This is an automated email from the ASF dual-hosted git repository.
potiuk 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 79991cd4db0 Fix id_to_job_info memory leak in the AWS Batch executor
(#72186)
79991cd4db0 is described below
commit 79991cd4db0c9346a28b23c453377f6df0c6b4ed
Author: PoAn Yang <[email protected]>
AuthorDate: Sun Sep 20 22:30:01 2026 +0800
Fix id_to_job_info memory leak in the AWS Batch executor (#72186)
Signed-off-by: PoAn Yang <[email protected]>
---
.../amazon/src/airflow/providers/amazon/aws/executors/batch/utils.py | 1 +
providers/amazon/tests/unit/amazon/aws/executors/batch/test_utils.py | 4 +---
2 files changed, 2 insertions(+), 3 deletions(-)
diff --git
a/providers/amazon/src/airflow/providers/amazon/aws/executors/batch/utils.py
b/providers/amazon/src/airflow/providers/amazon/aws/executors/batch/utils.py
index d47fc93ff5b..4fc127fca86 100644
--- a/providers/amazon/src/airflow/providers/amazon/aws/executors/batch/utils.py
+++ b/providers/amazon/src/airflow/providers/amazon/aws/executors/batch/utils.py
@@ -126,6 +126,7 @@ class BatchJobCollection:
del self.key_to_id[workload_key]
del self.id_to_key[job_id]
del self.id_to_failure_counts[job_id]
+ del self.id_to_job_info[job_id]
return workload_key
def remove_job(self, job_id: str) -> BatchJobWorkloadKey | None:
diff --git
a/providers/amazon/tests/unit/amazon/aws/executors/batch/test_utils.py
b/providers/amazon/tests/unit/amazon/aws/executors/batch/test_utils.py
index 3ee8dca2628..0bf8fc35c91 100644
--- a/providers/amazon/tests/unit/amazon/aws/executors/batch/test_utils.py
+++ b/providers/amazon/tests/unit/amazon/aws/executors/batch/test_utils.py
@@ -206,9 +206,7 @@ class TestBatchJobCollection:
assert len(self.collection) == 0
assert self.job_id1 not in self.collection.id_to_key
assert self.key1 not in self.collection.key_to_id
- # id_to_job_info is NOT removed by pop_by_id in the current
implementation.
- assert self.job_id1 in self.collection.id_to_job_info
- assert self.collection.id_to_job_info[self.job_id1].cmd == self.cmd1
+ assert self.job_id1 not in self.collection.id_to_job_info
# id_to_failure_counts is a defaultdict, so accessing a removed key
returns 0.
assert self.collection.id_to_failure_counts[self.job_id1] == 0
