feat: dagster_runs_to_otlp + dagster_runs_to_splunk_hec sensors (v0.10.19)

Fills the gap in the dagster_runs_to_<destination>_sensor family — we
had Prometheus (Push Gateway / VM remote_write) + InfluxDB (line
protocol), but neither OTLP-direct nor Splunk-HEC-direct. Now all four
destinations are covered with the same @run_status_sensor pair shape
(SUCCESS + FAILURE).

dagster_runs_to_otlp_sensor:
- OTLP/HTTP wire (one wire, many backends — Splunk via Splunk OTC,
  Datadog OTel ingest, Honeycomb, Sumo, Loki via OTC, etc.)
- LogRecord per event: body + severityNumber (9 INFO / 17 ERROR) +
  attributes (dagster.run_id, dagster.job_name, dagster.status,
  dagster.duration_seconds, dagster.failure_step on failures)
- Resource attrs: service.name + service.instance.id
- Failures get failure_step + first 500 chars of error in attrs
- Parallels the existing dagster_failures_to_otlp_sensor (which is
  failures-only) — pick whichever fits the destination

dagster_runs_to_splunk_hec_sensor:
- Splunk-native HEC payload (single-hop direct, no OTel Collector)
- One HEC event per run with structured fields:
  dagster_run_id, dagster_job_name, dagster_status,
  dagster_duration_seconds, dagster_failure_step
- Splunk Web query patterns documented in README
  (p95 duration, failure rate by job, failure step distribution)
- Distinct from compute_log_managers/splunk/ (op stdout/stderr,
  instance layer) and audit_logs_to_splunk (audit DataFrames,
  scheduled)

Manifest: 834 → 836 components.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: eric-thomas-dagster

manifest.json

@@ -1,7 +1,7 @@
 {
   "version": "1.0.0",
   "repository": "https://github.com/eric-thomas-dagster/dagster-component-templates",
-  "last_updated": "2026-05-28",
+  "last_updated": "2026-06-04",
   "components": [
     {
       "id": "ab_controls",
@@ -4472,6 +4472,41 @@
         "last_validated": "2026-05-28"
       }
     },
+    {
+      "id": "dagster_runs_to_otlp_sensor",
+      "name": "Dagster Runs \u2192 OTLP Sensor",
+      "category": "sensor",
+      "description": "Push Dagster run lifecycle events (success + failure) as OTLP/HTTP log records to any OTel-compatible backend (Splunk via Splunk OTC, Datadog, Honeycomb, Sumo, Loki via OTC).",
+      "version": "1.0.0",
+      "author": "Dagster Community",
+      "path": "sensors/dagster_runs_to_otlp_sensor",
+      "tags": [
+        "sensor",
+        "telemetry",
+        "otlp",
+        "opentelemetry",
+        "honeycomb",
+        "datadog",
+        "splunk",
+        "sumo",
+        "observability"
+      ],
+      "vendor": "OpenTelemetry",
+      "dependencies": {
+        "pip": [
+          "requests>=2.28"
+        ]
+      },
+      "readme_url": "https://raw.githubusercontent.com/eric-thomas-dagster/dagster-component-templates/main/sensors/dagster_runs_to_otlp_sensor/README.md",
+      "component_url": "https://raw.githubusercontent.com/eric-thomas-dagster/dagster-component-templates/main/sensors/dagster_runs_to_otlp_sensor/component.py",
+      "schema_url": "https://raw.githubusercontent.com/eric-thomas-dagster/dagster-component-templates/main/sensors/dagster_runs_to_otlp_sensor/schema.json",
+      "example_url": "https://raw.githubusercontent.com/eric-thomas-dagster/dagster-component-templates/main/sensors/dagster_runs_to_otlp_sensor/example.yaml",
+      "requirements_url": "https://raw.githubusercontent.com/eric-thomas-dagster/dagster-component-templates/main/sensors/dagster_runs_to_otlp_sensor/requirements.txt",
+      "validation": {
+        "level": "code",
+        "last_validated": "2026-06-04"
+      }
+    },
     {
       "id": "dagster_runs_to_prometheus_sensor",
       "name": "Dagster Runs \u2192 Prometheus Sensor",
@@ -4504,6 +4539,37 @@
         "last_validated": "2026-05-28"
       }
     },
+    {
+      "id": "dagster_runs_to_splunk_hec_sensor",
+      "name": "Dagster Runs \u2192 Splunk HEC Sensor",
+      "category": "sensor",
+      "description": "Push Dagster run lifecycle events (success + failure) directly to Splunk via HTTP Event Collector (HEC). One hop, no OTel Collector required.",
+      "version": "1.0.0",
+      "author": "Dagster Community",
+      "path": "sensors/dagster_runs_to_splunk_hec_sensor",
+      "tags": [
+        "sensor",
+        "telemetry",
+        "splunk",
+        "hec",
+        "observability"
+      ],
+      "vendor": "Splunk",
+      "dependencies": {
+        "pip": [
+          "requests>=2.28"
+        ]
+      },
+      "readme_url": "https://raw.githubusercontent.com/eric-thomas-dagster/dagster-component-templates/main/sensors/dagster_runs_to_splunk_hec_sensor/README.md",
+      "component_url": "https://raw.githubusercontent.com/eric-thomas-dagster/dagster-component-templates/main/sensors/dagster_runs_to_splunk_hec_sensor/component.py",
+      "schema_url": "https://raw.githubusercontent.com/eric-thomas-dagster/dagster-component-templates/main/sensors/dagster_runs_to_splunk_hec_sensor/schema.json",
+      "example_url": "https://raw.githubusercontent.com/eric-thomas-dagster/dagster-component-templates/main/sensors/dagster_runs_to_splunk_hec_sensor/example.yaml",
+      "requirements_url": "https://raw.githubusercontent.com/eric-thomas-dagster/dagster-component-templates/main/sensors/dagster_runs_to_splunk_hec_sensor/requirements.txt",
+      "validation": {
+        "level": "code",
+        "last_validated": "2026-06-04"
+      }
+    },
     {
       "id": "dagstermill_notebook_asset",
       "name": "DagstermillNotebookAsset",
@@ -24453,5 +24519,5 @@
   ],
   "schema_spec": "https://raw.githubusercontent.com/eric-thomas-dagster/dagster-component-templates/main/schema-spec.json",
   "schema_spec_local": "schema-spec.json",
-  "total": 834
+  "total": 836
 }

