forked from bendory/tekton-on-gcp
-
Notifications
You must be signed in to change notification settings - Fork 0
/
setup.sh
executable file
·168 lines (148 loc) · 8.17 KB
/
setup.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
#!/bin/sh
set -e
dir=$(dirname $0)
. "${dir}"/env.sh
# Start API enablements so we don't have to wait for them below.
# For reasons I don't know, running artifactregistry enablement --async here and
# blocking on it below just before repository creation often leads to an error
# indicating that the API is not yet enabled. I am guessing there is a
# propogation issue. Workaround: synchronously enable here.
${gcloud} services enable artifactregistry.googleapis.com
${gcloud} services enable binaryauthorization.googleapis.com --async # Binary Authorization
${gcloud} services enable cloudkms.googleapis.com --async # KMS
${gcloud} services enable compute.googleapis.com --async # GCE
${gcloud} services enable container.googleapis.com --async # GKE
${gcloud} services enable containeranalysis.googleapis.com --async # Container Analysis for provenance storage
${gcloud} services enable containerscanning.googleapis.com --async # Vulnerability Scanning
${gcloud} services enable containerfilesystem.googleapis.com --async # Streaming Images
${gcloud} services enable iam.googleapis.com --async # IAM
# Create the BUILDER_SA. The BUILDER_SA is the ServiceAccount identity that will
# be used to authenticate and authorize GCP calls during the build.
${gcloud} services enable iam.googleapis.com # Ensure IAM is enabled
${gcloud} iam service-accounts create "${BUILDER}" \
--description="Tekton Build-time Service Account" \
--display-name="Tekton Builder"
# Set up Artifact Registry: create a docker repository and authorize the
# BUILDER_SA to push images to it.
${gcloud} artifacts repositories create "${REPO}" \
--repository-format=docker --location="${LOCATION}"
${gcloud} projects add-iam-policy-binding "${PROJECT}" \
--member="serviceAccount:${BUILDER_SA}" --role='roles/artifactregistry.writer'
# Set up VERIFIER_SA.
${gcloud} iam service-accounts create "${VERIFIER}" \
--description="Tekton Chains Service Account" \
--display-name="Tekton Chains"
# Enable the VERIFIER_SA to write Notes and Occurrences in Container Analysis.
${gcloud} services enable containeranalysis.googleapis.com # Ensure Container Analysis is enabled.
${gcloud} projects add-iam-policy-binding "${PROJECT}" \
--role roles/containeranalysis.notes.editor \
--member "serviceAccount:${VERIFIER_SA}"
${gcloud} projects add-iam-policy-binding "${PROJECT}" \
--role roles/containeranalysis.occurrences.editor \
--member "serviceAccount:${VERIFIER_SA}"
# Configure Key Management Service. Set up a private key that will be used by
# VERIFIER_SA to sign attestations.
# NOTE: the below commands assume that key_setup.sh in this directory has
# already executed successfully.
${key_gcloud} services enable cloudkms.googleapis.com # Ensure KMS is available.
echo "Checking for existence of KEYRING ${KEYRING}..."
if ${key_gcloud} kms keyrings describe "${KEYRING}" --location "${LOCATION}"; then
echo "KEYRING ${KEYRING} found."
else
echo "KEYRING ${KEYRING} NOT found. That's OK, I'll create it now."
${key_gcloud} kms keyrings create "${KEYRING}" --location "${LOCATION}"
echo "KEYRING ${KEYRING} created successfully."
fi
${key_gcloud} kms keys create "${KEY}" \
--keyring "${KEYRING}" \
--location "${LOCATION}" \
--purpose "asymmetric-signing" \
--default-algorithm "rsa-sign-pkcs1-2048-sha256"
${key_gcloud} kms keys add-iam-policy-binding "${KEY}" \
--location="${LOCATION}" --keyring="${KEYRING}" \
--member "serviceAccount:${VERIFIER_SA}" --role "roles/cloudkms.cryptoOperator"
${key_gcloud} kms keys add-iam-policy-binding "${KEY}" \
--location="${LOCATION}" --keyring="${KEYRING}" \
--member "serviceAccount:${VERIFIER_SA}" --role "roles/cloudkms.viewer"
