| @@ -37,6 +37,18 @@ join the same traces via traceparent). · 36 unmodified lines |
| 37 | `mcp.tool.integration`, same outcome attrs. | 37 | `mcp.tool.integration`, same outcome attrs. |
| 38 | - `plugin.openapi.invoke` — `plugin.openapi.method` / `path_template` / | 38 | - `plugin.openapi.invoke` — `plugin.openapi.method` / `path_template` / |
| 39 | `base_url`, and since PR #992 `http.status_code`. | 39 | `base_url`, and since PR #992 `http.status_code`. |
| | 40 | - `executor.tools.list` — `executor.tools.filter.integration` / `.owner` / |
| | 41 | `.connection` (present only when the read was filtered), |
| | 42 | `executor.tools.result_count`, and the catalog-refresh counters |
| | 43 | `executor.tools.sync.candidates` (rows the stale scan returned) / |
| | 44 | `executor.tools.sync.synced`. A slow tools read is almost always |
| | 45 | `candidates` > 0: subtract the child span durations to confirm. |
| | 46 | - `executor.tools.sync` (child of the above, one per refreshed connection; |
| | 47 | older spans carry the previous name `executor.tools.sync_stale` and no |
| | 48 | trigger/outcome attrs) — `executor.integration`, |
| | 49 | `executor.connection`, `executor.tools.sync.trigger` |
| | 50 | (`stale_marked`/`config_revised`/`expired`) and |
| | 51 | `executor.tools.sync.outcome` (`ok`/`fail`). |
| 40 | - `mcp.request` (outer) — `mcp.auth.organization_id`, | 52 | - `mcp.request` (outer) — `mcp.auth.organization_id`, |
| 41 | `mcp.auth.account_id`, `mcp.tool.name`, CF edge fields (`cf.country`…), | 53 | `mcp.auth.account_id`, `mcp.tool.name`, CF edge fields (`cf.country`…), |
| 42 | MCP client fingerprint (`mcp.client.name`…). | 54 | MCP client fingerprint (`mcp.client.name`…). |