Add client-side exec-server RPC attempt metrics (#42883)

## What changed

- Record `exec_server_client_requests_total` for every client RPC call attempt,
  labeled by protocol method.
- Count attempts before local admission so rejected, timed-out, cancelled, and
  transport-failed calls are included, while notifications and responses are not.
- Export the counter to configured OpenTelemetry collectors while excluding it
  from the built-in Statsig metrics set.

## Testing

- Cover every RPC call entry point, successful responses, local and transport
  failures, timeouts, cancellation, notifications, and disabled metrics.
- Verify that the OTLP HTTP exporter retains the counter.

GitOrigin-RevId: d266e840d5871aa2c34bed8ce44f3345e2b4650f
This commit is contained in:
bkotsopoulos
2026-09-04 23:06:00 +00:00
committed by copyberry
parent a97cf1b72e
commit 574a36ff99
8 changed files with 310 additions and 0 deletions
+3
View File
@@ -2,6 +2,7 @@ use crate::config::OtelExporter;
use crate::metrics::Result;
use crate::metrics::names::API_CALL_COUNT_METRIC;
use crate::metrics::names::API_CALL_DURATION_METRIC;
use crate::metrics::names::EXEC_SERVER_CLIENT_REQUEST_COUNT_METRIC;
use crate::metrics::names::RESPONSES_API_ENGINE_IAPI_TTFT_DURATION_METRIC;
use crate::metrics::names::RESPONSES_API_ENGINE_SERVICE_TBT_DURATION_METRIC;
use crate::metrics::names::RESPONSES_API_ENGINE_SERVICE_TTFT_DURATION_METRIC;
@@ -23,6 +24,8 @@ const STATSIG_DISABLED_METRICS: &[&str] = &[
API_CALL_COUNT_METRIC,
API_CALL_DURATION_METRIC,
CONVERSATION_TURN_COUNT_METRIC,
// Caller-side executor volume belongs in configured observability collectors.
EXEC_SERVER_CLIENT_REQUEST_COUNT_METRIC,
RESPONSES_API_ENGINE_IAPI_TTFT_DURATION_METRIC,
RESPONSES_API_ENGINE_SERVICE_TBT_DURATION_METRIC,
RESPONSES_API_ENGINE_SERVICE_TTFT_DURATION_METRIC,
+2
View File
@@ -5,6 +5,8 @@ pub const ARTIFACT_OPERATION_STARTED_METRIC: &str = "codex.artifact.operation.st
pub const ARTIFACT_OPERATION_EXPECTED_OUTPUT_COUNT_METRIC: &str =
"codex.artifact.operation.expected_output_count";
pub const PROCESS_START_METRIC: &str = "codex.process.start";
/// Caller-side exec-server RPC attempts, including local admission and transport failures.
pub const EXEC_SERVER_CLIENT_REQUEST_COUNT_METRIC: &str = "exec_server_client_requests_total";
pub const API_CALL_COUNT_METRIC: &str = "codex.api_request";
pub const API_CALL_DURATION_METRIC: &str = "codex.api_request.duration_ms";
pub const SSE_EVENT_COUNT_METRIC: &str = "codex.sse_event";
+7
View File
@@ -605,6 +605,7 @@ mod tests {
use super::*;
use crate::metrics::API_CALL_COUNT_METRIC;
use crate::metrics::API_CALL_DURATION_METRIC;
use crate::metrics::EXEC_SERVER_CLIENT_REQUEST_COUNT_METRIC;
use crate::metrics::MetricsExporter;
use crate::metrics::RESPONSES_API_ENGINE_IAPI_TTFT_DURATION_METRIC;
use crate::metrics::RESPONSES_API_ENGINE_SERVICE_TBT_DURATION_METRIC;
@@ -731,6 +732,12 @@ mod tests {
metrics.counter(API_CALL_COUNT_METRIC, /*inc*/ 1, &[])?;
metrics.record_duration(API_CALL_DURATION_METRIC, Duration::from_millis(100), &[])?;
metrics.counter_with_description(
EXEC_SERVER_CLIENT_REQUEST_COUNT_METRIC,
"Client-side exec-server RPC attempts.",
/*inc*/ 1,
&[("method", "fs/readFile")],
)?;
metrics.counter("codex.conversation.turn.count", /*inc*/ 1, &[])?;
metrics.record_duration(
RESPONSES_API_ENGINE_IAPI_TTFT_DURATION_METRIC,