# Set up GKE with Workload Identity, Binary Authorization, and Image Streaming
# Workload Identity is used to map a Kubernetes Service Account to our desired
# BUILDER_SA.
# https://cloud.google.com/kubernetes-engine/docs/how-to/workload-identity
# Image Streaming enables faster loading of containers.
# https://cloud.google.com/kubernetes-engine/docs/how-to/image-streaming
${gcloud} services enable \
binaryauthorization.googleapis.com \
compute.googleapis.com \
container.googleapis.com \
containerfilesystem.googleapis.com
${gcloud} container clusters create "${TEKTON_CLUSTER}" \
--zone="${ZONE}" --workload-pool="${PROJECT}.svc.id.goog" \
--num-nodes=1 --image-type="COS_CONTAINERD" --enable-image-streaming \
--binauthz-evaluation-mode="PROJECT_SINGLETON_POLICY_ENFORCE" \
--enable-autoscaling --min-nodes=1 --max-nodes=5 \
--workload-metadata="GKE_METADATA"
${gcloud} iam service-accounts add-iam-policy-binding \
"${BUILDER_SA}" --role roles/iam.workloadIdentityUser \
--member "serviceAccount:${PROJECT}.svc.id.goog[default/default]"
${k_tekton} annotate serviceaccount \
--namespace default default iam.gke.io/gcp-service-account="${BUILDER_SA}"
# Install Tekton Pipelines CRDs.
${k_tekton} apply --filename https://storage.googleapis.com/tekton-releases/pipeline/${PIPELINE_VERSION}/release.yaml
# Wait for pipelines to be ready.
unset status
while [[ "${status}" -ne "Running" ]]; do
echo "Waiting for Tekton Pipelines installation to complete."
status=$(${k_tekton} get pods --namespace tekton-pipelines -o custom-columns=':status.phase' | sort -u)
done
echo "Tekton Pipelines installation completed."
# Install tasks
${tkn} hub install task git-clone
sleep 10 # No idea why a pause reduces flakes.
${tkn} hub install task kaniko || ${tkn} hub install task kaniko
# Install Tekton Chains. Tekton Chains will gather build provenance for images
# and attest to their provenance from this Tekton installation.
${k_tekton} apply --filename https://storage.googleapis.com/tekton-releases/chains/${CHAINS_VERSION}/release.yaml
unset status
while [[ "${status}" -ne "Running" ]]; do
echo "Waiting for Tekton Chains installation to complete."
status=$(${k_tekton} get pods --namespace "${CHAINS_NS}" -o custom-columns=':status.phase' | sort -u)
done
echo "Tekton Chains installation completed."
# Configure Workload Identity for the Chains namespace. Chains runs in a
# different namespace and as a different Kubernetes Service Account in order to
# separate responsibilities between pipelines and Chains.
${gcloud} iam service-accounts add-iam-policy-binding \
$VERIFIER_SA --role roles/iam.workloadIdentityUser \
--member "serviceAccount:$PROJECT.svc.id.goog[${CHAINS_NS}/${VERIFIER}]"
${k_tekton} annotate serviceaccount "${VERIFIER}" --namespace "${CHAINS_NS}" \
iam.gke.io/gcp-service-account=${VERIFIER_SA}
# Configure Tekton Chains to use simplesigning of images; TaskRuns will be
# captured using in-toto. Attestations for both will be signed with a KMS key
# and stored in *BOTH* grafeas (Container Analysis) and in OCI bundles alongside
# the image itself in Artifact Registry.
# NOTE: by not setting `storage.oci.repository`, we store the OCI alongside the
# image itself in the image registry.
${k_tekton} patch configmap chains-config -n "${CHAINS_NS}" \
-p='{"data":{
"artifacts.oci.format": "simplesigning",
"artifacts.oci.signer": "kms",
"artifacts.oci.storage": "grafeas,oci",
"artifacts.pipelinerun.format": "in-toto",
"artifacts.pipelinerun.signer": "kms",
"artifacts.pipelinerun.storage": "grafeas,oci",
"artifacts.taskrun.format": "in-toto",
"artifacts.taskrun.signer": "kms",
"artifacts.taskrun.storage": "grafeas,oci" }}'
# Configure the KMS signing key, storage project in Container Analysis, and
# builder identifier used by Tekton Chains.
${k_tekton} patch configmap chains-config -n "${CHAINS_NS}" \
-p="{\"data\": {\
\"signers.kms.kmsref\": \"${KMS_URI}\", \
\"storage.grafeas.projectid\": \"${PROJECT}\" }}"
# To store OCI attestations alongside the image in AR, VERIFIER_SA needs
# write permission.
${gcloud} projects add-iam-policy-binding "${PROJECT}" \
--member="serviceAccount:${VERIFIER_SA}" --role='roles/artifactregistry.writer'
# Apply pipeline.yaml; see https://tekton.dev/docs/how-to-guides/kaniko-build-push/
${k_tekton} apply --filename "${dir}/pipeline.yaml"
echo "Setup complete! Run ./run_pipeline.sh to build and push your first container."