Skip to content

fix(op-deployer): stabilize TestSuperchain flaky test#19582

Closed
ajsutton wants to merge 1 commit intodevelopfrom
aj/fix/flake-test-superchain
Closed

fix(op-deployer): stabilize TestSuperchain flaky test#19582
ajsutton wants to merge 1 commit intodevelopfrom
aj/fix/flake-test-superchain

Conversation

@ajsutton
Copy link
Copy Markdown
Contributor

@ajsutton ajsutton commented Mar 17, 2026

Summary

Fixes #19586

  • Root cause: testSuperchain called t.Parallel(), causing mainnet and sepolia subtests to run concurrently. Each starts a RetryProxy + Anvil forking from an external RPC. Under CI load, this resource contention causes timeouts.
  • Fix: Remove t.Parallel() — same fix applied to TestImplementations in op-deployer: harden TestImplementations and cli_runner #18019.
  • Flake count: 45

Test plan

  • go build ./op-deployer/... passes
  • CI passes on this PR

🤖 Generated with Claude Code

Remove t.Parallel() from testSuperchain to prevent concurrent Anvil
instances and external RPC calls from competing for resources under CI
load. This mirrors the same fix applied to TestImplementations in
3a372df which resolved the identical flake pattern in that test.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@ajsutton
Copy link
Copy Markdown
Contributor Author

Claude: Closing this PR — the RetryProxy hardening in #19593 (longer timeouts, more retries, startup race fix) should stabilize TestSuperchain as well, since both tests use devnet.NewForked which creates a RetryProxy internally. Removing t.Parallel() is unnecessary if the proxy is resilient enough.

@ajsutton ajsutton closed this Mar 17, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

flaky test: TestSuperchain

1 participant