Discuss what needs to be done still on celery, and usage
We started working towards making it possible to distribute tasks in lava using celery during this past cycle, and most of the pieces seem to be there, but with nothing set up to actually use it yet. We are in the process of setting up an internal cloud where we can launch instances to run celery (and other) tasks.
1. What is still needed to make this work?
* Do we need an API for our celery tasks to launch an instance in the cloud before starting?
* Does it make sense to deploy a new instance every time a new task comes in, or can we reuse them safely without adding too much complexity and have a dynamic pool of celery running instances that can grow/shrink with demand?
* Celery permissions bug - how critical is it?
* documentation/
2. Usage - what are the primary places we want to use this?
* LMC tasks has been mentioned before, but I'm sure we'll still need to limit
* add more before discussion if you have some
Blueprint information
- Status:
- Not started
- Approver:
- Paul Larson
- Priority:
- Undefined
- Drafter:
- None
- Direction:
- Needs approval
- Assignee:
- None
- Definition:
- Discussion
- Series goal:
- None
- Implementation:
-
Unknown
- Milestone target:
- None
- Started by
- Completed by