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "dagster-community-components"
-version = "0.10.18"
+version = "0.10.19"
 description = "Community-maintained Dagster components — ingestion, transforms, IO managers, sensors, sinks, resources, and more."
 readme = "README.md"
 requires-python = ">=3.10"

sensors/dagster_runs_to_otlp_sensor/README.md

@@ -0,0 +1,132 @@
+# dagster_runs_to_otlp_sensor
+
+Push **Dagster run lifecycle events** (success + failure) as OTLP/HTTP log records to any OpenTelemetry-compatible backend — Splunk via Splunk OTel Collector, Datadog OTel ingest, Honeycomb, Sumo Logic OTel, Grafana Loki via OTel Collector, etc.
+
+Operational telemetry — distinct from the data-plane `dataframe_to_otlp_logs` sink. Parallels the existing `dagster_runs_to_prometheus_sensor` (Push Gateway / VM) and `dagster_runs_to_influxdb_sensor` — same `@run_status_sensor` pair (SUCCESS + FAILURE), different wire protocol.
+
+For failure-only OTLP shipment with richer error context, see `dagster_failures_to_otlp_sensor` — both can run side-by-side.
+
+## What lands at the backend per run-status event
+
+One OTLP `LogRecord` per run completion:
+
+| Field | Value |
+|---|---|
+| `body.stringValue` | `"Dagster run <run_id> for job <job_name> success (Xs)"` (or `failure step=… err=…`) |
+| `severityNumber` | 9 (INFO) for success, 17 (ERROR) for failure |
+| `severityText` | `INFO` / `ERROR` |
+| `timeUnixNano` | run end_time |
+| `attributes.dagster.run_id` | run UUID |
+| `attributes.dagster.job_name` | job name |
+| `attributes.dagster.status` | `success` / `failure` |
+| `attributes.dagster.duration_seconds` | double — wall-clock run time |
+| `attributes.dagster.failure_step` | step_key (failure only) |
+| `resource.service.name` | `dagster` |
+| `resource.service.instance.id` | `location_label` (deployment / region) |
+
+[//]: # (FIELDS:START - auto-generated by tools/regen_readme_fields.py)
+
+## Fields
+
+### Required
+
+| Field | Type | Description |
+|---|---|---|
+| `sensor_name` | `str` | Dagster sensor name. |
+| `otlp_endpoint_env_var` | `str` | Env var with OTLP/HTTP endpoint base URL (e.g. https://api.honeycomb.io). The /v1/logs path is appended. |
+
+### Sensor configuration
+
+| Field | Type | Default | Description |
+|---|---|---|---|
+| `default_status` | `str` | `"STOPPED"` | STOPPED \| RUNNING |
+
+### Other
+
+| Field | Type | Default | Description |
+|---|---|---|---|
+| `otlp_headers_env_var` | `str` | — | Env var with comma-separated HTTP headers (OTel convention: `x-honeycomb-team=<key>,x-honeycomb-dataset=dagster`). |
+| `service_name` | `str` | `"dagster"` | OTLP resource attribute `service.name`. |
+| `location_label` | `str` | `"default"` | OTLP resource attribute `service.instance.id` — usually deployment / region. |
+| `monitor_jobs` | `List[str]` | — | If set, only emit for these Dagster job names. None = all. |
+| `request_timeout_seconds` | `int` | `15` | — |
