GLaDAP opened a new issue, #34451:
URL: https://github.com/apache/airflow/issues/34451
### Apache Airflow version
Other Airflow 2 version (please specify below)
### What happened
We are running Apache Airflow 2.6.1, Python 3.8 and
apache-airflow-providers-google 10.8.0.
We want to list the Vertex AI Endpoints in a Google Cloud Project with
the`airflow.providers.google.cloud.operators.vertex_ai.endpoint_service.ListEndpointsOperator`.
However, when running the DAG, the operator returns an error stating the
response is not an instance of Endpoint. It does display the information about
the Endpoint in the logs.
The stack trace is shown below:
```
[2023-09-18, 11:24:37 UTC] {taskinstance.py:1824} ERROR - Task failed with
exception
Traceback (most recent call last):
File
"/usr/local/lib/python3.8/site-packages/airflow/providers/google/cloud/operators/vertex_ai/endpoint_service.py",
line 456, in execute
return [Endpoint.to_dict(result) for result in results]
File
"/usr/local/lib/python3.8/site-packages/airflow/providers/google/cloud/operators/vertex_ai/endpoint_service.py",
line 456, in <listcomp>
return [Endpoint.to_dict(result) for result in results]
File "/usr/local/lib/python3.8/site-packages/proto/message.py", line 406,
in to_dict
cls.pb(instance),
File "/usr/local/lib/python3.8/site-packages/proto/message.py", line 291,
in pb
raise TypeError("%r is not an instance of %s" % (obj, cls.__name__,))
name: "projects/<project_id>/locations/europe-west4/endpoints/<endpoint_id>"
display_name: "demo-endpoint"
description: "model-endpoint"
deployed_models {
dedicated_resources {
machine_spec {
machine_type: "n1-standard-2"
}
min_replica_count: 1
max_replica_count: 1
}
id: "<endpoint_id>"
model: "projects/<project>/locations/europe-west4/models/<model_id>"
model_version_id: "1"
display_name: "endpoint_test"
create_time {
seconds: 1695031973
nanos: 292536000
}
service_account: <service_account>
enable_access_logging: true
}
traffic_split {
key: <key>
value: 100
}
etag: "<etag>"
labels {
key: "source"
value: "airflow"
}
create_time {
seconds: 1695031963
nanos: 480401000
}
update_time {
seconds: 1695032127
nanos: 514426000
}
is not an instance of Endpoint
[2023-09-18, 11:24:37 UTC] {taskinstance.py:1345} INFO - Marking task as
FAILED.
```
Note that sensitive information is replaced in the log above.
### What you think should happen instead
We expected a successful dag run listing the endpoints in the target project.
### How to reproduce
The DAG is shown below:
```
from datetime import datetime
from airflow.decorators import dag
from airflow.providers.google.cloud.operators.vertex_ai.endpoint_service
import (
ListEndpointsOperator,
)
@dag(
dag_id="vertexai-list-endpoints-dag",
description="Example of listing endpoints",
start_date=datetime(2023, 9, 15)
)
def list_endpoints_example():
list_endpoints_operators = ListEndpointsOperator(
task_id="vertexai-list-endpoints",
project_id="<PROJECT_ID>",
region="europe-west4"
)
list_endpoints_operators
list_endpoints_example()
```
### Operating System
Debian 11 Bullseye
### Versions of Apache Airflow Providers
apache-airflow-providers-google==10.8.0
### Deployment
Other Docker-based deployment
### Deployment details
_No response_
### Anything else
This problem occurs every run (fails always)
### Are you willing to submit PR?
- [ ] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of
Conduct](https://github.com/apache/airflow/blob/main/CODE_OF_CONDUCT.md)
--
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]
