karunpoudel opened a new issue #22702:
URL: https://github.com/apache/airflow/issues/22702


   ### Apache Airflow version
   
   2.2.2
   
   ### What happened
   
   Gantt chart doesn't show up if you provide vault annotation in 
executor_config in your task. Vault annotation generally needs a template 
string where you need to use double quotes and newline like shown below.
   
![image](https://user-images.githubusercontent.com/62040859/161346729-fb19bbed-a23d-488b-b874-c011e9a0dd4e.png)
   
   Following is console log from google chrome browser when opening Gantt chart 
page
   
![image](https://user-images.githubusercontent.com/62040859/161347034-230a9512-cf11-4c8a-b894-43c931768097.png)
   which refers to this line of code
   
![image](https://user-images.githubusercontent.com/62040859/161347106-5064cd04-44c4-46fa-a183-56a6803255e7.png)
   
   Browser is not able to parse the json string because it not generated 
correctly to handle backslash that might show up in taskinstance log (as shown 
above).
   
   This throws error:
   ```
   JSON.parse('{"vault.hashicorp.com/agent-inject-template-artifactory":"\n     
 {{- with secret \"secret/data/nastdomains/artifactory\" -}}\n      {{- 
.Data.data | toJSON -}}\n      {{- end }}"}')
   Unexpected token in JSON at position 58
   ```
   
   ### What you think should happen instead
   
   The taskinstance result should be properly escaped for backslash character 
when writing to html document.
   Instead of exporting 
   `{"vault.hashicorp.com/agent-inject-template-artifactory": "\n      {{- with 
secret \"path/to/secrets\" -}}\n      {{- .Data.data | toJSON -}}\n      {{- 
end }}"}`
   it should be exported as
   `{"vault.hashicorp.com/agent-inject-template-artifactory": "\\n      {{- 
with secret \\"path/to/secrets\\" -}}\\n      {{- .Data.data | toJSON -}}\\n    
  {{- end }}"}`
   So that is can be correctly consumed by JSON.parse() javascript function.
   
   ### How to reproduce
   
   Create a DAG with following code. Trigger the DAG. After it completes 
(either success or failure), go to Gantt view.
   ```
   from airflow import DAG
   from airflow.operators.dummy import DummyOperator
   from kubernetes.client import models as k8s
   import pendulum
   
   annotations = {
       'vault.hashicorp.com/agent-inject': 'true',
       'vault.hashicorp.com/role': f'vault-role',
       'vault.hashicorp.com/agent-pre-populate-only': 'true',
       f'vault.hashicorp.com/agent-inject-secret-filename': 'path/to/secret',
       f'vault.hashicorp.com/agent-inject-template-filename': f"""
         {{{{- with secret "path/to/secret" -}}}}
         {{{{- .Data | toJSON -}}}}
         {{{{- end }}}}"""
   }
   
   dag = DAG(
       dag_id="example2",
       start_date=pendulum.datetime(2021, 1, 1),
       schedule_interval=None,
       max_active_runs=1,
       catchup=False,
   )
   
   with dag:
       t0 = DummyOperator(
           task_id='start',
           executor_config={
               "pod_override": k8s.V1Pod(
                   metadata=k8s.V1ObjectMeta(annotations=annotations),
               )
           }
       )
   ```
   
   ### Operating System
   
   Debian GNU/Linux 11 (bullseye)
   
   ### Versions of Apache Airflow Providers
   
   apache-airflow-providers-cncf-kubernetes==1!2.1.0
   
   ### Deployment
   
   Astronomer
   
   ### Deployment details
   
   _No response_
   
   ### Anything else
   
   _No response_
   
   ### 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]


Reply via email to