Files
honeyDueAPI/deploy/scripts/deploy_prod.sh
Trey t 7690f07a2b Harden API security: input validation, safe auth extraction, new tests, and deploy config
Comprehensive security hardening from audit findings:
- Add validation tags to all DTO request structs (max lengths, ranges, enums)
- Replace unsafe type assertions with MustGetAuthUser helper across all handlers
- Remove query-param token auth from admin middleware (prevents URL token leakage)
- Add request validation calls in handlers that were missing c.Validate()
- Remove goroutines in handlers (timezone update now synchronous)
- Add sanitize middleware and path traversal protection (path_utils)
- Stop resetting admin passwords on migration restart
- Warn on well-known default SECRET_KEY
- Add ~30 new test files covering security regressions, auth safety, repos, and services
- Add deploy/ config, audit digests, and AUDIT_FINDINGS documentation

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-02 09:48:01 -06:00

398 lines
13 KiB
Bash
Executable File

#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
DEPLOY_DIR="$(cd "${SCRIPT_DIR}/.." && pwd)"
REPO_DIR="$(cd "${DEPLOY_DIR}/.." && pwd)"
STACK_TEMPLATE="${DEPLOY_DIR}/swarm-stack.prod.yml"
CLUSTER_ENV="${DEPLOY_DIR}/cluster.env"
REGISTRY_ENV="${DEPLOY_DIR}/registry.env"
PROD_ENV="${DEPLOY_DIR}/prod.env"
SECRET_POSTGRES="${DEPLOY_DIR}/secrets/postgres_password.txt"
SECRET_APP_KEY="${DEPLOY_DIR}/secrets/secret_key.txt"
SECRET_EMAIL_PASS="${DEPLOY_DIR}/secrets/email_host_password.txt"
SECRET_FCM_KEY="${DEPLOY_DIR}/secrets/fcm_server_key.txt"
SECRET_APNS_KEY="${DEPLOY_DIR}/secrets/apns_auth_key.p8"
SKIP_BUILD="${SKIP_BUILD:-0}"
SKIP_HEALTHCHECK="${SKIP_HEALTHCHECK:-0}"
log() {
printf '[deploy] %s\n' "$*"
}
warn() {
printf '[deploy][warn] %s\n' "$*" >&2
}
die() {
printf '[deploy][error] %s\n' "$*" >&2
exit 1
}
require_cmd() {
command -v "$1" >/dev/null 2>&1 || die "Missing required command: $1"
}
contains_placeholder() {
local value="$1"
[[ -z "${value}" ]] && return 0
local lowered
lowered="$(printf '%s' "${value}" | tr '[:upper:]' '[:lower:]')"
case "${lowered}" in
*changeme*|*replace_me*|*example.com*|*your-*|*todo*|*fill_me*|*paste_here*)
return 0
;;
*)
return 1
;;
esac
}
ensure_file_from_example() {
local path="$1"
local example="${path}.example"
if [[ -f "${path}" ]]; then
return
fi
if [[ -f "${example}" ]]; then
cp "${example}" "${path}"
die "Created ${path} from template. Fill it in and rerun."
fi
die "Missing required file: ${path}"
}
require_var() {
local name="$1"
local value="${!name:-}"
[[ -n "${value}" ]] || die "Missing required value: ${name}"
if contains_placeholder "${value}"; then
die "Value still uses placeholder text: ${name}=${value}"
fi
}
require_secret_file() {
local path="$1"
local label="$2"
ensure_file_from_example "${path}"
local contents
contents="$(tr -d '\r' < "${path}" | sed 's/[[:space:]]*$//')"
[[ -n "${contents}" ]] || die "Secret file is empty: ${path}"
if contains_placeholder "${contents}"; then
die "Secret file still has placeholder text (${label}): ${path}"
fi
}
print_usage() {
cat <<'EOF'
Usage:
./.deploy_prod
Optional environment flags:
SKIP_BUILD=1 Deploy existing image tags without rebuilding/pushing.
SKIP_HEALTHCHECK=1 Skip final HTTP health check.
DEPLOY_TAG=<tag> Override image tag (default: git short sha).
