prole/scripts/test-cnpg-knoe-db.sh
chrisfu a069989315 Rename prole-db to knoe-db, add knoe-auth as cluster-internal KDC
Itemized changes:

1. knoe-auth: New cluster-internal KDC and SSO gateway service
   - Created etc/init_knoe_auth.sh based on init_kdc.sh with knoe-auth naming
   - Namespace defaults to SERVICE_NAMESPACE (knoe-system)
   - ConfigMap: knoe-auth-kdc-config, Secret: knoe-auth-secrets
   - Legacy cleanup removes old auth/dog/authority deployments

2. Orchestration: knoe-auth initializes before CloudNativePG
   - Updated prole.sh to insert init_knoe_auth.sh as step 2 (before CNPG)
   - Renumbered all subsequent initialization steps

3. Kong routing: Updated init_kong.sh to route to knoe-auth in SERVICE_NAMESPACE

4. Comment/reference updates for knoe-auth
   - Updated init_common_services.sh, init_service_layer.sh, init_kerberos.sh

5. prole-db renamed to knoe-db across the entire codebase
   - Renamed prole-db/ directory to knoe-db/
   - Renamed all prole-db Kubernetes manifests (deploy/opentofu, k8s/)
   - Renamed scripts: docker-root-knoe-db.sh, docker-run-knoe-db.sh, test-cnpg-knoe-db.sh
   - Renamed etc/init_prole-db-reset.sh to etc/init_knoe-db-reset.sh
   - Renamed etc/prole-db-passwwd.sh to etc/knoe-db-passwwd.sh
   - Renamed mock_val counterparts accordingly
   - Renamed tests/etc/test_init_prole-db-reset.sh to test_init_knoe-db-reset.sh
   - Renamed docs/prole-db-documentation-mcp-architecture.md to knoe-db variant
   - Renamed modes/k3d/prole-db/ to modes/k3d/knoe-db/
   - Renamed prole-db.iml to knoe-db.iml

6. Configuration updates
   - Updated conf/dev, conf/prod, conf/test, conf/service prole.cfg files
   - Updated conf/port-mapping.cfg
   - Updated etc/prole_cfg.sh and mock_val/prole_cfg.sh
   - Updated service/prole.cfg

7. Kubernetes manifests and deploy configuration
   - Updated deploy/opentofu/k3s ArgoCD application YAMLs
   - Updated kong-configmap.yaml and kustomization.yaml
   - Updated k3s/kong-config.yml and prole-resources.yaml
   - Updated prole-mssql-db deployment YAMLs
   - Updated supabase helm render and deploy scripts

8. Infrastructure and GCP Terraform
   - Updated deploy/gcp/terraform: folders, groups, IAM, service-projects

9. Python/installer code updates
   - Updated knoe/core: actions, build_context, controller, env, milestones
   - Updated knoe/milestone.py
   - Updated knoe/ui/screens: cfg, database, database_options, deploy, docker,
     navigation, security, services, validate
   - Updated knoe.spec, status.py

10. Shell script updates
    - Updated etc/: build_db, init_cloudnative_pg, init_cnpg_backup,
      init_db_manager, init_forgejo, init_gitlab, init_monitoring, init_openbao,
      init_port_forwards, init_postgrest, init_supabase_ports, status
    - Updated mock_val/ counterparts for all above scripts
    - Updated prole-net/init-prole-dns.sh
    - Updated bin/prole-kpf.sh, gitea/deploy.sh, supabase/deploy.sh

11. Test updates
    - Updated tests/etc/: test_init_cloudnative_pg*, test_init_cnpg_backup*,
      test_init_kdc*, test_init_kerberos*, test_init_kong*, test_prole_cfg*
    - Updated tests/installer/: test_actions_helpers, test_cfg_save_kubecontext,
      test_controller, test_core_classes, test_milestones, test_milestones_extended,
      test_namespace_propagation
    - Updated tests/: test_database_options, test_navigation,
      test_render_supabase_hostname, test_docker_build_fix,
      test_all_prole_home_fixes, silent_install_test, final_test

12. Documentation updates
    - Updated docs/: DOCKER-BUILD-FIX, PROLE-CFG-SECRETS, PROLE-HOME-DIRECTORY,
      build-system, patent
    - Updated scan/network_description.txt
    - Updated pom.xml

13. Miscellaneous script updates
    - Updated root-level: _adopt_replica_pvcs, _fix_replica_merlin, _import_pi,
      _patch_cluster, _prebind_pvcs, _rebind_d002, _rebind_d002b, test_resolve
    - Updated scripts/generate_spec.py

