snjypl commented on code in PR #28161:
URL: https://github.com/apache/airflow/pull/28161#discussion_r1043664573
##########
airflow/utils/log/file_task_handler.py:
##########
@@ -173,88 +170,15 @@ def _read(self, ti: TaskInstance, try_number: int,
metadata: dict[str, Any] | No
log = f"*** Failed to load local log file: {location}\n"
log += f"*** {str(e)}\n"
return log, {"end_of_log": True}
- elif conf.get("core", "executor") == "KubernetesExecutor":
- try:
- from airflow.kubernetes.kube_client import get_kube_client
-
- kube_client = get_kube_client()
-
- if len(ti.hostname) >= 63:
- # Kubernetes takes the pod name and truncates it for the
hostname. This truncated hostname
- # is returned for the fqdn to comply with the 63 character
limit imposed by DNS standards
- # on any label of a FQDN.
- pod_list =
kube_client.list_namespaced_pod(conf.get("kubernetes_executor", "namespace"))
- matches = [
- pod.metadata.name
- for pod in pod_list.items
- if pod.metadata.name.startswith(ti.hostname)
- ]
- if len(matches) == 1:
- if len(matches[0]) > len(ti.hostname):
- ti.hostname = matches[0]
-
- log += f"*** Trying to get logs (last 100 lines) from worker
pod {ti.hostname} ***\n\n"
-
- res = kube_client.read_namespaced_pod_log(
- name=ti.hostname,
- namespace=conf.get("kubernetes_executor", "namespace"),
- container="base",
- follow=False,
- tail_lines=100,
- _preload_content=False,
- )
-
- for line in res:
- log += line.decode()
-
- except Exception as f:
- log += f"*** Unable to fetch logs from worker pod
{ti.hostname} ***\n{str(f)}\n\n"
- return log, {"end_of_log": True}
else:
- import httpx
- url = self._get_log_retrieval_url(ti, log_relative_path)
- log += f"*** Log file does not exist: {location}\n"
- log += f"*** Fetching from: {url}\n"
- try:
- timeout = None # No timeout
- try:
- timeout = conf.getint("webserver", "log_fetch_timeout_sec")
- except (AirflowConfigException, ValueError):
- pass
-
- signer = JWTSigner(
- secret_key=conf.get("webserver", "secret_key"),
- expiration_time_in_seconds=conf.getint(
- "webserver", "log_request_clock_grace", fallback=30
- ),
- audience="task-instance-logs",
- )
- response = httpx.get(
- url,
- timeout=timeout,
- headers={"Authorization":
signer.generate_signed_token({"filename": log_relative_path})},
- )
- response.encoding = "utf-8"
-
- if response.status_code == 403:
- log += (
- "*** !!!! Please make sure that all your Airflow
components (e.g. "
- "schedulers, webservers and workers) have "
- "the same 'secret_key' configured in 'webserver'
section and "
- "time is synchronized on all your machines (for
example with ntpd) !!!!!\n***"
- )
- log += (
- "*** See more at
https://airflow.apache.org/docs/apache-airflow/";
- "stable/configurations-ref.html#secret-key\n***"
- )
- # Check if the resource was properly fetched
- response.raise_for_status()
-
- log += "\n" + response.text
- except Exception as e:
- log += f"*** Failed to fetch log file from worker. {str(e)}\n"
- return log, {"end_of_log": True}
+ executor = ExecutorLoader.get_default_executor()
+ task_log = executor.get_task_log(ti,
log_relative_path=log_relative_path)
+
+ if isinstance(task_log, tuple):
+ return task_log
Review Comment:
task_log is the returned tuple | str. incase of exception, we are
returning the tuple
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
