This is an automated email from the ASF dual-hosted git repository.
shahar pushed a commit to branch v3-1-test
in repository https://gitbox.apache.org/repos/asf/airflow.git
The following commit(s) were added to refs/heads/v3-1-test by this push:
new e5d67c4b1dd [v3-1-test] Fix flaky OTel integration test with DNS
health check (#61070) (#61242) (#61286)
e5d67c4b1dd is described below
commit e5d67c4b1dd0070a762801f43fe2e142ded1848e
Author: github-actions[bot]
<41898282+github-actions[bot]@users.noreply.github.com>
AuthorDate: Sat Jan 31 16:49:48 2026 +0200
[v3-1-test] Fix flaky OTel integration test with DNS health check (#61070)
(#61242) (#61286)
* Fix flaky OTel integration test with DNS health check (#61070)
* Update airflow-core/tests/integration/otel/test_otel.py
---------
(cherry picked from commit 8ac25dd658e9ea6a29aaa85215e17c180765c821)
Co-authored-by: Abhishek Mishra <[email protected]>
Co-authored-by: Henry Chen <[email protected]>
---
airflow-core/tests/integration/otel/test_otel.py | 51 +++++++++++++++++++++++-
1 file changed, 49 insertions(+), 2 deletions(-)
diff --git a/airflow-core/tests/integration/otel/test_otel.py
b/airflow-core/tests/integration/otel/test_otel.py
index e218820b292..51f70c51291 100644
--- a/airflow-core/tests/integration/otel/test_otel.py
+++ b/airflow-core/tests/integration/otel/test_otel.py
@@ -20,6 +20,7 @@ import json
import logging
import os
import signal
+import socket
import subprocess
import time
@@ -52,6 +53,44 @@ from tests_common.test_utils.version_compat import
AIRFLOW_V_3_0_PLUS, AIRFLOW_V
log = logging.getLogger("integration.otel.test_otel")
+def wait_for_otel_collector(host: str, port: int, timeout: int = 120) -> None:
+ """
+ Wait for the OTel collector to be reachable before running tests.
+
+ This prevents flaky test failures caused by transient DNS resolution issues
+ (e.g., 'Temporary failure in name resolution' for breeze-otel-collector).
+
+ Note: If the collector is not reachable after timeout, logs a warning but
+ does not fail - allows tests to run and fail naturally if needed.
+ """
+ deadline = time.monotonic() + timeout
+ last_error = None
+ while time.monotonic() < deadline:
+ try:
+ # Test DNS resolution and TCP connectivity
+ with socket.create_connection((host, port), timeout=5):
+ pass
+ log.info("OTel collector at %s:%d is reachable.", host, port)
+ return
+ except (socket.gaierror, TimeoutError, OSError) as e:
+ last_error = e
+ log.debug(
+ "OTel collector at %s:%d not reachable: %s. Retrying...",
+ host,
+ port,
+ e,
+ )
+ time.sleep(2)
+ log.warning(
+ "OTel collector at %s:%d is not reachable after %ds. Last error: %s. "
+ "Tests will proceed but may fail if collector is required.",
+ host,
+ port,
+ timeout,
+ last_error,
+ )
+
+
def unpause_trigger_dag_and_get_run_id(dag_id: str) -> str:
unpause_command = ["airflow", "dags", "unpause", dag_id]
@@ -611,9 +650,17 @@ class TestOtelIntegration:
@classmethod
def setup_class(cls):
+ otel_host = "breeze-otel-collector"
+ otel_port = 4318
+
+ # Wait for OTel collector to be reachable before running tests.
+ # This prevents flaky test failures caused by transient DNS resolution
issues
+ # during scheduler handoff (see
https://github.com/apache/airflow/issues/61070).
+ wait_for_otel_collector(otel_host, otel_port)
+
os.environ["AIRFLOW__TRACES__OTEL_ON"] = "True"
- os.environ["AIRFLOW__TRACES__OTEL_HOST"] = "breeze-otel-collector"
- os.environ["AIRFLOW__TRACES__OTEL_PORT"] = "4318"
+ os.environ["AIRFLOW__TRACES__OTEL_HOST"] = otel_host
+ os.environ["AIRFLOW__TRACES__OTEL_PORT"] = str(otel_port)
if cls.use_otel != "true":
os.environ["AIRFLOW__TRACES__OTEL_DEBUGGING_ON"] = "True"