Co-authored-by: Junie <junie@jetbrains.com>
2026-03-22 22:16:21 -07:00

264 lines
7.4 KiB
Bash
Executable File

#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
PROLE_HOME=${PROLE_HOME:-$(cd "$SCRIPT_DIR/.." && pwd)}
CNPG_TEST_NAMESPACE=${CNPG_TEST_NAMESPACE:-prole-cnpg-test}
CNPG_CLUSTER_NAME=${CNPG_CLUSTER_NAME:-knoe-db}
CNPG_OPERATOR_NAMESPACE=${CNPG_OPERATOR_NAMESPACE:-cnpg-system}
CNPG_CLUSTER_MANIFEST=${CNPG_CLUSTER_MANIFEST:-$PROLE_HOME/k8s/prole/knoe-db.yaml}
CNPG_USER_SECRET_FILE=${CNPG_USER_SECRET_FILE:-$PROLE_HOME/knoe-db/knoe-db-user-secret.yaml}
CNPG_WAIT_TIMEOUT=${CNPG_WAIT_TIMEOUT:-300}
CNPG_FALLBACK_VERSION=${CNPG_FALLBACK_VERSION:-1.27.0}
SKIP_BUILD=${SKIP_BUILD:-false}
IMAGE_REPO=${IMAGE_REPO:-knoe-db}
IMAGE_LOAD=${IMAGE_LOAD:-}
IMAGE_LOAD_CMD=${IMAGE_LOAD_CMD:-}
log() {
printf "[%s] %s\n" "$(date +%H:%M:%S)" "$*"
}
require_tool() {
command -v "$1" >/dev/null 2>&1 || { echo "Missing required tool: $1" >&2; exit 1; }
}
get_latest_cnpg_version() {
local version=""
if command -v python3 >/dev/null 2>&1; then
version=$(python3 - <<'PY'
import json,sys,urllib.request
try:
with urllib.request.urlopen("https://api.github.com/repos/cloudnative-pg/cloudnative-pg/releases/latest", timeout=10) as resp:
data=json.load(resp)
tag=data.get("tag_name", "")
print(tag.lstrip("v"))
except Exception:
pass
PY
) || true
fi
if [[ -z "$version" ]]; then
version=$(curl -fsSL "https://api.github.com/repos/cloudnative-pg/cloudnative-pg/releases/latest" 2>/dev/null \
| sed -n 's/.*"tag_name": *"v\{0,1\}\([^"]*\)".*/\1/p' | head -n1) || true
fi
if [[ -z "$version" ]]; then
echo "$CNPG_FALLBACK_VERSION"
else
echo "$version"
fi
}
resolve_cnpg_manifest_url() {
if [[ -n "${CNPG_MANIFEST_URL:-}" ]]; then
echo "$CNPG_MANIFEST_URL"
return 0
fi
local version
version=${CNPG_VERSION:-$(get_latest_cnpg_version)}
local minor
minor=$(echo "$version" | awk -F. '{print $1"."$2}')
echo "https://raw.githubusercontent.com/cloudnative-pg/cloudnative-pg/release-${minor}/releases/cnpg-${version}.yaml"
}
compute_image_tag() {
local pg_version release
pg_version=$(tr -d '[:space:]' < "$PROLE_HOME/conf/postgresql/.version" 2>/dev/null || true)
release=$(tr -d '[:space:]' < "$PROLE_HOME/knoe-db/.version" 2>/dev/null || true)
if [[ -z "$pg_version" ]]; then
pg_version="17.7"
fi
if [[ -z "$release" ]]; then
release="000"
fi
if [[ "$release" =~ ^[0-9]+$ ]]; then
release=$(printf "%03d" "$release")
fi
echo "${pg_version}-${release}"
}
ensure_namespace() {
if ! kubectl get namespace "$CNPG_TEST_NAMESPACE" >/dev/null 2>&1; then
log "Creating namespace $CNPG_TEST_NAMESPACE"
kubectl create namespace "$CNPG_TEST_NAMESPACE" >/dev/null
fi
}
ensure_cnpg_operator() {
if kubectl get crd clusters.postgresql.cnpg.io >/dev/null 2>&1; then
log "CNPG CRDs detected; operator install skipped."
return 0
fi
local url
url=$(resolve_cnpg_manifest_url)
log "Installing CloudNative-PG operator from: $url"
kubectl apply --server-side -f "$url"
if kubectl -n "$CNPG_OPERATOR_NAMESPACE" get deploy cnpg-controller-manager >/dev/null 2>&1; then
kubectl -n "$CNPG_OPERATOR_NAMESPACE" rollout status deploy/cnpg-controller-manager --timeout=120s || true
fi
}
build_image() {
local image_tag
image_tag=$(compute_image_tag)
IMAGE="${IMAGE_REPO}:${image_tag}"
if [[ "$SKIP_BUILD" == "true" ]]; then
log "Skipping docker build (SKIP_BUILD=true)."
return 0
fi
log "Building image $IMAGE"
docker build -t "$IMAGE" "$PROLE_HOME/knoe-db"
}
load_image_if_requested() {
if [[ -n "$IMAGE_LOAD_CMD" ]]; then
log "Loading image with custom command: $IMAGE_LOAD_CMD"
