jupyterhub / jupyterhub/simpervisor
Document ways to make sure our state is always synchronized
Open
- Dominant language
- Python
- Stars
- 14
- Forks
- 9
- PR merge metrics
- No merged PRs in 30d
Description
We've several flags like running and _killed that *must* be synchronized with the real state of the process we are supervising. We potentially lose control with every `await` call, so need to be careful where we set the flags. Also need to realize that the flags could be set to something else when we come back from the await call.
I suspect we should count lines between await calls as something similar to critical sections and set our flags there. And not assume any flags set before an await are the same after.
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.