Skip to content

Conversation

@rbtr
Copy link
Collaborator

@rbtr rbtr commented Jan 19, 2022

Signed-off-by: Evan Baker rbtr@users.noreply.github.com

Reason for Change:

Instead of watching all NNCs (and Nodes and Pods) and filtering on the client-side, we should modify the Manager's cache to only Watch objects with the appropriate fields set which match our Node. This filtering happens server-side as part of the WatchList initialization and should effectively scope CNS's apiserver traffic to its own Node's objects instead of letting it scale with Node and Pod count.

Issue Fixed:

Requirements:

Notes:

Signed-off-by: Evan Baker <rbtr@users.noreply.github.com>
@rbtr rbtr requested a review from thatmattlong January 19, 2022 01:18
@rbtr rbtr self-assigned this Jan 19, 2022
Signed-off-by: Evan Baker <rbtr@users.noreply.github.com>
@thatmattlong thatmattlong merged commit 1b68ce1 into Azure:master Jan 21, 2022
@rbtr rbtr deleted the cns/fix/server-side-filter branch January 21, 2022 22:44
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants