-
Notifications
You must be signed in to change notification settings - Fork 180
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Resource allocation and netplugin #8
Comments
@mapuri - distributed lock in etcd can offer the global allocation be done locally, in a clustered environment. A very common use case of distributed lock is to work on a compare-and-swap operations that otherwise would need to be done by a controller/central entity. Etcd docs recommend using this implementation |
@jainvipin, I was not just referring to global lock problem but the requirement for hosts to read states of other hosts in order to come up with next available value. Actually a distributed lock in this case might slow down things depending on amount of state that needs to be kept locked whenever a host needs to allocate a resource. I think this logic can be much simpler when done centrally at one place. |
closing the issue, as it is addressed by in basic form by PR #23 |
Move away from using etcd modifiedIndex in compare-and-swap
updating bgp object
Right now the resource allocation is implemented inside the netplugin using a global configuration state that contains state about configurable pools of IP, vlans etc and it publishes by some operational state about resource state (like consumed/free ids etc).
This quickly get's tricky in a multi-host environment where each netplugin implementation now needs to do resource allocation on it's own and publish back operation state.
A brief discussion of issues is here as well (issue #6). This issue tracks changes for a working solution.
The text was updated successfully, but these errors were encountered: