Refactor OpenTelemetry environment variable handling to use ray_constants (#57910)

1. **Remove direct environment variable access patterns**
- Replace all instances of `os.getenv("RAY_enable_open_telemetry") ==
"1"`
- Standardize to use `ray_constants.RAY_ENABLE_OPEN_TELEMETRY`
consistently throughout the codebase

2. **Unify default value format for RAY_enable_open_telemetry**
   - Standardize the default value to `"true"` | `"false"` 
- Previously, the codebase had mixed usage of `"1"` and `"true"`, which
is now unified

3. **Backward compatibility maintained**
- Carefully verified that the existing `RAY_ENABLE_OPEN_TELEMETRY`
constant properly handles both `"1"` and `"true"` values
   - This change will not introduce any breaking behavior
   - The `env_bool` helper function already supports both formats:
```python
RAY_ENABLE_OPEN_TELEMETRY = env_bool("RAY_enable_open_telemetry", False)
def env_bool(key, default):
    if key in os.environ:
        return (
            True
            if os.environ[key].lower() == "true" or os.environ[key] == "1"
            else False
        )
    return default
```

---
Most of the current code uses: `RAY_enable_open_telemetry: "1"`

A smaller portion (not zero) uses: `RAY_enable_open_telemetry: "true"`

https://github.com/ray-project/ray/blob/fe7ad00f9720a722fde5fecba5bb681234bcdb63/python/ray/tests/test_metrics_agent.py#L497

My personal preference is "true"—it’s concise and unambiguous. If it’s
"1", I have to think/guess whether it means "true" or "false".

---------

Signed-off-by: justwph <[email protected]>

Author: wph95

python/ray/dashboard/modules/reporter/tests/test_reporter.py

@@ -204,9 +204,9 @@ def enable_open_telemetry(request):
     Fixture to enable OpenTelemetry for the test.
     """
     if request.param:
-        os.environ["RAY_enable_open_telemetry"] = "1"
+        os.environ["RAY_enable_open_telemetry"] = "true"
     else:
-        os.environ["RAY_enable_open_telemetry"] = "0"
+        os.environ["RAY_enable_open_telemetry"] = "false"
     yield
     os.environ.pop("RAY_enable_open_telemetry", None)
 

python/ray/tests/BUILD.bazel

@@ -85,7 +85,7 @@ py_test_module_list(
 py_test_module_list(
     size = "medium",
     env = {
-        "RAY_enable_open_telemetry": "1",
+        "RAY_enable_open_telemetry": "true",
     },
     files = [
         "test_metric_cardinality.py",

python/ray/tests/test_metric_cardinality.py

@@ -18,6 +18,7 @@
     TASK_OR_ACTOR_NAME_TAG_KEY,
 )
 
+from ray._private.ray_constants import RAY_ENABLE_OPEN_TELEMETRY
 
 try:
     import prometheus_client
@@ -41,7 +42,7 @@ def _setup_cluster_for_test(request, ray_start_cluster):
             "metrics_report_interval_ms": 1000,
             "enable_metrics_collection": True,
             "metric_cardinality_level": core_metric_cardinality_level,
-            "enable_open_telemetry": os.getenv("RAY_enable_open_telemetry") == "1",
+            "enable_open_telemetry": RAY_ENABLE_OPEN_TELEMETRY,
         }
     )
     cluster.wait_for_nodes()
@@ -151,7 +152,7 @@ def test_cardinality_recommended_and_legacy_levels(
 # implementation doesn't support low cardinality.
 @pytest.mark.skipif(prometheus_client is None, reason="Prometheus not installed")
 @pytest.mark.skipif(
-    os.getenv("RAY_enable_open_telemetry") != "1",
+    not RAY_ENABLE_OPEN_TELEMETRY,
     reason="OpenTelemetry is not enabled",
 )
 @pytest.mark.parametrize(

python/ray/tests/test_metrics_agent.py

@@ -22,7 +22,10 @@
     Gauge as MetricsAgentGauge,
     PrometheusServiceDiscoveryWriter,
 )
-from ray._private.ray_constants import PROMETHEUS_SERVICE_DISCOVERY_FILE
+from ray._private.ray_constants import (
+    PROMETHEUS_SERVICE_DISCOVERY_FILE,
+    RAY_ENABLE_OPEN_TELEMETRY,
+)
 from ray._private.test_utils import (
     PrometheusTimeseries,
     fetch_prometheus_metric_timeseries,
@@ -204,7 +207,7 @@ def _setup_cluster_for_test(request, ray_start_cluster):
             "event_stats_print_interval_ms": 500,
             "event_stats": True,
             "enable_metrics_collection": enable_metrics_collection,
-            "enable_open_telemetry": os.getenv("RAY_enable_open_telemetry") == "1",
+            "enable_open_telemetry": RAY_ENABLE_OPEN_TELEMETRY,
         }
     )
     # Add worker nodes.
@@ -320,7 +323,7 @@ def test_cases():
                 "test_driver_counter_total",
                 "test_gauge",
             ]
-            if os.environ.get("RAY_enable_open_telemetry") != "1"
+            if not RAY_ENABLE_OPEN_TELEMETRY
             else [
                 "test_counter_total",
                 "test_driver_counter_total",
@@ -744,7 +747,7 @@ def wrap_test_case_for_retry():
 
 @pytest.mark.skipif(sys.platform == "win32", reason="Not working in Windows.")
 @pytest.mark.skipif(
-    os.environ.get("RAY_enable_open_telemetry") == "1",
+    RAY_ENABLE_OPEN_TELEMETRY,
     reason="OpenTelemetry backend does not support Counter exported as gauge.",
 )
 def test_counter_exported_as_gauge(shutdown_only):

python/ray/util/metrics.py

@@ -1,9 +1,9 @@
 import logging
-import os
 import re
 import warnings
 from typing import Any, Dict, List, Optional, Tuple, Union
 
+from ray._private.ray_constants import env_bool
 from ray._raylet import (
     Count as CythonCount,
     Gauge as CythonGauge,
@@ -198,7 +198,7 @@ def __init__(
         if self._discard_metric:
             self._metric = None
         else:
-            if os.environ.get("RAY_enable_open_telemetry") == "1":
+            if env_bool("RAY_enable_open_telemetry", False):
                 """
                 For the new opentelemetry implementation, we'll correctly use Counter
                 rather than Sum.

src/ray/stats/tests/BUILD.bazel

@@ -5,7 +5,7 @@ ray_cc_test(
     size = "small",
     srcs = ["metric_with_open_telemetry_test.cc"],
     env = {
-        "RAY_enable_open_telemetry": "1",
+        "RAY_enable_open_telemetry": "true",
     },
     tags = ["team:core"],
     deps = [