-
Notifications
You must be signed in to change notification settings - Fork 898
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
Added Daemonset to istio-cni and fixed the script #2782
Added Daemonset to istio-cni and fixed the script #2782
Conversation
Signed-off-by: biswajit-9776 <[email protected]>
Signed-off-by: biswajit-9776 <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think you need to drop --cluster-specific such that the hack script works without a running cluster. Why are you using the default profile in line 66? Isn't there a CNI profile?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
[biswa@fedora bin]$ istioctl profile list
Istio configuration profiles:
ambient
default
demo
empty
minimal
openshift
openshift-ambient
preview
remote
stable
I don't see it explicitly here and it's default in istio-cni-1-17 README.md:
https://github.com/kubeflow/manifests/tree/v1.8-branch/common/istio-cni-1-17#:~:text=%24%20cd%20%24ISTIO_NEW%0A%24%20istioctl%20profile%20dump%20default%20%3E%20profile.yaml
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Also I found this:
https://istio.io/latest/docs/setup/additional-setup/config-profiles/#:~:text=The%20components%20marked,minimal
Do you think CNI is a component of ambient profile instead of default?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, maybe it changed recently. Then let's drop only the --cluster-specific and test on your cluster whether it really works.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I have pushed the commit for the updated script and here are the cluster resources:
[biswa@fedora manifests]$ kubectl get daemonsets -n kube-system
NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE
istio-cni-node 3 3 3 3 3 kubernetes.io/os=linux 103s
kindnet 3 3 3 3 3 kubernetes.io/os=linux 2m29s
kube-proxy 3 3 3 3 3 kubernetes.io/os=linux 2m30s
I also see 3 healthy istio-cni pods in kube-system namespace in each cluster node:
[biswa@fedora manifests]$ kubectl get pods -n kube-system
NAME READY STATUS RESTARTS AGE
coredns-76f75df574-db5lj 1/1 Running 0 3m57s
coredns-76f75df574-pdml9 1/1 Running 0 3m57s
etcd-kind-control-plane 1/1 Running 0 4m12s
istio-cni-node-28r44 1/1 Running 0 3m25s
istio-cni-node-rpx22 1/1 Running 0 3m25s
istio-cni-node-tlg2b 1/1 Running 0 3m25s
kindnet-bmfbv 1/1 Running 0 3m57s
kindnet-mk64l 1/1 Running 0 3m51s
kindnet-pjz6s 1/1 Running 0 3m52s
kube-apiserver-kind-control-plane 1/1 Running 0 4m12s
kube-controller-manager-kind-control-plane 1/1 Running 0 4m12s
kube-proxy-k5rcg 1/1 Running 0 3m51s
kube-proxy-qkbk7 1/1 Running 0 3m57s
kube-proxy-zjvpn 1/1 Running 0 3m52s
kube-scheduler-kind-control-plane 1/1 Running 0 4m12s
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Do you think we should maybe check if the ip tables are modified?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, please check a few initcontainers
I tried deploying two containers httpbin and sleep from
|
These logs of istio-cni-node present in the same worker node point out that ip-tables are configured for the pods that are created in those namespaces and saved as logs in the istio-cni-node |
/lgtm |
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: juliusvonkohout The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
* Added Daemonset to istio-cni and fixed the script Signed-off-by: biswajit-9776 <[email protected]> * Removed --cluster-specific from script in hack Signed-off-by: biswajit-9776 <[email protected]> --------- Signed-off-by: biswajit-9776 <[email protected]> Signed-off-by: hansinikarunarathne <[email protected]>
Pull Request Template for Kubeflow manifests Issues
✏️ A brief description of the changes
📦 List any dependencies that are required for this change
🐛 If this PR is related to an issue, please put the link of the issue here.
✅ Unit Test Checklist
✅ Contributor checklist
DCO
check)cla/google
check)