fix(usage): stop 429s with an activity-dependent poll cadence + manual refresh

The usage monitor polled the undocumented OAuth usage endpoint every 60s and
earned 429s. It now polls every 5 min while any task is Running and every
15 min while idle (usage_poll_interval_active_seconds / _idle_seconds, both
clamped to >= 60; the old single usage_poll_interval_seconds key is gone).

A 429 comes back as UsageRateLimitedException carrying Retry-After and adds
exponential backoff on top, capped at 30 min and never shorter than the normal
cadence; the strike count resets on the first success. The schedule arithmetic
is the pure static UsagePollSchedule.NextDelay.

Since the idle cadence is slow on purpose, WorkerHub.RefreshUsage drives
UsageMonitorService.RefreshNowAsync behind a Refresh now button in the Usage
Monitor modal: an out-of-band poll that pushes the loop's next-due time out so
no double poll follows, with a 10s cooldown so click-spam can't earn a 429.

Staleness now measures against the slower (idle) interval so an idle worker
isn't flagged stale just for not polling.
This commit is contained in:
mika kuns
2026-08-05 16:40:34 +02:00
parent f6cb8250bb
commit 2700c3d817
24 changed files with 615 additions and 50 deletions
+16 -1
View File
@@ -189,6 +189,7 @@ public sealed class WorkerHub : Microsoft.AspNetCore.SignalR.Hub
private readonly Data.Git.GitService? _git;
private readonly UsageSnapshotBuilder? _usageSnapshotBuilder;
private readonly ITranscriptUsageReader? _usageReader;
private readonly UsageMonitorService? _usageMonitor;
public WorkerHub(
QueueService queue,
@@ -220,7 +221,8 @@ public sealed class WorkerHub : Microsoft.AspNetCore.SignalR.Hub
WorktreeManager? worktreeManager = null,
Data.Git.GitService? git = null,
UsageSnapshotBuilder? usageSnapshotBuilder = null,
ITranscriptUsageReader? usageReader = null)
ITranscriptUsageReader? usageReader = null,
UsageMonitorService? usageMonitor = null)
{
_queue = queue;
_waker = waker;
@@ -252,6 +254,7 @@ public sealed class WorkerHub : Microsoft.AspNetCore.SignalR.Hub
_git = git;
_usageSnapshotBuilder = usageSnapshotBuilder;
_usageReader = usageReader;
_usageMonitor = usageMonitor;
}
// Persistence boundary for the session_skills JSON-array columns (task/list/global).
@@ -1049,6 +1052,18 @@ public sealed class WorkerHub : Microsoft.AspNetCore.SignalR.Hub
return _usageSnapshotBuilder.BuildAsync(Context.ConnectionAborted);
});
/// <summary>
/// Manual "refresh now" for the usage monitor. Polls the endpoint out of band and returns the
/// fresh snapshot; a refresh inside the monitor's cooldown reuses the last poll's result
/// instead of risking a 429.
/// </summary>
public Task<UsageSnapshotDto> RefreshUsage() => HubGuard(() =>
{
if (_usageMonitor is null)
throw new InvalidOperationException("Usage monitor is not configured.");
return _usageMonitor.RefreshNowAsync(Context.ConnectionAborted);
});
public Task<IReadOnlyList<ModelUsageRowDto>> GetModelUsage(DateOnly from, DateOnly to) => HubGuard(async () =>
{
if (_usageReader is null)