Fix flaky PBTS tests by anchoring the proposal wait deadline to now

[masih]

TestTooFarInThePastProposal (and the rest of the PBTS harness tests) time out waiting for the height-5-pattern proposal under -race on CI. observedValidatorProposerHeight waited for that proposal with time.Until(previousBlockTime.Add(ensureTimeout)), but previousBlockTime is a simulated block timestamp (firstBlockTime + height4ProposedBlockOffset, which several tests leave at zero), so the deadline is effectively "five seconds after the first block's timestamp" - a budget shared by every preceding height of the run rather than by this one step. Each height takes roughly half a second with the race detector on, so about 4.5 s of that budget is spent before the final height starts and any scheduler noise makes the remainder expire. In the failing run the first block was at 15:43:12.38 and the timeout fired at 15:43:17.41.

The wait now uses the same ensureTimeout measured from the moment the harness starts waiting, like every other ensure* helper, and the now-unused previousBlockTime parameter is dropped.

Flaked in: https://github.com/sei-protocol/sei-chain/actions/runs/34989800080/job/104451193066?pr=4184

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment, CI, and merge conflict monitoring

[github-actions]

The latest Buf updates on your PR. Results from workflow Buf / buf (pull_request).

Build	Format	Lint	Breaking	Updated (UTC)
✅ passed	✅ passed	✅ passed	✅ passed	Sep 16, 2026, 9:31 AM

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 65.52%. Comparing base (7e3ca24) to head (7e7e78f).
⚠️ Report is 5 commits behind head on giga-1.

Additional details and impacted files

@@            Coverage Diff             @@
##           giga-1    #4195      +/-   ##
==========================================
- Coverage   65.55%   65.52%   -0.03%     
==========================================
  Files        2081     2076       -5     
  Lines      157460   157021     -439     
==========================================
- Hits       103222   102895     -327     
+ Misses      54097    53985     -112     
  Partials      141      141              

Flag	Coverage Δ
sei-chain-pr	83.66% <ø> (?)
sei-db	?

Flags with carried forward coverage won't be shown. Click here to find out more.
see 7 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[cursor]

PR Summary

Low Risk
Test-only change to proposal wait timing in the PBTS harness; no production consensus behavior is modified.

Overview
Fixes flaky PBTS harness timeouts under -race on CI by changing how observedValidatorProposerHeight waits for the observed validator’s proposal.

Before: ensureProposalWithTimeout used time.Until(previousBlockTime.Add(ensureTimeout)), so the deadline was tied to a simulated prior block time (often firstBlockTime + height4ProposedBlockOffset). That budget was effectively shared across earlier heights in the run, so by height 5 little wall-clock time remained and the wait could expire spuriously.

After: The wait uses a fixed ensureTimeout from when the helper starts waiting—the same pattern as nextHeight and other ensure* helpers. The unused previousBlockTime parameter is removed from observedValidatorProposerHeight and all call sites (run, height5).

^{Reviewed by Cursor Bugbot for commit 7e7e78f. Bugbot is set up for automated code reviews on this repo. Configure here.}

[seidroid]

Anchors the PBTS harness's proposal wait to ensureTimeout from the moment waiting starts instead of a simulated block timestamp, which correctly fixes a shared-budget flake and makes the call site consistent with every other ensure* helper. No test depends on the old (shorter) deadline expiring, and the removed parameter's inputs remain used elsewhere, so nothing is left dangling.

Findings: 0 blocking | 0 non-blocking | 0 posted inline

Blockers

• None at the file/PR level.

Non-blocking

• None at the file/PR level.