Joseph Doherty
0b7c53f64f
Comparing full-suite runs on this branch against master in a clean worktree: both fail 11 tests, 9 identical. Master's two extras are load-flaky unit tests (AbCip Probe_loops, Galaxy EventPumpBoundedChannel); this branch's two extras were both Host.IntegrationTests deploy-path tests — the area this change re-times — so they are attributable here rather than to background flakiness. Cause is margin, not correctness. These waits used to observe a coordinator that sealed instantly on an empty expected-ack set; they now observe a real ApplyAck round-trip from every configured node. 15s was enormous margin against "instant" and thin against the real thing, so they failed only under full-suite CPU contention. Hoisted to TwoNodeClusterHarness.DeploySealTimeout (45s) so the reason is recorded once rather than as four unexplained numbers. DriverReconnectE2eTests is deliberately left alone: it seeds both ClusterNode rows itself and unconditionally, so its expected-ack set is identical before and after this change. Widening its timeout would be papering over a flake this change did not cause. Host.IntegrationTests 195/201; sole failure AbCip_Green_AgainstSim, verified failing on master. Claude-Session: https://claude.ai/code/session_01GASWkNEi68FSCtvr6rLoEW
85 lines
3.6 KiB
C#
85 lines
3.6 KiB
C#
using Microsoft.Extensions.DependencyInjection;
|
|
using Shouldly;
|
|
using Xunit;
|
|
using ZB.MOM.WW.OtOpcUa.Commons.Interfaces;
|
|
using ZB.MOM.WW.OtOpcUa.Commons.Messages.Admin;
|
|
|
|
namespace ZB.MOM.WW.OtOpcUa.Host.IntegrationTests;
|
|
|
|
/// <summary>
|
|
/// End-to-end <see cref="IFleetDiagnosticsClient"/> round-trip via the cluster:
|
|
/// admin node asks node-B's <c>DriverHostActor</c> for a snapshot via <c>ActorSelection</c>.
|
|
/// Verifies the cross-node Ask/Reply works and the snapshot reflects the target node's
|
|
/// view (NodeId + CurrentRevision after a deploy).
|
|
/// </summary>
|
|
public sealed class FleetDiagnosticsRoundTripTests
|
|
{
|
|
private static CancellationToken Ct => TestContext.Current.CancellationToken;
|
|
|
|
/// <summary>Verifies that get diagnostics returns a snapshot with the target node ID.</summary>
|
|
[Fact]
|
|
public async Task GetDiagnostics_returns_snapshot_with_target_NodeId()
|
|
{
|
|
await using var harness = await TwoNodeClusterHarness.StartAsync();
|
|
|
|
// Resolve target NodeId from the cluster — second member ordered by address.
|
|
var members = Akka.Cluster.Cluster.Get(harness.NodeASystem).State.Members
|
|
.OrderBy(m => m.Address.ToString())
|
|
.ToArray();
|
|
members.Length.ShouldBe(2);
|
|
var targetAddress = members[1].Address;
|
|
var targetNodeId = Commons.Types.NodeId.Parse($"{targetAddress.Host}:{targetAddress.Port}");
|
|
|
|
await using var scope = harness.NodeA.Services.CreateAsyncScope();
|
|
var client = scope.ServiceProvider.GetRequiredService<IFleetDiagnosticsClient>();
|
|
|
|
var snapshot = await client.GetDiagnosticsAsync(targetNodeId, Ct);
|
|
|
|
snapshot.NodeId.ShouldBe(targetNodeId);
|
|
snapshot.Drivers.ShouldBeEmpty(); // No driver children yet (F7).
|
|
snapshot.AsOfUtc.ShouldBeGreaterThan(DateTime.UtcNow.AddSeconds(-30));
|
|
}
|
|
|
|
/// <summary>Verifies that get diagnostics after deploy reports the current revision.</summary>
|
|
[Fact]
|
|
public async Task GetDiagnostics_after_deploy_reports_current_revision()
|
|
{
|
|
await using var harness = await TwoNodeClusterHarness.StartAsync();
|
|
await harness.SeedDefaultClusterAsync();
|
|
|
|
await using var scope = harness.NodeA.Services.CreateAsyncScope();
|
|
var adminOps = scope.ServiceProvider.GetRequiredService<IAdminOperationsClient>();
|
|
var diagnostics = scope.ServiceProvider.GetRequiredService<IFleetDiagnosticsClient>();
|
|
|
|
var deploy = await adminOps.StartDeploymentAsync(createdBy: "alice@test", Ct);
|
|
deploy.Outcome.ShouldBe(StartDeploymentOutcome.Accepted);
|
|
var expectedRev = deploy.RevisionHash!.Value;
|
|
|
|
// Wait until both DriverHostActors have caught up to the deployed revision.
|
|
var members = Akka.Cluster.Cluster.Get(harness.NodeASystem).State.Members
|
|
.OrderBy(m => m.Address.ToString())
|
|
.Select(m => Commons.Types.NodeId.Parse($"{m.Address.Host}:{m.Address.Port}"))
|
|
.ToArray();
|
|
|
|
foreach (var nodeId in members)
|
|
{
|
|
await WaitForAsync(async () =>
|
|
{
|
|
var snap = await diagnostics.GetDiagnosticsAsync(nodeId, Ct);
|
|
return snap.CurrentRevision == expectedRev;
|
|
}, TwoNodeClusterHarness.DeploySealTimeout);
|
|
}
|
|
}
|
|
|
|
private static async Task WaitForAsync(Func<Task<bool>> condition, TimeSpan timeout)
|
|
{
|
|
var deadline = DateTime.UtcNow + timeout;
|
|
while (DateTime.UtcNow < deadline)
|
|
{
|
|
if (await condition()) return;
|
|
await Task.Delay(200);
|
|
}
|
|
throw new TimeoutException($"Condition not met within {timeout}");
|
|
}
|
|
}
|