https://pulumi.com logo
#general
Title
# general
e

echoing-activity-32278

12/29/2021, 1:25 PM
Anyone has success with installing cert-manager in k8s? For me, Pulumi always get stuck in the middle, hanging there forever.
w

wonderful-twilight-70958

12/29/2021, 1:45 PM
Many times. Have you looked at the cluster to see the state / logs?
e

echoing-activity-32278

12/29/2021, 1:57 PM
Are you using the helm.release or using the k8s.chart?
w

wonderful-twilight-70958

12/29/2021, 2:16 PM
Currently the release, but I think I may have used the chart previously also, can't remember.
q

quiet-wolf-18467

12/29/2021, 2:20 PM
I've started publishing Pulumi SDKs too: https://www.npmjs.com/package/@pulumiverse/cert-manager
e

echoing-activity-32278

12/29/2021, 4:51 PM
Using the chart, once again hanging for 15mins now:
Copy code
// cert manager chart
// see <https://github.com/jetstack/cert-manager/blob/master/deploy/charts/cert-manager/values.yaml>
const certManagerName = utils.resourceName`cert-manager`;
const certManager = new k8s.helm.v3.Chart(
  certManagerName,
  {
    namespace: cluster.namespace.metadata.name,
    chart: "cert-manager",
    // see <https://cert-manager.io/docs/installation/helm/#4-install-cert-manager>
    version: "v1.5.4",
    fetchOpts: { repo: "<https://charts.jetstack.io>" },
    values: {
      // see <https://stackoverflow.com/a/66042994/855160>
      extraArgs: [
        "--dns01-recursive-nameservers-only",
        "--dns01-recursive-nameservers=1.1.1.1:53,1.0.0.1:53",
      ],
      installCRDs: "true",
    },
    transformations: [
      (obj) => {
        // NB do transformations on the yaml to set all resources to the namespace
        if (obj.metadata) {
          obj.metadata.namespace = cluster.namespace.metadata.name;
        }
      },
    ],
  },
  { provider: cluster.provider, customTimeouts: { create: "3m", update: "3m" } }
);
Last time I forcibly use Ctrl-c to quit Pulumi, and when I run Pulumi again it complains NO stack found.
w

wonderful-twilight-70958

12/29/2021, 4:58 PM
So, when this is hanging, what do the (k8s) logs show? Are pods created?
e

echoing-activity-32278

12/29/2021, 5:04 PM
The apicheck pod was created, but failed 6 times, reaching the backoff limit.
But pulumi is waiting on it.
The job (pod) was run immediately when pulumi runs the helm chart. A more desired time would be a little bit later because most cert manager resources have not been created.
For the missing stack data issue, basically I am experiencing the same things as: https://github.com/pulumi/pulumi/issues/4679