You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@airflow.apache.org by GitBox <gi...@apache.org> on 2021/04/21 22:30:37 UTC
[GitHub] [airflow] zachliu commented on issue #15000: When an ECS Task fails to start, ECS Operator raises a CloudWatch exception
zachliu commented on issue #15000:
URL: https://github.com/apache/airflow/issues/15000#issuecomment-824404502
Yes. And Fargate tasks are more likely to experience this `failed-to-start` issue than the conventional EC2 tasks. I've been back and forth with AWS support on this one for a long long time. The AWS's diagnostic is this:
* Randomly AWS could fail to provision the ENI (elastic network interface) when running your Fargate task, and when this happens, the Fargate task fails to start **silently**. The only way to avoid this is to write your own "baby-sitting" routine to poll the `eni_status` in your ecs task after you trigger the task.
But luckily we only see around 1% Fargate tasks failed to start due to this random issue.
I didn't close https://github.com/apache/airflow/issues/11663 just to remind me to create a new issue here but apparently @kanga333 beat me to it :grin:
--
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.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org