fix(pegboard): validate drain_grace_period < request_lifespan

[abcxff]

when drain_grace_period >= request_life_span, we get undefined behaviour. This enforces this on the server side befoe pool config is upserted

[abcxff]

Warning

This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
Learn more

• fix(pegboard): validate drain_grace_period < request_lifespan #4998 👈 (View in Graphite)
• fix(rivetkit): bind methods through createWriteThroughProxy #4987
• chore(rivetkit): log inspector errors + standardize pino errorKey error #4986
• [slop]fix(rivetkit): remove async drizzle db proxy #4984 : 2 other dependent PRs (#4985 , #4991 )
• main

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

[railway-app]

🚅 Deployed to the rivet-pr-4998 environment in rivet-frontend

Service	Status	Web	Updated (UTC)
website	✅ Success (View Logs)	Web	May 9, 2026 at 1:50 am
mcp-hub	✅ Success (View Logs)	Web	May 7, 2026 at 8:18 pm
kitchen-sink	❌ Build Failed (View Logs)	Web	May 7, 2026 at 8:18 pm
frontend-cloud	❌ Build Failed (View Logs)	Web	May 7, 2026 at 8:18 pm
frontend-inspector	❌ Build Failed (View Logs)	Web	May 7, 2026 at 8:18 pm
ladle	❌ Build Failed (View Logs)	Web	May 7, 2026 at 8:15 pm

[claude]

Code Review

Overview

This PR adds a validation guard in pegboard_runner_config_upsert to reject serverless runner configs where drain_grace_period >= request_lifespan. The change is minimal and targeted: it destructures request_lifespan from the match arm and inserts a single comparison check before the existing drain_grace_period > actor_stop_threshold check.

---

What's Good

• Follows existing patterns exactly. The error type, format style, and placement within the validation block are consistent with the surrounding checks (e.g. slots_per_runner == 0, header length).
• Clear error message. The format string surfaces both values and their units ({drain_grace_period}s >= {request_lifespan}s), giving operators an actionable diagnostic.
• Correct comparison. Both fields are u32 seconds, and >= is the right operator since equal values (e.g. both 30s) are also semantically invalid — the drain would expire at the exact moment the request does, leaving no margin.

---

Issues / Suggestions

Missing test coverage

No test is added for this new validation path. This is the most significant gap. The PR description checklist item for tests is unchecked. Suggested: add a test that calls pegboard_runner_config_upsert with drain_grace_period == request_lifespan and with drain_grace_period > request_lifespan, verifying RunnerConfig::Invalid is returned in both cases.

Edge case: request_lifespan = 0

If a caller passes request_lifespan = 0, the new check drain_grace_period >= 0 is always true for a u32, so the validation correctly rejects it — but the error message (must be less than request_lifespan (Xs >= 0s)) will be confusing. Consider whether request_lifespan > 0 should be validated explicitly with its own error before this check, similar to the slots_per_runner == 0 guard.

PR is still a draft

The description template is unfilled and all checklist items are unchecked. Nothing blocking from a code standpoint, but worth completing before requesting review.

---

Summary

The logic is correct and well-placed. Main asks before merging: add a test covering the new validation branch, and consider an explicit request_lifespan > 0 guard to avoid a misleading error message when that field is zero.