fix(audit-log): explicit long CommandTimeout on partition-switch and per-channel purge maintenance paths

This commit is contained in:
Joseph Doherty
2026-07-09 06:43:23 -04:00
parent 1b6b0aab2a
commit b59aa2d717
11 changed files with 231 additions and 17 deletions
@@ -51,6 +51,11 @@ public class AuditLogPurgeActorTests : TestKit, IClassFixture<MsSqlMigrationFixt
public DateTime? ThrowOnBoundary { get; set; }
public Exception? BoundaryException { get; set; }
// Task 4: captures the maintenance command timeout the actor threads through to the
// switch-out + per-channel purge, so tests can assert the options value reaches the repo.
public List<TimeSpan?> SwitchTimeouts { get; } = new();
public List<TimeSpan?> ChannelTimeouts { get; } = new();
// M5.5 (T3): records every per-channel purge call as
// (channel, threshold, batchSize) so tests can assert which channels the
// actor chose to purge and with what window.
@@ -69,8 +74,9 @@ public class AuditLogPurgeActorTests : TestKit, IClassFixture<MsSqlMigrationFixt
AuditLogQueryFilter filter, AuditLogPaging paging, CancellationToken ct = default) =>
Task.FromResult<IReadOnlyList<AuditEvent>>(Array.Empty<AuditEvent>());
public Task<long> SwitchOutPartitionAsync(DateTime monthBoundary, CancellationToken ct = default)
public Task<long> SwitchOutPartitionAsync(DateTime monthBoundary, TimeSpan? commandTimeout = null, CancellationToken ct = default)
{
SwitchTimeouts.Add(commandTimeout);
if (ThrowOnBoundary.HasValue && monthBoundary == ThrowOnBoundary.Value)
{
throw BoundaryException ?? new InvalidOperationException("simulated switch failure");
@@ -87,9 +93,10 @@ public class AuditLogPurgeActorTests : TestKit, IClassFixture<MsSqlMigrationFixt
}
public Task<long> PurgeChannelOlderThanAsync(
string channel, DateTime threshold, int batchSize, CancellationToken ct = default)
string channel, DateTime threshold, int batchSize, TimeSpan? commandTimeout = null, CancellationToken ct = default)
{
ChannelPurges.Add((channel, threshold, batchSize));
ChannelTimeouts.Add(commandTimeout);
return Task.FromResult(RowsPerChannel(channel));
}
@@ -491,4 +498,49 @@ public class AuditLogPurgeActorTests : TestKit, IClassFixture<MsSqlMigrationFixt
Assert.Empty(repo.ChannelPurges);
}
// ---------------------------------------------------------------------
// 10. PurgeTick_Passes_MaintenanceCommandTimeout_To_Repository (Task 4)
// ---------------------------------------------------------------------
[Fact]
public void PurgeTick_Passes_MaintenanceCommandTimeout_To_Repository()
{
// The actor must thread AuditLogPurgeOptions.ResolvedMaintenanceCommandTimeout through to
// BOTH the partition switch-out and each per-channel purge, so the ~30s ADO.NET default
// cannot self-lock the drop-and-rebuild dance (arch-review 04, S2).
var boundary = new DateTime(2025, 5, 1, 0, 0, 0, DateTimeKind.Utc);
var repo = new RecordingRepo
{
Boundaries = new List<DateTime> { boundary },
RowsPerBoundary = _ => 7L,
};
var purgeOptions = FastTickOptions();
purgeOptions.MaintenanceCommandTimeoutMinutes = 45;
// A per-channel override shorter than the global window so the channel purge also fires.
var auditOptions = Options.Create(new AuditLogOptions
{
RetentionDays = 365,
PerChannelRetentionDays = new Dictionary<string, int> { ["ApiOutbound"] = 30 },
});
var sp = BuildScopedProvider(repo);
Sys.ActorOf(Props.Create(() => new AuditLogPurgeActor(
sp,
Options.Create(purgeOptions),
auditOptions,
NullLogger<AuditLogPurgeActor>.Instance)));
var expected = TimeSpan.FromMinutes(45);
AwaitAssert(
() =>
{
Assert.Contains(expected, repo.SwitchTimeouts);
Assert.Contains(expected, repo.ChannelTimeouts);
},
duration: TimeSpan.FromSeconds(3),
interval: TimeSpan.FromMilliseconds(50));
}
}