EOF
}
while (($# > 0)); do
case "$1" in
-h|--help)
print_usage
exit 0
;;
*)
die "Unknown argument: $1"
;;
esac
done
require_cmd docker
require_cmd ssh
require_cmd scp
require_cmd git
require_cmd awk
require_cmd sed
require_cmd grep
require_cmd mktemp
require_cmd date
require_cmd curl
ensure_file_from_example "${CLUSTER_ENV}"
ensure_file_from_example "${REGISTRY_ENV}"
ensure_file_from_example "${PROD_ENV}"
require_secret_file "${SECRET_POSTGRES}" "Postgres password"
require_secret_file "${SECRET_APP_KEY}" "SECRET_KEY"
require_secret_file "${SECRET_EMAIL_PASS}" "SMTP password"
require_secret_file "${SECRET_FCM_KEY}" "FCM server key"
require_secret_file "${SECRET_APNS_KEY}" "APNS private key"
set -a
# shellcheck disable=SC1090
source "${CLUSTER_ENV}"
# shellcheck disable=SC1090
source "${REGISTRY_ENV}"
# shellcheck disable=SC1090
source "${PROD_ENV}"
set +a
DEPLOY_MANAGER_SSH_PORT="${DEPLOY_MANAGER_SSH_PORT:-22}"
DEPLOY_STACK_NAME="${DEPLOY_STACK_NAME:-casera}"
DEPLOY_REMOTE_DIR="${DEPLOY_REMOTE_DIR:-/opt/casera/deploy}"
DEPLOY_WAIT_SECONDS="${DEPLOY_WAIT_SECONDS:-420}"
DEPLOY_TAG="${DEPLOY_TAG:-$(git -C "${REPO_DIR}" rev-parse --short HEAD)}"
PUSH_LATEST_TAG="${PUSH_LATEST_TAG:-true}"
require_var DEPLOY_MANAGER_HOST
require_var DEPLOY_MANAGER_USER
require_var DEPLOY_STACK_NAME
require_var DEPLOY_REMOTE_DIR
require_var REGISTRY
require_var REGISTRY_NAMESPACE
require_var REGISTRY_USERNAME
require_var REGISTRY_TOKEN
require_var ALLOWED_HOSTS
require_var CORS_ALLOWED_ORIGINS
require_var BASE_URL
require_var NEXT_PUBLIC_API_URL
require_var DB_HOST
require_var DB_PORT
require_var POSTGRES_USER
require_var POSTGRES_DB
require_var DB_SSLMODE
require_var REDIS_URL
require_var EMAIL_HOST
require_var EMAIL_PORT
require_var EMAIL_HOST_USER
require_var DEFAULT_FROM_EMAIL
require_var APNS_AUTH_KEY_ID
require_var APNS_TEAM_ID
require_var APNS_TOPIC
if [[ ! "$(tr -d '\r\n' < "${SECRET_APNS_KEY}")" =~ BEGIN[[:space:]]+PRIVATE[[:space:]]+KEY ]]; then
die "APNS key file does not look like a private key: ${SECRET_APNS_KEY}"
fi
app_secret_len="$(tr -d '\r\n' < "${SECRET_APP_KEY}" | wc -c | tr -d ' ')"
if (( app_secret_len < 32 )); then
die "deploy/secrets/secret_key.txt must be at least 32 characters."
fi
REGISTRY_PREFIX="${REGISTRY%/}/${REGISTRY_NAMESPACE#/}"
API_IMAGE="${REGISTRY_PREFIX}/casera-api:${DEPLOY_TAG}"
WORKER_IMAGE="${REGISTRY_PREFIX}/casera-worker:${DEPLOY_TAG}"
ADMIN_IMAGE="${REGISTRY_PREFIX}/casera-admin:${DEPLOY_TAG}"
SSH_KEY_PATH="${DEPLOY_SSH_KEY_PATH:-}"
if [[ -n "${SSH_KEY_PATH}" ]]; then
SSH_KEY_PATH="${SSH_KEY_PATH/#\~/${HOME}}"
fi
SSH_TARGET="${DEPLOY_MANAGER_USER}@${DEPLOY_MANAGER_HOST}"
SSH_OPTS=(-p "${DEPLOY_MANAGER_SSH_PORT}")
SCP_OPTS=(-P "${DEPLOY_MANAGER_SSH_PORT}")
if [[ -n "${SSH_KEY_PATH}" ]]; then
SSH_OPTS+=(-i "${SSH_KEY_PATH}")
SCP_OPTS+=(-i "${SSH_KEY_PATH}")
fi
log "Validating SSH access to ${SSH_TARGET}"
if ! ssh "${SSH_OPTS[@]}" "${SSH_TARGET}" "echo ok" >/dev/null 2>&1; then
die "SSH connection failed to ${SSH_TARGET}"
fi
remote_swarm_state="$(ssh "${SSH_OPTS[@]}" "${SSH_TARGET}" "docker info --format '{{.Swarm.LocalNodeState}} {{.Swarm.ControlAvailable}}'" 2>/dev/null || true)"
if [[ -z "${remote_swarm_state}" ]]; then
die "Could not read Docker Swarm state on manager. Is Docker installed/running?"
