 Summary Bumps reth dependencies to `1.3.8`. ## ✅ I have completed the following steps: * [x] Run `make lint` * [x] Run `make test` * [ ] Added tests (if applicable)

danyalprout added 7 commits January 11, 2026 16:11

review feedback

49f6d1c

chore: inline config extraction in extension constructors

b439aa8

Move config extraction directly into extension constructor calls to reduce intermediate variables in main.rs.

migrate back to generic install_ext

7aa0f89

minimal changes to integrate project

d0e7185

fix running tests locally (we have oltp env vars on macs

2f7eac2

fix: add native dependencies to udeps CI job

8391c04

The tss-esapi-sys crate requires libtss2-dev to build, which was missing from the udeps job causing CI failures.

danyalprout force-pushed the danyal/merge-builder branch from 09cc540 to 8391c04 Compare January 12, 2026 01:03

danyalprout requested review from cody-wang-cb, niran and refcell January 12, 2026 01:33

danyalprout marked this pull request as ready for review January 12, 2026 02:00

Base automatically changed from danyal/part-eleven-gha-script to main January 12, 2026 04:35

haardikk21 previously approved these changes Jan 12, 2026

View reviewed changes

Merge branch 'main' of github.com:base/node-reth into danyal/merge-bu…

fe50834

…ilder

haardikk21 dismissed their stale review via fe50834 January 12, 2026 05:50

haardikk21 mentioned this pull request Jan 12, 2026

feat: create a system tests environment #357

Open

refcell mentioned this pull request Jan 12, 2026

chore(ci): Cache APT Packages #360

Open

refcell reviewed Jan 12, 2026

View reviewed changes

refcell approved these changes Jan 12, 2026

View reviewed changes

danyalprout added this pull request to the merge queue Jan 12, 2026

Merged via the queue into main with commit ce34d66 Jan 12, 2026
27 checks passed

danyalprout deleted the danyal/merge-builder branch January 12, 2026 14:32

This was referenced Jan 12, 2026

chore(workspace): Remove Legacy tips-core Dependency #361

Open

chore(tdx): Migrate the TDX Binary to bin/ #362

Closed

haardikk21 mentioned this pull request Jan 12, 2026

chore: remove dependency on tips-core for builder #363

Open

This was referenced Jan 12, 2026

chore(tdx): Cleanup lib.rs #364

Closed

chore(tdx): Split Out Mock Provider #365

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

chore: merge builder code #354

chore: merge builder code #354

danyalprout commented Jan 12, 2026 •

edited

Loading

Uh oh!

cb-heimdall commented Jan 12, 2026 •

edited

Loading

Uh oh!

haardikk21 left a comment •

edited

Loading

Uh oh!

refcell Jan 12, 2026

Uh oh!

danyalprout Jan 12, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

20 participants

chore: merge builder code #354

chore: merge builder code #354

Conversation

danyalprout commented Jan 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Follow ups

Uh oh!

cb-heimdall commented Jan 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

✅ Heimdall Review Status

Uh oh!

haardikk21 left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

refcell Jan 12, 2026

Choose a reason for hiding this comment

Uh oh!

danyalprout Jan 12, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

20 participants

danyalprout commented Jan 12, 2026 •

edited

Loading

cb-heimdall commented Jan 12, 2026 •

edited

Loading

haardikk21 left a comment •

edited

Loading