[router] Count requests/responses at the HTTP edge for true intake (#29467)
This commit is contained in:
@@ -20,10 +20,11 @@ The dashboard graphs every family the router emits:
|
||||
|
||||
| Metric | Type | What it shows |
|
||||
|---|---|---|
|
||||
| `sgl_router_requests_total` | Counter | Dispatches by `worker_url`, `model_id`, `mode`, `outcome` |
|
||||
| `sgl_router_requests_total` | Counter | **Edge intake** — every request received at the router HTTP boundary, by `route`, `method`, counted before worker dispatch (true intake) |
|
||||
| `sgl_router_responses_total` | Counter | **Edge responses** — every response returned, by `route`, `method`, `status_code` (incl. early-exit 400/413/503). `requests_total - responses_total` = received-but-not-answered |
|
||||
| `sgl_router_worker_requests_total` | Counter | Per-worker **dispatches** by `worker_url`, `model_id`, `mode`, `outcome` (recorded after dispatch; blind to pre-dispatch drops) |
|
||||
| `sgl_router_request_duration_seconds` | Histogram | End-to-end request latency by `model_id` |
|
||||
| `sgl_router_ttft_seconds` | Histogram | Time to first token (streaming) by `model_id` |
|
||||
| `sgl_router_responses_total` | Counter | Client-visible HTTP `status_code` |
|
||||
| `sgl_router_overlap_blocks` | Histogram | Cache-aware-zmq overlap blocks by `model_id` |
|
||||
| `sgl_router_active_load` | Gauge | Per-worker prefill-token / decode-block load |
|
||||
| `sgl_router_workers` | Gauge | Registered worker count by `mode` |
|
||||
|
||||
Reference in New Issue
Block a user