fi
if [[ "${remote_swarm_state}" != "active true" ]]; then
die "Remote node must be a Swarm manager. Got: ${remote_swarm_state}"
fi
if [[ "${SKIP_BUILD}" != "1" ]]; then
log "Logging in to ${REGISTRY}"
printf '%s' "${REGISTRY_TOKEN}" | docker login "${REGISTRY}" -u "${REGISTRY_USERNAME}" --password-stdin >/dev/null
log "Building API image ${API_IMAGE}"
docker build --target api -t "${API_IMAGE}" "${REPO_DIR}"
log "Building Worker image ${WORKER_IMAGE}"
docker build --target worker -t "${WORKER_IMAGE}" "${REPO_DIR}"
log "Building Admin image ${ADMIN_IMAGE}"
docker build --target admin -t "${ADMIN_IMAGE}" "${REPO_DIR}"
log "Pushing deploy images"
docker push "${API_IMAGE}"
docker push "${WORKER_IMAGE}"
docker push "${ADMIN_IMAGE}"
if [[ "${PUSH_LATEST_TAG}" == "true" ]]; then
log "Updating :latest tags"
docker tag "${API_IMAGE}" "${REGISTRY_PREFIX}/casera-api:latest"
docker tag "${WORKER_IMAGE}" "${REGISTRY_PREFIX}/casera-worker:latest"
docker tag "${ADMIN_IMAGE}" "${REGISTRY_PREFIX}/casera-admin:latest"
docker push "${REGISTRY_PREFIX}/casera-api:latest"
docker push "${REGISTRY_PREFIX}/casera-worker:latest"
docker push "${REGISTRY_PREFIX}/casera-admin:latest"
fi
else
warn "SKIP_BUILD=1 set. Using prebuilt images for tag: ${DEPLOY_TAG}"
fi
DEPLOY_ID_RAW="${DEPLOY_TAG}-$(date +%Y%m%d%H%M%S)"
DEPLOY_ID="$(printf '%s' "${DEPLOY_ID_RAW}" | tr -c 'a-zA-Z0-9_-' '_')"
POSTGRES_PASSWORD_SECRET="${DEPLOY_STACK_NAME}_postgres_password_${DEPLOY_ID}"
SECRET_KEY_SECRET="${DEPLOY_STACK_NAME}_secret_key_${DEPLOY_ID}"
EMAIL_HOST_PASSWORD_SECRET="${DEPLOY_STACK_NAME}_email_host_password_${DEPLOY_ID}"
FCM_SERVER_KEY_SECRET="${DEPLOY_STACK_NAME}_fcm_server_key_${DEPLOY_ID}"
APNS_AUTH_KEY_SECRET="${DEPLOY_STACK_NAME}_apns_auth_key_${DEPLOY_ID}"
TMP_DIR="$(mktemp -d)"
cleanup() {
rm -rf "${TMP_DIR}"
}
trap cleanup EXIT
cp "${STACK_TEMPLATE}" "${TMP_DIR}/swarm-stack.prod.yml"
cp "${PROD_ENV}" "${TMP_DIR}/prod.env"
cp "${REGISTRY_ENV}" "${TMP_DIR}/registry.env"
mkdir -p "${TMP_DIR}/secrets"
cp "${SECRET_POSTGRES}" "${TMP_DIR}/secrets/postgres_password.txt"
cp "${SECRET_APP_KEY}" "${TMP_DIR}/secrets/secret_key.txt"
cp "${SECRET_EMAIL_PASS}" "${TMP_DIR}/secrets/email_host_password.txt"
cp "${SECRET_FCM_KEY}" "${TMP_DIR}/secrets/fcm_server_key.txt"
cp "${SECRET_APNS_KEY}" "${TMP_DIR}/secrets/apns_auth_key.p8"
cat > "${TMP_DIR}/runtime.env" <<EOF
API_IMAGE=${API_IMAGE}
WORKER_IMAGE=${WORKER_IMAGE}
ADMIN_IMAGE=${ADMIN_IMAGE}
API_REPLICAS=${API_REPLICAS:-3}
WORKER_REPLICAS=${WORKER_REPLICAS:-2}
