ci(#606): Gitea-compatible trx upload + deterministic fixes for the CI-load test flakes

- v2-ci.yml: upload-artifact@v4 -> @v3 + continue-on-error (act_runner reports as GHES and
  v4 refuses GHES, failing every run); runner-capacity diagnostic step; AKKA_TEST_TIMEFACTOR=3
  on the test step.
- Runtime/ControlPlane harnesses: akka.test.timefactor from AKKA_TEST_TIMEFACTOR
  (TestTimeFactor), single-expect-default 15s; ControlPlane gains PresenceBudget.
- Leaked spinner threads: the infinite-loop timeout tests (Calculation, Host.IntegrationTests
  RoslynVirtualTagEvaluator, 3x Core.Scripting) pinned a core for the rest of the test process
  (abandoned, never stopped). Now a bounded 5 s wall-clock spin; outer compile-inclusive bound 30 s.
- Real test races fixed: ScriptedAlarmEngine push-during-drain (Inactive is also the start
  state + persist-before-update => wait for the Cleared save); DriverReconnectE2e (both nodes
  host the driver; Created[] is last-writer-wins and ForceReconnect is a no-op while
  Connecting); ControlPlaneStarterOrder (lost-race heal waited for a 60 s refresh tick).
- Fixed sleeps -> poll-for-condition: ReconnectSupervisor LastError, Modbus deadband steps
  (wait for observed reads), Modbus dispose hygiene precondition, FOCAS probe-loop log.
- Short presence budgets -> PresenceBudget in the Runtime/ControlPlane tests seen failing on
  Gitea; Calculation Settle 10 -> 30 s. No assertion removed, nothing skipped.
- infra/README.md §10 CI; stale TreatWarningsAsErrors comment in Directory.Build.props.

Claude-Session: https://claude.ai/code/session_01Ef2ozxS7QPanGT8uoATbTe
This commit is contained in:
Joseph Doherty
2026-10-08 19:44:31 -04:00
parent 8801631739
commit 55ae230fa4
36 changed files with 449 additions and 128 deletions
@@ -13,6 +13,26 @@ namespace ZB.MOM.WW.OtOpcUa.Core.Scripting.Tests;
[Trait("Category", "Unit")]
public sealed class TimedScriptEvaluatorTests
{
/// <summary>
/// A CPU-bound script that outlives every timeout in this file but still ENDS.
/// </summary>
/// <remarks>
/// A timed-out script is abandoned, not stopped — its pool thread keeps running (the documented
/// <see cref="TimedScriptEvaluator{TContext, TResult}"/> leak). The previous body,
/// <c>long acc = 0; while (true) { acc++; if (acc &lt; 0) break; }</c>, needs 2^63 increments to
/// overflow, so each of the three tests using it pinned one core at 100 % for the rest of the test
/// process — CPU the parallel suites on a small CI runner were starving for (#606). A 5 s wall-clock
/// spin is still one to two orders of magnitude past the 30–80 ms timeouts/cancels asserted here, so
/// the script cannot finish first even under heavy load, and the abandoned thread exits on its own.
/// <c>System.DateTime</c> is outside the sandbox deny-list; Thread/Environment/Diagnostics are not.
/// </remarks>
private const string BoundedSpinScript =
"""
var end = System.DateTime.UtcNow.AddSeconds(5);
while (System.DateTime.UtcNow < end) { }
return 1;
""";
/// <summary>Verifies that fast scripts complete under timeout and return value.</summary>
[Fact]
public async Task Fast_script_completes_under_timeout_and_returns_value()
@@ -33,13 +53,8 @@ public sealed class TimedScriptEvaluatorTests
{
// Scripts can't reach the sandbox-denied process surface (System.Threading.Thread
// and System.Environment are both denied — Core.Scripting-001). A tight CPU loop
// that never returns exceeds any short timeout without touching a forbidden type.
var inner = ScriptEvaluator<FakeScriptContext, int>.Compile(
"""
long acc = 0;
while (true) { acc++; if (acc < 0) break; }
return 1;
""");
// that outlives the timeout exceeds it without touching a forbidden type.
var inner = ScriptEvaluator<FakeScriptContext, int>.Compile(BoundedSpinScript);
var timed = new TimedScriptEvaluator<FakeScriptContext, int>(
inner, TimeSpan.FromMilliseconds(50));
@@ -56,12 +71,7 @@ public sealed class TimedScriptEvaluatorTests
// A CPU-bound script that would otherwise timeout; external ct fires first.
// Expected: OperationCanceledException (not ScriptTimeoutException) so shutdown
// paths aren't misclassified as timeouts.
var inner = ScriptEvaluator<FakeScriptContext, int>.Compile(
"""
long acc = 0;
while (true) { acc++; if (acc < 0) break; }
return 1;
""");
var inner = ScriptEvaluator<FakeScriptContext, int>.Compile(BoundedSpinScript);
var timed = new TimedScriptEvaluator<FakeScriptContext, int>(
inner, TimeSpan.FromSeconds(5));
@@ -127,12 +137,7 @@ public sealed class TimedScriptEvaluatorTests
[Fact]
public async Task ScriptTimeoutException_message_points_at_diagnostic_path()
{
var inner = ScriptEvaluator<FakeScriptContext, int>.Compile(
"""
long acc = 0;
while (true) { acc++; if (acc < 0) break; }
return 1;
""");
var inner = ScriptEvaluator<FakeScriptContext, int>.Compile(BoundedSpinScript);
var timed = new TimedScriptEvaluator<FakeScriptContext, int>(
inner, TimeSpan.FromMilliseconds(30));