feat: batcher alt-DA CLI flags and service wiring by patnir · Pull Request #3615 · base/base

patnir · 2026-06-18T00:44:35Z

wire batcher CLI flags for alt-DA dual-write on privacy devnet:

Summary

Add BATCHER_ALTDA_ENABLED and BATCHER_ALTDA_DA_SERVER CLI flags
Inject base-alt-da client via binary adapter (keeps crate layering)

Context

Depends on #3614. Enables dual-write on the privacy L3 devnet once merged; privacy-enclave compose follow-up sets flags on base-batcher after this stack lands.

Test plan

cargo test -p base-batcher-bin
cargo check -p base-batcher-bin

type=routine
risk=low
impact=sev5

linear · 2026-06-18T00:44:39Z

PRIV-1972

github-actions · 2026-06-18T00:48:37Z

+            alt_da: if self.altda_enabled {
+                let server_url = self
+                    .altda_da_server
+                    .ok_or_else(|| eyre::eyre!("--altda-enabled requires --altda-da-server"))?;
+                let client = base_alt_da::Client::new(server_url)
+                    .map_err(|e| eyre::eyre!("failed to create alt-da client: {e}"))?;
+                Some(std::sync::Arc::new(AltDaClientAdapter(client))
+                    as base_batcher_core::DynAltDaClient)
+            } else {
+                None
+            },


nit: --altda-da-server without --altda-enabled is silently discarded. An operator could set the server URL, forget the enable flag, and wonder why alt-DA isn't working. Consider warning in validate() (or here) when altda_da_server.is_some() && !altda_enabled:

if !self.altda_enabled && self.altda_da_server.is_some() { tracing::warn!("--altda-da-server is set but --altda-enabled is false; alt-DA is disabled"); }

github-actions · 2026-06-18T00:48:39Z

 url.workspace = true
 eyre.workspace = true
+base-alt-da.workspace = true
 base-protocol.workspace = true
 base-cli-utils.workspace = true
+async-trait.workspace = true
 base-batcher-core.workspace = true
 alloy-signer-local.workspace = true


Per project convention (CLAUDE.md): dependencies should be sorted by line length (waterfall style). async-trait (shorter) is listed after base-cli-utils (longer). Suggested order:

Suggested change

url.workspace = true

eyre.workspace = true

base-alt-da.workspace = true

base-protocol.workspace = true

base-cli-utils.workspace = true

async-trait.workspace = true

base-batcher-core.workspace = true

alloy-signer-local.workspace = true

url.workspace = true

eyre.workspace = true

async-trait.workspace = true

base-alt-da.workspace = true

base-protocol.workspace = true

base-cli-utils.workspace = true

base-batcher-core.workspace = true

alloy-signer-local.workspace = true

github-actions · 2026-06-18T06:45:33Z

+        if self.alt_da.is_some() && self.encoder_config.da_type != DaType::Calldata {
+            eyre::bail!(
+                "alt-da dual-write requires --data-availability-type calldata; \
+                 blob submissions are not dual-written to the DA server"
+            );
+        }


When alt_da is enabled and force_blobs_when_throttling is true (the default), a DA-backlog throttle event will silently switch the encoder to blob mode and stop dual-writing to alt-DA — the submission queue only clones alt_da_body for DaType::Calldata. An operator relying on the dual-write shadow path for validation may not realize it has gaps during throttle windows.

Consider warning here (or rejecting the combination) so operators are aware:

if self.alt_da.is_some() && self.force_blobs_when_throttling { tracing::warn!( "alt-da dual-write will pause during DA-backlog throttle events \ (throttling forces blob mode, which skips alt-DA); \ pass --no-force-blobs-when-throttling to keep dual-write continuous" ); }

- Client::put uploads batch bytes and returns a validated 34-byte generic commitment for L1 posting - read a bounded snippet of the server error body on non-2xx responses so operators see diagnostics - encode_commitment_tx_data prefixes DERIVATION_VERSION_1 for the commitment L1 tx - CommitmentError decouples commitment validation from the server Error enum - drop redundant reqwest dev-dependency

Move test imports to mod scope and drop redundant DERIVATION_VERSION_1 import.

Fire-and-forget PUT plus commitment txs alongside calldata, with shadow concurrency cap and reorg abort.

Log submission ids in shadow tasks, warn on missing commitment block number, and avoid redundant alt_da clone in SubmissionQueue::new.

Fix lib.rs re-export ordering and shadow-path indentation.

Add BATCHER_ALTDA_ENABLED and BATCHER_ALTDA_DA_SERVER; inject base-alt-da client via binary adapter.

Warn when alt-DA server URL is set without enable flag or when dual-write conflicts with throttle-driven blob override; sort batcher bin deps by line length.

Wrap long warn! string to satisfy workspace rustfmt.

github-actions · 2026-06-18T20:23:43Z

Review Summary

Clean PR that adds alt-DA CLI flags and wires the base-alt-da client into the batcher via a well-structured adapter pattern. The binary-level adapter (AltDaClientAdapter) correctly keeps the crate layering intact — base-batcher-core defines the trait, the binary bridges to the concrete base-alt-da client, and the service crate stays unaware of the infra dependency.

Architecture

The adapter pattern here is the right call. The AltDaClient trait in base-batcher-core takes alloy_primitives::Bytes while base_alt_da::Client::put takes &[u8]; the adapter bridges this cleanly with body.as_ref(). The GenericCommitment types are structurally identical [u8; 34] arrays in both crates, so the implicit coercion through the Result mapping is sound.

