Ilker ManapandClaude Opus 5.5
f8572fb5ee
"Why was this request slow?" had no answer. Nothing tied a request to the replica that served it, the time admission held it went out in a response header and was never kept, and the only per-request log lines were for disconnects, early stream ends and upstream retries. Every request now has an id: the client's X-Request-Id when it is a plain string of reasonable length, otherwise one the gateway makes. It is set on the response before anything else, so a request refused as malformed can be named too, and forwarded to the engine. One access log line per request carries it with the model asked for and served, the replica, status, time in the queue, time to the first token, total time, tokens, and the key, project and user. Usage records gain the id, the replica, queue time and time to first token, tried against a restored copy of the live store before being shipped: 591 rows kept, columns and index in place. `llmgw ctl request ID` reads one back through the admin API. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
14 lines
819 B
SQL
14 lines
819 B
SQL
-- What a request went through, beside what it cost: which replica answered,
|
|
-- how long it queued in admission, and how long the first token took. Without
|
|
-- them "why was my request slow" had no answer — the queue time went out in a
|
|
-- response header and nowhere else.
|
|
--
|
|
-- request_id is the id the client sent or the gateway made, echoed in the
|
|
-- X-Request-Id header and the access log, so one request can be followed from
|
|
-- the client's error to the replica that served it.
|
|
ALTER TABLE usage ADD COLUMN request_id TEXT NOT NULL DEFAULT '';
|
|
ALTER TABLE usage ADD COLUMN replica TEXT NOT NULL DEFAULT '';
|
|
ALTER TABLE usage ADD COLUMN queue_ms INTEGER NOT NULL DEFAULT 0;
|
|
ALTER TABLE usage ADD COLUMN ttft_ms INTEGER NOT NULL DEFAULT 0;
|
|
CREATE INDEX usage_by_request ON usage(request_id) WHERE request_id <> '';
|