CNI 1.1.0 support #1273
CNI 1.1.0 support #1273
Conversation
|
Need to discuss: May need to disable GC for now because it may happens unexpected resource removal by GC (if same CNI plugin is used for both, cluster network and net-attach-def, it mihgt be happen) |
Changed. |
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
|
/cc |
|
This pull request is stale because it has been open 90 days with no activity. Remove stale label or comment or this will be closed in 7 days. |
|
This pull request is stale because it has been open 90 days with no activity. Remove stale label or comment or this will be closed in 7 days. |
This change supports up to date CNI 1.1 command, GC and STATUS for cluster network.
This code changes CNI's GC command argument. Previously it just passes from parent CNI runtime, however, it may causes unexpected resource deletion if one CNI plugin is used in both cluster network and net-attach-def. This change generates valid attachments from multus CNI cache and passed to delegate CNI plugin.
There was a problem hiding this comment.
Some notes from our call:
- GC and STATUS
- Hard to invoke STATUS in a multus context, e.g. no status for macvlan as a delegated plugin, for example.
- Maybe only for cluster networks
- GC more complicated
- Only stores cluster network object.
- GC and STATUS are just delegated.
So, this delegation is what Multus is designed to do, I think this is the right place to start.
This PR support CNI SPEC 1.1.0, with new CNI commands, GC and STATUS.