Calisti Installation Error

[philiwon]

Hello,

I tried to follow the guide to install Calisti on my k8s cluster (1.22.12) with 2 nodes (1 master and 1 slave).  As you can see below, the installation stopped at the following point...(in verbose mode). Any guidance on how to resolve this is much appreciated. thanks.

regards

Phil.

...

smm/wait/readiness ❯ pending name=smm-federation-gateway, namespace=smm-system, apiVersion=apps/v1, kind=Deployment

Error with stack:

timed out waiting for the condition

wwwin-github.cisco.com/cisco-app-networking/smm/pkg/common/operator/reconciler.(*Dispatcher).Handle

/mnt/workspace/N_backyards_backyards_smm_v1.9.1/clean_state_for_goreleaser/pkg/common/operator/reconciler/component.go:160

wwwin-github.cisco.com/cisco-app-networking/smm/pkg/common/operator/reconciler.(*Dispatcher).Reconcile

/mnt/workspace/N_backyards_backyards_smm_v1.9.1/clean_state_for_goreleaser/pkg/common/operator/reconciler/component.go:91

wwwin-github.cisco.com/cisco-app-networking/smm/services/cli/internal/deploy.ReconcileRequestWithRetry.func1

/mnt/workspace/N_backyards_backyards_smm_v1.9.1/clean_state_for_goreleaser/services/cli/internal/deploy/reconcile.go:29

...

[Geevarghese Cheria]

Hi Phil,

   Are you following this documentation https://smm-docs.eticloud.io/docs/quickstart/ and using Free tier for the installation ? I think we need one master and 2 worker nodes. Please let us know on which platform have you created the k8s cluster and trying for the Calisti installation. https://calisti.app/

Thanks and Regards,

Geevarghese

[philiwon]

Thanks for replying. I tried both CentOS and Ubuntu (1 master plus 2 slaves version 1.21 and 1.22). Error is the same. it aborted at the same point mentioned on the original post. kind of "time out waiting...". I am behind a proxy but i have coded correctly and all pulling images of containerd works ok. all nodes are with 16 vCPUs, 32GB RAM and 100GB disk. any further clue? thanks.

[yawming]

Did you check on the cluster see any pod issue ?

[yawming]

Can you please also attach entire installation log ? Today I got the "Error with stack:
timed out waiting for the condition..." like to see if the same as yours

[yawming]

Which SMM binary you downloaded? Mac or Linux ?

First thing may want to check if the resource meet the min request.

https://calisti.app/quickstart - Prerequisites

Also check your cluster

"kubectl get pods -A" see if smm created pods are running ok

[tiswanso]

Hi Phil,

Thanks for trying Calisti!  My initial suspicion is that this sounds like it matches symptoms seen when the cluster-name auto-learned by SMM does not comply to Istio resource naming validation ("_" or "." or ":" characters).  On the cluster you could check this via "kubectl get clusters" to get the name of the cluster object.  If that's the case you should:

smm uninstall -a
CLUSNM=$(kubectl get clusters -o jsonpath='{.items[0].metadata.name}')
kubectl delete cluster $CLUSNM
smm install -a --cluster-name <any compliant name>

https://smm-docs.eticloud.io/docs/installation/single-cluster-setup/#install-hahahugoshortcode-s1-hbhb

(edit -- fixed the link to the external URL)

[philiwon]

Thanks all for your advice. Finally figured out that my lab is behind a proxy; i need to disable the proxy before the kubeadm init. Now Calisti is up and running.