+
+[//]: # (FIELDS:END)
+
+## Example — Honeycomb
+
+```yaml
+type: dagster_community_components.DagsterRunsToOtlpSensorComponent
+attributes:
+  sensor_name: dagster_runs_to_honeycomb
+  otlp_endpoint_env_var: HONEYCOMB_OTLP_ENDPOINT
+  otlp_headers_env_var: HONEYCOMB_OTLP_HEADERS
+  service_name: dagster
+  location_label: prod-east
+```
+
+Required env:
+
+```
+HONEYCOMB_OTLP_ENDPOINT=https://api.honeycomb.io
+HONEYCOMB_OTLP_HEADERS=x-honeycomb-team=<api_key>,x-honeycomb-dataset=dagster_runs
+```
+
+## Example — Datadog OTel ingest
+
+```yaml
+type: dagster_community_components.DagsterRunsToOtlpSensorComponent
+attributes:
+  sensor_name: dagster_runs_to_datadog
+  otlp_endpoint_env_var: DD_OTLP_ENDPOINT
+  otlp_headers_env_var: DD_OTLP_HEADERS
+  service_name: dagster
+```
+
+Required env:
+
+```
+DD_OTLP_ENDPOINT=https://otlp-intake.logs.datadoghq.com
+DD_OTLP_HEADERS=DD-API-KEY=<api_key>
+```
+
+## Example — Splunk via Splunk OTel Collector
+
+```yaml
+type: dagster_community_components.DagsterRunsToOtlpSensorComponent
+attributes:
+  sensor_name: dagster_runs_to_splunk_otc
+  otlp_endpoint_env_var: SPLUNK_OTC_ENDPOINT
+  service_name: dagster
+```
+
+Required env:
+
+```
+SPLUNK_OTC_ENDPOINT=http://splunk-otc.svc:4318
+```
+
+For Splunk HEC direct (no OTel Collector), use `dagster_runs_to_splunk_hec_sensor` — same shape, native HEC payload.
+
+## Why two sensors (SUCCESS + FAILURE)?
+
+Dagster's `@run_status_sensor` decorator filters on a single `run_status`. The component wires up two — one for SUCCESS, one for FAILURE — so both event types push their own LogRecord with the right `severityNumber` (9 vs 17) and `dagster.status` attribute.
+
+## Relationship to other telemetry sensors
+
+| Sensor | Scope | Wire |
+|---|---|---|
+| **`dagster_runs_to_otlp_sensor`** (this one) | success + failure | OTLP/HTTP |
+| `dagster_runs_to_prometheus_sensor` | success + failure | Prometheus Push Gateway / VM remote_write |
+| `dagster_runs_to_influxdb_sensor` | success + failure | InfluxDB line protocol |
+| `dagster_runs_to_splunk_hec_sensor` | success + failure | Splunk HEC direct |
+| `dagster_failures_to_otlp_sensor` | failure only (richer error context) | OTLP/HTTP |
+| `dagster_materializations_to_prometheus_sensor` | asset materialization events | Prometheus Push Gateway / VM |
+
+Use whichever combination matches your observability stack. They're independent and can run side-by-side.
+
+## See also
+
+- [OTLP/HTTP logs spec](https://opentelemetry.io/docs/specs/otlp/#otlphttp-logs-binary-protocol)
+- `dataframe_to_otlp_logs` (in `assets/sinks/`) — **distinct** — data-plane DataFrame sink, not run events
+- `compute_log_managers/otlp/` — **distinct** — instance-level CLM shipping op stdout/stderr (not run events)

sensors/dagster_runs_to_otlp_sensor/__init__.py

@@ -0,0 +1,3 @@
+from .component import DagsterRunsToOtlpSensorComponent
+
+__all__ = ["DagsterRunsToOtlpSensorComponent"]