This is an automated email from the ASF dual-hosted git repository.
jason810496 pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/airflow.git
The following commit(s) were added to refs/heads/main by this push:
new 19cc828aed8 Add running_pod_log_lines config option to
KubernetesExecutor (#69301)
19cc828aed8 is described below
commit 19cc828aed8359057743d2bc9431f734a6bb2006
Author: Jason(Zhe-You) Liu <[email protected]>
AuthorDate: Mon Jul 20 16:43:18 2026 +0800
Add running_pod_log_lines config option to KubernetesExecutor (#69301)
* Resolve conflict in KubernetesExecutor.__init__ and drop redundant comment
The running_pod_log_lines config addition conflicted with the
pod-launch-failure requeue state added to __init__ after this branch
diverged; keep both initializations. Also drops a comment that only
restated the line below it, per review feedback.
* Clarify fallback reference in KubernetesExecutor running_pod_log_lines
self.RUNNING_POD_LOG_LINES looked like a self-referential fallback since
it isn't set on the instance until this assignment completes; referencing
the class attribute directly makes clear it falls back to the class default.
---
providers/cncf/kubernetes/provider.yaml | 9 +++++++++
.../cncf/kubernetes/executors/kubernetes_executor.py | 8 ++++++++
.../providers/cncf/kubernetes/get_provider_info.py | 7 +++++++
.../cncf/kubernetes/executors/test_kubernetes_executor.py | 15 +++++++++++++++
4 files changed, 39 insertions(+)
diff --git a/providers/cncf/kubernetes/provider.yaml
b/providers/cncf/kubernetes/provider.yaml
index 047f6b410c0..9e211647790 100644
--- a/providers/cncf/kubernetes/provider.yaml
+++ b/providers/cncf/kubernetes/provider.yaml
@@ -287,6 +287,15 @@ config:
type: boolean
example: ~
default: "False"
+ running_pod_log_lines:
+ description: |
+ Number of lines read from the end of a running task's pod log when
the task log
+ is served through the kube API, e.g. when viewing logs of a running
task in the UI.
+ The value must be greater than 0.
+ version_added: 10.20.0
+ type: integer
+ example: ~
+ default: "100"
pod_template_file:
description: |
Path to the YAML pod file that forms the basis for
KubernetesExecutor workers.
diff --git
a/providers/cncf/kubernetes/src/airflow/providers/cncf/kubernetes/executors/kubernetes_executor.py
b/providers/cncf/kubernetes/src/airflow/providers/cncf/kubernetes/executors/kubernetes_executor.py
index ba0355504b1..67616ea5989 100644
---
a/providers/cncf/kubernetes/src/airflow/providers/cncf/kubernetes/executors/kubernetes_executor.py
+++
b/providers/cncf/kubernetes/src/airflow/providers/cncf/kubernetes/executors/kubernetes_executor.py
@@ -154,6 +154,14 @@ class KubernetesExecutor(BaseExecutor):
# instead of requeuing. The orphaned task instance itself is still
recovered by the
# scheduler's adopt_or_reset_orphaned_tasks(), which re-queues it with
a fresh attempt.
self.pod_launch_attempts: dict[TaskInstanceKey, _PodLaunchAttempt] = {}
+ self.RUNNING_POD_LOG_LINES = self.conf.getint(
+ "kubernetes_executor", "running_pod_log_lines",
fallback=KubernetesExecutor.RUNNING_POD_LOG_LINES
+ )
+ if self.RUNNING_POD_LOG_LINES <= 0:
+ raise ValueError(
+ "The [kubernetes_executor] running_pod_log_lines configuration
must be greater than 0, "
+ f"got {self.RUNNING_POD_LOG_LINES}."
+ )
self.completed: dict[tuple[str, str], KubernetesResults] = {}
self.create_pods_after: datetime | None = None
diff --git
a/providers/cncf/kubernetes/src/airflow/providers/cncf/kubernetes/get_provider_info.py
b/providers/cncf/kubernetes/src/airflow/providers/cncf/kubernetes/get_provider_info.py
index 34c9857a315..df11e510921 100644
---
a/providers/cncf/kubernetes/src/airflow/providers/cncf/kubernetes/get_provider_info.py
+++
b/providers/cncf/kubernetes/src/airflow/providers/cncf/kubernetes/get_provider_info.py
@@ -158,6 +158,13 @@ def get_provider_info():
"example": None,
"default": "False",
},
+ "running_pod_log_lines": {
+ "description": "Number of lines read from the end of a
running task's pod log when the task log\nis served through the kube API, e.g.
when viewing logs of a running task in the UI.\nThe value must be greater than
0.\n",
+ "version_added": "10.20.0",
+ "type": "integer",
+ "example": None,
+ "default": "100",
+ },
"pod_template_file": {
"description": "Path to the YAML pod file that forms
the basis for KubernetesExecutor workers.\n",
"version_added": None,
diff --git
a/providers/cncf/kubernetes/tests/unit/cncf/kubernetes/executors/test_kubernetes_executor.py
b/providers/cncf/kubernetes/tests/unit/cncf/kubernetes/executors/test_kubernetes_executor.py
index e5716e30ffc..a71101dc555 100644
---
a/providers/cncf/kubernetes/tests/unit/cncf/kubernetes/executors/test_kubernetes_executor.py
+++
b/providers/cncf/kubernetes/tests/unit/cncf/kubernetes/executors/test_kubernetes_executor.py
@@ -641,6 +641,21 @@ class TestAirflowKubernetesScheduler:
assert kube_executor.RUNNING_POD_LOG_LINES == 100
assert kube_executor_2.RUNNING_POD_LOG_LINES == 200
+ @conf_vars({("kubernetes_executor", "running_pod_log_lines"): "500"})
+ def test_running_pod_log_lines_from_config(self):
+ kube_executor = KubernetesExecutor()
+
+ assert kube_executor.RUNNING_POD_LOG_LINES == 500
+ assert KubernetesExecutor.RUNNING_POD_LOG_LINES == 100
+
+ @pytest.mark.parametrize("invalid_value", ["0", "-1"])
+ def test_running_pod_log_lines_invalid_config(self, invalid_value):
+ with conf_vars({("kubernetes_executor", "running_pod_log_lines"):
invalid_value}):
+ with pytest.raises(
+ ValueError, match="running_pod_log_lines configuration must be
greater than 0"
+ ):
+ KubernetesExecutor()
+
class TestKubernetesExecutor:
"""
