fix: submit channel refactors #94

dylanlott · 2025-05-22T15:36:12Z

fix: submit channel refactors

This PR includes several fixes and refactors to the channel submission logic.

Handle 403 errors with Skips instead of erroring
Adds retry_count increments to some code paths that weren't previously ticking the count
Adds per-slot limiting to the block building loop so that the retry function doesn't simply get called again after 3 attempts
Explicitly logs the call results of the blob transaction to aid with debugging
Adds gas bumping logic to the submit task to facilitate transaction replacement

Additionally, there are some configuration changes that needed to be made.

Additional broadcast hosts was just pointing back at the host RPC, which was adding noise to the logs and interfering with nonce debugging. This PR makes it optional and un-sets the duplicate host-rpc URL.
Slot offset and start timestamp were incorrectly configured and have been updated to their correct values

Closes ENG-1079

See: Builder Retry Logic

dylanlott · 2025-05-22T15:36:31Z

fix: submit channel refactors #94 : 2 dependent PRs (#96 , #98 ) 👈 (View in Graphite)
main

This stack of pull requests is managed by Graphite. Learn more about stacking.

Cargo.toml

Evalir

nice progress. some nits:

src/tasks/submit.rs

src/tasks/block/sim.rs

src/tasks/submit.rs

prestwich · 2025-05-28T12:26:23Z

src/tasks/submit.rs

-                "error in transaction submission"
-            );
-
+            // NB: These errors are all handled the same way but are logged for debugging purposes
            if e.as_revert_data()


style: DRY

macro_rules! check_selector { ($e:ident, $rd:ident, $selector:ident, $msg:literal) => { if $rd.starts_with(&$selector) { debug!("incorrect host block"); bail!(e) } } } if let Some(rd) = e.as_revert_data() { check_selector!(e, rd, IncorrectHostBlock::SELECTOR, "incorrect host block") check_selector!(e, rd, Zenith::BadSignature::SELECTOR, "...") ... bail!(e) }

started messing around here
https://github.com/init4tech/builder/pull/98/files

src/tasks/submit.rs

Evalir · 2025-05-30T07:15:57Z

Cargo.toml

+init4-bin-base = { version = "0.4.1", features = ["perms"] }

-signet-constants = { git = "https://github.com/init4tech/signet-sdk", rev = "b8251ff0fec7cb14ca87e6f95c14f56bc2593049" }
-signet-sim = { git = "https://github.com/init4tech/signet-sdk", rev = "b8251ff0fec7cb14ca87e6f95c14f56bc2593049" }
-signet-tx-cache = { git = "https://github.com/init4tech/signet-sdk", rev = "b8251ff0fec7cb14ca87e6f95c14f56bc2593049" }
-signet-types = { git = "https://github.com/init4tech/signet-sdk", rev = "b8251ff0fec7cb14ca87e6f95c14f56bc2593049" }
-signet-zenith = { git = "https://github.com/init4tech/signet-sdk", rev = "b8251ff0fec7cb14ca87e6f95c14f56bc2593049" }
+signet-constants = { git = "https://github.com/init4tech/signet-sdk", rev = "ba5894f6fac35299d495ae115cd465a1f0791637" }


note that technically there's a version mismatch here: bin-base will pull signet-constants from commit bd183b627dcb0eb682da801093b13f1f8311446b while this dep will be exclusively using this commit. This shouldn't cause problems right now, but it is something to be aware of. We could bump signet-constants as a patch release to make sure it's at the commit? @prestwich

