CI flake: Lints

[clockwork-labs-bot]

Detected by CI Gremlin after a failed GitHub Actions attempt succeeded on rerun.

Tests:

• Lints

Error:
Error: command ["pnpm", "lint"] exited with code 133
2026-07-29T21:26:26.7037822Z

Failed job runs:

• https://github.com/clockworklabs/SpacetimeDB/actions/runs/30492104931/job/90712317587 (Lints, attempt 1 -> 2)

Timeout detection:

• none

Context:

• none

[clockwork-labs-bot]

Diagnosis for the exact linked job/attempt: CI run 30492104931, job 90712317587, Lints, attempt 1.

This job was for PR #5618 (bfops/consolidate-default-tests) at head 42b607d73b54fb0c87517f8ec4bcdfffe5bbc186 on runner ghr-github-self-hosted-ubuntu-14900k-05-spacetimedb-14-73f087b2.

The job completed checkout, Rust/.NET/Node setup, and pnpm setup. The failure happened in the Run ci lint step while cargo ci lint was running pnpm lint. The failing workspace package was crates/bindings-typescript/test-app:

crates/bindings-typescript/test-app lint$ eslint . && prettier . --check --ignore-path ../../../.prettierignore
crates/bindings-typescript/test-app lint: # Fatal error in , line 0
crates/bindings-typescript/test-app lint: # unreachable code
crates/bindings-typescript/test-app lint: #FailureMessage Object: 0x7f55671fc2a0
crates/bindings-typescript/test-app lint:  2: 0x2a2193b V8_Fatal(char const*, ...) [node]
crates/bindings-typescript/test-app lint:  5: 0x2ff6dc3 v8::internal::compiler::turboshaft::BuildGraph(...)
crates/bindings-typescript/test-app lint: Failed
ERR_PNPM_RECURSIVE_RUN_FIRST_FAIL @clockworklabs/test-app@0.0.1 lint: `eslint . && prettier . --check --ignore-path ../../../.prettierignore`
Exit status 133
Error: command ["pnpm", "lint"] exited with code 133

Root cause visible in this exact attempt: Node/V8 aborted internally with fatal unreachable code in the Turboshaft optimizer path while running the TypeScript test-app lint command. This is not an ESLint rule failure, not a Prettier formatting diff, and not PR code directly failing a lint assertion; the Node process crashed underneath the lint task.

Suggested updated title:

CI flake: Lints test-app lint crashes in Node/V8 unreachable code

Duplicate check in Test Flakes:

• Top duplicate candidate: CI flake: v8 internal error during build #5366, the open public tracker for the Node/V8 Fatal error ... unreachable code internal-abort class.
• Exact repeat history: CI flake: Lints test-app lint crashes in Node/V8 unreachable code #5628 was created for the same run/job/attempt and the same crates/bindings-typescript/test-app lint crash, then closed as a duplicate of CI flake: v8 internal error during build #5366.
• Related but weaker: private Moves pnpm generate scripts into rust tools scripts #3539 has the same broad Node/V8 unreachable code / signal-5 class during pnpm format / Prettier, but in a different repo/job/workload. CI flake: Smoketests - pnpm build failed despite reporting success #5546 involves TypeScript package build instability, but its exact log did not show the V8 fatal unreachable code crash, so it is related rather than a close duplicate.

Top duplicate candidate: #5366. I am leaving this ticket open because #5366 is not one of the duplicate targets authorized for automatic closure by the followup request.