[v1.10] Trim clusterrole permissions of Cilium #19674
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
maintainer-s-little-helper
bot
added
the
kind/backports
aanm
force-pushed
the
pr/trim-rbac-1.10
branch
2 times, most recently
from
820edfb
to
e6fd9bc
Compare
|
/test |
|
/test |
|
/test-backport-1.10 |
|
/test-backport-1.10 |
|
/test-gke |
|
The conformance tests are failing because of the helm changes. This needs to be merged into the stable branch and then verify if the CI started failing because of these changes. |
sayboras
approved these changes
May 4, 2022
[ upstream commit 900f668 ] Since Cilium does not set any finalizer in the owner of the CEP, a Pod, it does not make sense to set "BlockOwnerDeletion: true". Regardless of this option being `true` or `false`, the Pod dependent, in this case the CEP, is always* Garbage Collected by Kubernetes. *Only if the user specifies the pod deletion with the "orphan" deletion cascading strategy that the CEP will be kept. However, Cilium Operator will garbage collect orphaned Cilium Endpoints every 5 minutes by default. Signed-off-by: André Martins <andre@cilium.io>
[ upstream commit 2d63c9b ] Cilium does not need to perform any Pod update thus this permission can be removed from Cilium's Cluster Role. Signed-off-by: André Martins <andre@cilium.io>
Trimmed down clustermesh-apiserver's ClusterRole to the exact permissions that clustermesh-apiserver requires. Signed-off-by: André Martins <andre@cilium.io>
[ upstream commit 73d6cae ] Since this option only existed to set up annotations in Kubernetes Nodes before the introduction of CiliumNodes, contrary to the upstream commit this option will be kept to 'true' with the possibility for users to change it to 'false'. Signed-off-by: André Martins <andre@cilium.io>
[ upstream commit f612c97 ] To decrease the amount of permissions Cilium's requires to operate in a cluster, the node taint removal and the setup of the node condition NetworkUnavailable can be set through cilium-operator. Cilium-operator will remove, if set, the Cilium's specific node taints from the Kubernetes nodes as well as setting up the NetworkUnavailable node condition to 'false' once it detects there is a "Ready" Cilium pod in that node. Signed-off-by: André Martins <andre@cilium.io>
[ upstream commit 9014253 ] We can use the "status" subresource to update node annotations which also allow us to reduce the clusterrole's permissions of the cilium DaemonSet even further. Signed-off-by: André Martins <andre@cilium.io>
|
/test-backport-1.10 |
1 similar comment
|
/test-backport-1.10 |
joestringer
approved these changes
May 4, 2022
michi-covalent
added a commit
to cilium/cilium-cli
that referenced
this pull request
May 19, 2022
Ref: cilium/cilium#19674 Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
michi-covalent
added a commit
to cilium/cilium-cli
that referenced
this pull request
May 19, 2022
Ref: cilium/cilium#19674 Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
michi-covalent
added a commit
to cilium/cilium-cli
that referenced
this pull request
May 20, 2022
- Set remove-cilium-node-taints to true - Limit nodes write access from cilium-agent to nodes/status field. - Add node write access to cilium-operator. Ref: cilium/cilium#19674 Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
michi-covalent
added a commit
to cilium/cilium-cli
that referenced
this pull request
May 20, 2022
- Set remove-cilium-node-taints to true - Limit nodes write access from cilium-agent to nodes/status field. - Add node write access to cilium-operator. - Add 'unix://' prefix for hubble-relay peer-service config. Ref: cilium/cilium#19674 Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
michi-covalent
added a commit
to cilium/cilium-cli
that referenced
this pull request
May 20, 2022
- Set remove-cilium-node-taints to true - Limit nodes write access from cilium-agent to nodes/status field. - Add node write access to cilium-operator. - Add 'unix://' prefix for hubble-relay peer-service config. - Run 'sudo cilium' for external workload. Ref: cilium/cilium#19674 Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
tklauser
pushed a commit
to cilium/cilium-cli
that referenced
this pull request
May 20, 2022
- Set remove-cilium-node-taints to true - Limit nodes write access from cilium-agent to nodes/status field. - Add node write access to cilium-operator. - Add 'unix://' prefix for hubble-relay peer-service config. Ref: cilium/cilium#19674 Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
tklauser
pushed a commit
to cilium/cilium-cli
that referenced
this pull request
May 20, 2022
- Set remove-cilium-node-taints to true - Limit nodes write access from cilium-agent to nodes/status field. - Add node write access to cilium-operator. - Add 'unix://' prefix for hubble-relay peer-service config. Ref: cilium/cilium#19674 Signed-off-by: Michi Mutsuzaki <michi@isovalent.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Manual backport of #19074 and #19590
The main differences between the upstream changes and this PR is that "annotateK8sNodes" does not change its default to "false" but instead it keeps it to "true" to avoid breaking existing clusters.