Closes ENG-1109

Evalir

just a few nits, but i think we can proceed with this

Evalir · 2025-06-04T13:56:14Z

Cargo.toml

 init4-bin-base = { version = "0.4.1", features = ["perms"] }

-signet-constants = { git = "https://github.com/init4tech/signet-sdk", rev = "bd183b627dcb0eb682da801093b13f1f8311446b" }
-signet-sim = { git = "https://github.com/init4tech/signet-sdk", rev = "bd183b627dcb0eb682da801093b13f1f8311446b" }
-signet-tx-cache = { git = "https://github.com/init4tech/signet-sdk", rev = "bd183b627dcb0eb682da801093b13f1f8311446b" }
-signet-types = { git = "https://github.com/init4tech/signet-sdk", rev = "bd183b627dcb0eb682da801093b13f1f8311446b" }
-signet-zenith = { git = "https://github.com/init4tech/signet-sdk", rev = "bd183b627dcb0eb682da801093b13f1f8311446b" }
+signet-constants = { git = "https://github.com/init4tech/signet-sdk", rev = "ba5894f6fac35299d495ae115cd465a1f0791637" }
+signet-sim = { git = "https://github.com/init4tech/signet-sdk", rev = "ba5894f6fac35299d495ae115cd465a1f0791637" }
+signet-tx-cache = { git = "https://github.com/init4tech/signet-sdk", rev = "ba5894f6fac35299d495ae115cd465a1f0791637" }
+signet-types = { git = "https://github.com/init4tech/signet-sdk", rev = "ba5894f6fac35299d495ae115cd465a1f0791637" }
+signet-zenith = { git = "https://github.com/init4tech/signet-sdk", rev = "ba5894f6fac35299d495ae115cd465a1f0791637" }


let's upgrade this to the 0.2.0 tag (we can now just use tag instead)

unless we need test-utils or rpc, we can actually skip the tag = and just use published?

src/tasks/submit.rs

prestwich

just nits

prestwich · 2025-06-04T17:38:01Z

src/tasks/block/sim.rs

-        let block = block_build.build().await;
-        debug!(block = ?block, "finished block simulation");
+        let built_block = block_build.build().await;
+        debug!(block_number = ?built_block.block_number(), "finished building block");


the ? is unnecessary, as block_number is a u64 which is a valid tracing value

prestwich · 2025-06-04T17:38:26Z

src/tasks/block/sim.rs

@@ -155,8 +155,7 @@ impl Simulator {

            // If no env, skip this run
            let Some(block_env) = self.block_env.borrow_and_update().clone() else { return };
-
-            debug!(block_env = ?block_env, "building on block");
+            debug!(block_env = ?block_env, "building on block env");


debug!(?block_env, ...)
no need to do = as we're not renaming or invoking any expressions

prestwich · 2025-06-04T17:39:54Z

src/tasks/submit.rs

+
+        // manually retrieve nonce
+        let nonce =
+            self.provider().get_transaction_count(self.provider().default_signer_address()).await?;


good target for a utility function

prestwich · 2025-06-04T17:40:30Z

src/tasks/submit.rs

        let SendableTx::Envelope(tx) = self.provider().fill(tx).await? else {
            bail!("failed to fill transaction")
        };
+        debug!(tx_hash = ?tx.hash(), host_block_number = %resp.req.host_block_number, "sending transaction to network");


no need for the % on the host block number. it's a u64, right?

dylanlott · 2025-06-04T17:41:39Z

Merge activity

Jun 4, 5:41 PM UTC: @dylanlott merged this pull request with Graphite.

dylanlott changed the title ~~fix: remove arc and clones~~ fix: submit channel refactors May 22, 2025

dylanlott changed the base branch from main to prestwich/env-cache-integration May 23, 2025 02:57

dylanlott force-pushed the dylan/submit-channel-refactors branch from 9a88965 to 8a1e80b Compare May 23, 2025 04:05

dylanlott changed the base branch from prestwich/env-cache-integration to main May 23, 2025 04:06

dylanlott changed the base branch from main to prestwich/env-cache-integration May 23, 2025 04:06

dylanlott changed the base branch from prestwich/env-cache-integration to main May 23, 2025 04:07

dylanlott force-pushed the dylan/submit-channel-refactors branch from 8a1e80b to 0498aae Compare May 23, 2025 17:02

dylanlott changed the base branch from main to prestwich/env-cache-integration May 23, 2025 17:03

dylanlott commented May 23, 2025

View reviewed changes

Cargo.toml Outdated Show resolved Hide resolved

prestwich force-pushed the prestwich/env-cache-integration branch from a239bcb to 2fd44a7 Compare May 23, 2025 18:02

