Yifan Gu yifan-gu

tectonic install assets --dir=my-test-cluster

* module.assets_base.local.identity_client_ca_cert: local.identity_client_ca_cert: file: open /home/yifan/gopher/src/github.com/coreos/tectonic-installer/my-test-cluster/generated/tls/root-ca.crt: no such file or directory in:

${file("${local.tls_path}/root-ca.crt")}
* module.assets_base.local.aggregator_ca_cert_pem: local.aggregator_ca_cert_pem: file: open /home/yifan/gopher/src/github.com/coreos/tectonic-installer/my-test-cluster/generated/tls/aggregator-ca.crt: no such file or directory in:

${file("${local.tls_path}/aggregator-ca.crt")}
* module.assets_base.local.identity_server_cert_pem: local.identity_server_cert_pem: file: open /home/yifan/gopher/src/github.com/coreos/tectonic-installer/my-test-cluster/generated/tls/identity-server.crt: no such file or directory in:

Remove the selector section in the pod-checkpointer spec:

cat<<EOF | kubectl replace -f -
apiVersion: extensions/v1beta1
kind: DaemonSet
metadata:
  labels:
    k8s-app: pod-checkpointer
    tier: control-plane

To reschedule the tco pods to master nodes, run:

kubectl -n tectonic-system patch deployment tectonic-channel-operator -p '{"spec":{"template":{"spec":{"nodeSelector":{"node-role.kubernetes.io/master":""},"tolerations":[{"key":"node-role.kubernetes.io/master","operator":"Exists","effect":"NoSchedule"}]}}}}'

This works around the issue that tco can't run on rhel node because it expects a ssl cert dir that's not in rhel distros.

Hit some patch conflict when updating console

original_manifest:

apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: tectonic-console
  namespace: tectonic-system

Create a manifest

cat << EOF | kubectl create -f -
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  annotations:
  labels:
    k8s-app: tectonic-test-operator

Upload a payload to CoreUpdate

cd tectonic-installer

# make and sign the payload.
make payload
SIGN_WITH_TEST_KEY=true ./modules/update-payload/sign-payload.sh

# upload the payload

	[INFO] [19:08:41-0700] Running tests against existing cluster...
	[INFO] [19:08:41-0700] Running parallel tests N=<default>
	I0709 19:08:41.741853 10764 test.go:86] Extended test version v3.10.0-alpha.0+e63afaa-1228-dirty
	Running Suite: Extended
	=======================
	Random Seed: 1531188522 - Will randomize all specs
	Will run 447 specs

	Running in parallel across 5 nodes

	[INFO] [13:42:19-0700] Running tests against existing cluster...
	[INFO] [13:42:19-0700] Running parallel tests N=<default>
	I0709 13:42:20.323666 26542 test.go:86] Extended test version v3.10.0-alpha.0+e63afaa-1228
	Running Suite: Extended
	=======================
	Random Seed: 1531168941 - Will randomize all specs
	Will run 447 specs

	Running in parallel across 5 nodes

	I1031 20:06:45.955006 1 main.go:290] Should start checkpoint kube-system/pod-checkpointer-wqbq7
	I1031 20:06:45.955056 1 main.go:385] Checkpoint manifest for "kube-system/pod-checkpointer-wqbq7" already exists. Skipping
	I1031 20:06:48.975356 1 main.go:385] Checkpoint manifest for "kube-system/pod-checkpointer-wqbq7" already exists. Skipping
	I1031 20:06:48.992369 1 main.go:385] Checkpoint manifest for "kube-system/kube-etcd-0000" already exists. Skipping
	I1031 20:06:49.005260 1 main.go:385] Checkpoint manifest for "kube-system/kube-apiserver-qtmd7" already exists. Skipping
	I1031 20:06:49.010436 1 main.go:385] Checkpoint manifest for "kube-system/kube-etcd-network-checkpointer-2npz7" already exists. Skipping
	I1031 20:06:49.023779 1 main.go:232] API GC: skipping inactive checkpoint kube-system/kube-apiserver-qtmd7
	I1031 20:06:49.023799 1 main.go:232] API GC: skipping inactive checkpoint kube-system/kube-etcd-0000
	I1031 20:06:49.023805 1 main.go:232] API GC: sk

	*filter
	:KUBE-SERVICES - [0:0]
	COMMIT
	*nat
	:KUBE-SERVICES - [0:0]
	:KUBE-NODEPORTS - [0:0]
	:KUBE-POSTROUTING - [0:0]
	:KUBE-MARK-MASQ - [0:0]
	:KUBE-SVC-4RFU3GGCUIJHFZEZ - [0:0]
	:KUBE-SEP-CG3BO6CZOYEJLW7B - [0:0]