eval "$IMAGE_LOAD_CMD"
return 0
fi
case "$IMAGE_LOAD" in
kind)
require_tool kind
log "Loading image into kind: $IMAGE"
kind load docker-image "$IMAGE"
;;
minikube)
require_tool minikube
log "Loading image into minikube: $IMAGE"
minikube image load "$IMAGE"
;;
k3d)
require_tool k3d
log "Importing image into k3d: $IMAGE"
if [[ -n "${K3D_CLUSTER_NAME:-}" ]]; then
k3d image import "$IMAGE" -c "$K3D_CLUSTER_NAME"
else
k3d image import "$IMAGE"
fi
;;
"")
;;
*)
echo "Unknown IMAGE_LOAD value: $IMAGE_LOAD" >&2
exit 2
;;
esac
}
render_cluster_manifest() {
local out_file=$1
if [[ ! -f "$CNPG_CLUSTER_MANIFEST" ]]; then
echo "Cluster manifest not found: $CNPG_CLUSTER_MANIFEST" >&2
exit 1
fi
awk -v image="$IMAGE" '
BEGIN { IGNORECASE=1 }
/^ imageName:/ { print " imageName: " image; next }
/^ instances:/ { print " instances: 3"; next }
/postgis_tiger_geocoder/ { next }
/schema tiger/ { next }
{ print }
' "$CNPG_CLUSTER_MANIFEST" > "$out_file"
}
collect_error_pod_logs() {
local out_dir=$1
local pods
pods=$(kubectl -n "$CNPG_TEST_NAMESPACE" get pods -l "cnpg.io/cluster=$CNPG_CLUSTER_NAME" --no-headers 2>/dev/null \
| awk '$3 ~ /(Error|CrashLoopBackOff|Init:Error|Init:CrashLoopBackOff)/ {print $1}')
if [[ -z "$pods" ]]; then
log "No error pods detected."
return 0
fi
mkdir -p "$out_dir"
for pod in $pods; do
kubectl -n "$CNPG_TEST_NAMESPACE" describe pod "$pod" > "$out_dir/${pod}.describe" 2>&1 || true
kubectl -n "$CNPG_TEST_NAMESPACE" logs "$pod" --all-containers --prefix=true --timestamps=true > "$out_dir/${pod}.log" 2>&1 || true
kubectl -n "$CNPG_TEST_NAMESPACE" logs "$pod" --all-containers --prefix=true --timestamps=true --previous > "$out_dir/${pod}.previous.log" 2>&1 || true
done
log "Error pod logs written to $out_dir"
}
wait_for_primary_ready() {
log "Waiting for CNPG primary pod to be Ready (timeout ${CNPG_WAIT_TIMEOUT}s)..."
local deadline
deadline=$(( $(date +%s) + CNPG_WAIT_TIMEOUT ))
while [[ $(date +%s) -lt $deadline ]]; do
if kubectl -n "$CNPG_TEST_NAMESPACE" get pods \
-l "cnpg.io/cluster=$CNPG_CLUSTER_NAME,cnpg.io/instanceRole=primary" \
--no-headers 2>/dev/null | grep -q .; then
if kubectl -n "$CNPG_TEST_NAMESPACE" wait --for=condition=Ready pod \
-l "cnpg.io/cluster=$CNPG_CLUSTER_NAME,cnpg.io/instanceRole=primary" \
--timeout=30s; then
return 0
fi
fi
sleep 5
done
return 1
}
main() {
require_tool kubectl
require_tool docker
build_image
load_image_if_requested
ensure_namespace
ensure_cnpg_operator
if [[ -f "$CNPG_USER_SECRET_FILE" ]]; then
log "Applying CNPG user secret: $CNPG_USER_SECRET_FILE"
kubectl -n "$CNPG_TEST_NAMESPACE" apply -f "$CNPG_USER_SECRET_FILE"
else
log "Skipping CNPG user secret (file not found): $CNPG_USER_SECRET_FILE"
fi
local tmp_dir
tmp_dir=$(mktemp -d)
local rendered_manifest="$tmp_dir/cluster.yaml"
render_cluster_manifest "$rendered_manifest"
log "Applying CNPG cluster manifest: $rendered_manifest"
kubectl -n "$CNPG_TEST_NAMESPACE" apply -f "$rendered_manifest"
if ! wait_for_primary_ready; then
local log_dir
local base_log_dir
base_log_dir="${PROLE_LOGS:-$PROLE_HOME/logs}"
base_log_dir="${base_log_dir%/}"
log_dir=${CNPG_TEST_LOG_DIR:-$base_log_dir/cnpg-test-$(date +%Y%m%dT%H%M%S)}
collect_error_pod_logs "$log_dir"
kubectl -n "$CNPG_TEST_NAMESPACE" get pods -l "cnpg.io/cluster=$CNPG_CLUSTER_NAME" || true
echo "CNPG primary pod did not become Ready within timeout." >&2
exit 1
fi
kubectl -n "$CNPG_TEST_NAMESPACE" get pods -l "cnpg.io/cluster=$CNPG_CLUSTER_NAME" || true
log "CNPG test completed successfully."
}
main "$@"