dylanlott force-pushed the dylan/submit-channel-refactors branch 2 times, most recently from 444d0f3 to f762131 Compare May 23, 2025 21:38

dylanlott changed the base branch from prestwich/env-cache-integration to main May 23, 2025 21:38

prestwich force-pushed the dylan/submit-channel-refactors branch from f762131 to 8c0eaa5 Compare May 27, 2025 12:34

dylanlott force-pushed the dylan/submit-channel-refactors branch 2 times, most recently from 2d47dc8 to 455791f Compare May 27, 2025 21:48

dylanlott self-assigned this May 27, 2025

dylanlott added the bug Something isn't working label May 27, 2025

dylanlott marked this pull request as ready for review May 27, 2025 22:00

dylanlott requested a review from Evalir as a code owner May 27, 2025 22:00

dylanlott mentioned this pull request May 28, 2025

feat: add host fills from built block to block header #96

Merged

Evalir requested changes May 28, 2025

View reviewed changes

prestwich requested changes May 28, 2025

View reviewed changes

prestwich reviewed May 29, 2025

View reviewed changes

src/tasks/submit.rs Outdated Show resolved Hide resolved

prestwich mentioned this pull request May 29, 2025

feat: simrevert type #98

Open

dylanlott force-pushed the dylan/submit-channel-refactors branch 3 times, most recently from 9ef1835 to 70591dd Compare May 29, 2025 22:31

dylanlott changed the base branch from main to evalir/chore-deps-upgrade-to-latest-alloy-sdk-trevm May 29, 2025 22:43

dylanlott temporarily deployed to dev May 30, 2025 02:06 — with GitHub Actions Inactive

dylanlott temporarily deployed to dev May 30, 2025 02:28 — with GitHub Actions Inactive

Base automatically changed from evalir/chore-deps-upgrade-to-latest-alloy-sdk-trevm to main May 30, 2025 04:22

dylanlott temporarily deployed to dev May 30, 2025 04:31 — with GitHub Actions Inactive

dylanlott temporarily deployed to dev May 30, 2025 05:01 — with GitHub Actions Inactive

dylanlott temporarily deployed to dev May 30, 2025 05:19 — with GitHub Actions Inactive

Evalir reviewed May 30, 2025

View reviewed changes

dylanlott force-pushed the dylan/submit-channel-refactors branch from 7371bc5 to 0a3a830 Compare May 30, 2025 20:01

Evalir and others added 10 commits May 30, 2025 15:10

chore(deps): upgrade to latest alloy/sdk/trevm

587f620

Closes ENG-1109

refactors: submit channel refactors

44b50ff

wip: detect and sleep during 403 slots

1b376b0

wip: adjusting mpfpg and gas limit to replace transaction in pool

6c2ed53

fix: retry logic bugs

d850791

refactor sim_with_call function

ba696d4

gas calculation refactors

1990579

address review feedback

934788c

bump to sdk version with block number updates

601c2d5

remove clone

0e8eda5

dylanlott force-pushed the dylan/submit-channel-refactors branch from 0a3a830 to 0e8eda5 Compare May 30, 2025 21:12

comment

e40f408

dylanlott mentioned this pull request Jun 4, 2025

passes sim result to the submit tasks #100

Draft

dylanlott requested review from Evalir and prestwich June 4, 2025 03:06

Evalir approved these changes Jun 4, 2025

View reviewed changes

prestwich approved these changes Jun 4, 2025

View reviewed changes

dylanlott merged commit a87da01 into main Jun 4, 2025
6 checks passed

dylanlott deleted the dylan/submit-channel-refactors branch June 4, 2025 17:41

fix: submit channel refactors #94

fix: submit channel refactors #94

Uh oh!

Conversation

dylanlott commented May 22, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

fix: submit channel refactors

Uh oh!

dylanlott commented May 22, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Evalir left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Evalir left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

prestwich left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

dylanlott commented Jun 4, 2025

Merge activity

Uh oh!

Uh oh!

dylanlott commented May 22, 2025 •

edited

Loading

dylanlott commented May 22, 2025 •

edited

Loading