Subhashini2610 opened a new issue, #29031:
URL: https://github.com/apache/airflow/issues/29031

   ### Apache Airflow version
   
   2.5.0
   
   ### What happened
   
   I have an instance of Airflow running on a K8s cluster. For some strange 
reason, the scheduler pods are restarted every now and then due to a liveness 
probe failure. 
   ```
   Warning  Unhealthy  18m (x298 over 8d)  kubelet  Liveness probe failed: HTTP 
probe failed with statuscode: 503
   ```
   
   The current liveness probe command being used is 
   ```
   http-get http://:8641/health delay=10s timeout=20s period=60s #success=1 
#failure=5
   ```
   
   I have also tried using a different liveness probe
   ```
   airflow jobs check --job-type TriggererJob --local
   ```
   
   
   ### What you think should happen instead
   
   The scheduler pods must not be restarted randomly
   
   ### How to reproduce
   
   A simple K8s deployment of Airflow v2.5.0 with KubernetesExecutor and 
external Postgres for the DB creates the issue. 
   
   
   ### Operating System
   
   K8s deployment on AKS
   
   ### Versions of Apache Airflow Providers
   
   _No response_
   
   ### Deployment
   
   Official Apache Airflow Helm Chart
   
   ### Deployment details
   
   
   I have a customised image with Dockerfile as follows:
   ```
   FROM apache/airflow:2.5.0
   COPY requirements.txt /
   RUN pip install --no-cache-dir -r /requirements.txt
   USER airflow
   ```
   
   K8s version: 1.23.12 on Azure Kubernetes Service
   
   ### 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