Validation

The validation has good layered coverage:

CLI-level: warns on --altda-da-server without --altda-enabled, warns on throttle+alt-DA interaction
Config-level: validate() rejects alt_da.is_some() when DA type is not calldata
Client construction: errors if --altda-enabled is set without --altda-da-server

Test coverage is adequate — the three new tests cover the missing-server-URL, happy-path, and blob-DA-rejection cases.

Existing inline comments

The three inline comments from the previous review run are still relevant and cover the notable findings. No additional issues found.

github-actions · 2026-06-18T20:30:32Z

❌ base-std fork tests did not run

The build or setup step failed before any tests could execute. Check the workflow logs for details.

- Stub alt_da: None in batcher service until #3615 wires the client - Collapse nested if in shadow path for clippy - Point UPGRADES.md hardfork test link at l3 branch

patnir · 2026-06-19T16:55:41Z

Combined into #3614 for PRIV-1972.

github-actions Bot reviewed Jun 18, 2026

View reviewed changes

patnir force-pushed the rahulpatni/priv-1972-3-batcher-shadow branch from 6306d4b to fb85817 Compare June 18, 2026 06:37

patnir force-pushed the rahulpatni/priv-1972-4-batcher-cli branch from c8801b6 to bb1063d Compare June 18, 2026 06:37

patnir force-pushed the rahulpatni/priv-1972-3-batcher-shadow branch from fb85817 to 4c02168 Compare June 18, 2026 06:42

patnir force-pushed the rahulpatni/priv-1972-4-batcher-cli branch from bb1063d to 005bfaa Compare June 18, 2026 06:42

github-actions Bot reviewed Jun 18, 2026

View reviewed changes

patnir added 4 commits June 18, 2026 13:05

chore: address alt-da PR review nits

66e1b7f

Move test imports to mod scope and drop redundant DERIVATION_VERSION_1 import.

feat: batcher alt-DA shadow dual-write

4437505

Fire-and-forget PUT plus commitment txs alongside calldata, with shadow concurrency cap and reorg abort.

fix: address alt-DA shadow submission review nits

3ca9749

Log submission ids in shadow tasks, warn on missing commitment block number, and avoid redundant alt_da clone in SubmissionQueue::new.

patnir force-pushed the rahulpatni/priv-1972-3-batcher-shadow branch from 4c02168 to 3ca9749 Compare June 18, 2026 20:08

patnir force-pushed the rahulpatni/priv-1972-4-batcher-cli branch from 005bfaa to c0c56ed Compare June 18, 2026 20:09

patnir mentioned this pull request Jun 18, 2026

feat: batcher alt-DA dual-write alongside calldata #3597

Closed

patnir added 3 commits June 18, 2026 13:11

chore: rustfmt batcher core

20e24f1

Fix lib.rs re-export ordering and shadow-path indentation.

feat: batcher alt-DA CLI flags and service wiring

a90d9e4

Add BATCHER_ALTDA_ENABLED and BATCHER_ALTDA_DA_SERVER; inject base-alt-da client via binary adapter.

chore: address alt-DA CLI review nits

9bfee0f

Warn when alt-DA server URL is set without enable flag or when dual-write conflicts with throttle-driven blob override; sort batcher bin deps by line length.

patnir force-pushed the rahulpatni/priv-1972-4-batcher-cli branch from c0c56ed to 9bfee0f Compare June 18, 2026 20:11

chore: rustfmt batcher CLI warnings

9099b32

Wrap long warn! string to satisfy workspace rustfmt.

patnir force-pushed the rahulpatni/priv-1972-3-batcher-shadow branch from 20e24f1 to 2e8679c Compare June 18, 2026 20:56

patnir added a commit that referenced this pull request Jun 18, 2026

fix: unblock PR 3614 CI failures

4f1fae8

- Stub alt_da: None in batcher service until #3615 wires the client - Collapse nested if in shadow path for clippy - Point UPGRADES.md hardfork test link at l3 branch

patnir closed this Jun 19, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat: batcher alt-DA CLI flags and service wiring#3615

feat: batcher alt-DA CLI flags and service wiring#3615
patnir wants to merge 8 commits into
rahulpatni/priv-1972-3-batcher-shadowfrom
rahulpatni/priv-1972-4-batcher-cli

patnir commented Jun 18, 2026 •

edited

Loading

Uh oh!

linear Bot commented Jun 18, 2026

Uh oh!

github-actions Bot Jun 18, 2026

Uh oh!

github-actions Bot Jun 18, 2026

Uh oh!

github-actions Bot Jun 18, 2026

Uh oh!

github-actions Bot commented Jun 18, 2026

Uh oh!

github-actions Bot commented Jun 18, 2026

Uh oh!

patnir commented Jun 19, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

patnir commented Jun 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Context

Test plan

Uh oh!

linear Bot commented Jun 18, 2026

Uh oh!

github-actions Bot Jun 18, 2026

Choose a reason for hiding this comment

Uh oh!

github-actions Bot Jun 18, 2026

Choose a reason for hiding this comment

Uh oh!

github-actions Bot Jun 18, 2026

Choose a reason for hiding this comment

Uh oh!

github-actions Bot commented Jun 18, 2026

Review Summary

Architecture

Validation

Existing inline comments

Uh oh!

github-actions Bot commented Jun 18, 2026

❌ base-std fork tests did not run

Uh oh!

patnir commented Jun 19, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

patnir commented Jun 18, 2026 •

edited

Loading