ADMIN_REPLICAS=${ADMIN_REPLICAS:-1}
API_PORT=${API_PORT:-8000}
ADMIN_PORT=${ADMIN_PORT:-3000}
DOZZLE_PORT=${DOZZLE_PORT:-9999}
POSTGRES_PASSWORD_SECRET=${POSTGRES_PASSWORD_SECRET}
SECRET_KEY_SECRET=${SECRET_KEY_SECRET}
EMAIL_HOST_PASSWORD_SECRET=${EMAIL_HOST_PASSWORD_SECRET}
FCM_SERVER_KEY_SECRET=${FCM_SERVER_KEY_SECRET}
APNS_AUTH_KEY_SECRET=${APNS_AUTH_KEY_SECRET}
EOF
log "Uploading deploy bundle to ${SSH_TARGET}:${DEPLOY_REMOTE_DIR}"
ssh "${SSH_OPTS[@]}" "${SSH_TARGET}" "mkdir -p '${DEPLOY_REMOTE_DIR}/secrets'"
scp "${SCP_OPTS[@]}" "${TMP_DIR}/swarm-stack.prod.yml" "${SSH_TARGET}:${DEPLOY_REMOTE_DIR}/swarm-stack.prod.yml"
scp "${SCP_OPTS[@]}" "${TMP_DIR}/prod.env" "${SSH_TARGET}:${DEPLOY_REMOTE_DIR}/prod.env"
scp "${SCP_OPTS[@]}" "${TMP_DIR}/registry.env" "${SSH_TARGET}:${DEPLOY_REMOTE_DIR}/registry.env"
scp "${SCP_OPTS[@]}" "${TMP_DIR}/runtime.env" "${SSH_TARGET}:${DEPLOY_REMOTE_DIR}/runtime.env"
scp "${SCP_OPTS[@]}" "${TMP_DIR}/secrets/postgres_password.txt" "${SSH_TARGET}:${DEPLOY_REMOTE_DIR}/secrets/postgres_password.txt"
scp "${SCP_OPTS[@]}" "${TMP_DIR}/secrets/secret_key.txt" "${SSH_TARGET}:${DEPLOY_REMOTE_DIR}/secrets/secret_key.txt"
scp "${SCP_OPTS[@]}" "${TMP_DIR}/secrets/email_host_password.txt" "${SSH_TARGET}:${DEPLOY_REMOTE_DIR}/secrets/email_host_password.txt"
scp "${SCP_OPTS[@]}" "${TMP_DIR}/secrets/fcm_server_key.txt" "${SSH_TARGET}:${DEPLOY_REMOTE_DIR}/secrets/fcm_server_key.txt"
scp "${SCP_OPTS[@]}" "${TMP_DIR}/secrets/apns_auth_key.p8" "${SSH_TARGET}:${DEPLOY_REMOTE_DIR}/secrets/apns_auth_key.p8"
log "Creating Docker secrets and deploying stack on manager"
ssh "${SSH_OPTS[@]}" "${SSH_TARGET}" "bash -s -- '${DEPLOY_REMOTE_DIR}' '${DEPLOY_STACK_NAME}'" <<'EOF'
set -euo pipefail
REMOTE_DIR="$1"
STACK_NAME="$2"
set -a
# shellcheck disable=SC1090
source "${REMOTE_DIR}/registry.env"
# shellcheck disable=SC1090
source "${REMOTE_DIR}/prod.env"
# shellcheck disable=SC1090
source "${REMOTE_DIR}/runtime.env"
set +a
create_secret() {
local name="$1"
local src="$2"
if docker secret inspect "${name}" >/dev/null 2>&1; then
echo "[remote] secret exists: ${name}"
else
docker secret create "${name}" "${src}" >/dev/null
echo "[remote] created secret: ${name}"
fi
}
printf '%s' "${REGISTRY_TOKEN}" | docker login "${REGISTRY}" -u "${REGISTRY_USERNAME}" --password-stdin >/dev/null
rm -f "${REMOTE_DIR}/registry.env"
create_secret "${POSTGRES_PASSWORD_SECRET}" "${REMOTE_DIR}/secrets/postgres_password.txt"
create_secret "${SECRET_KEY_SECRET}" "${REMOTE_DIR}/secrets/secret_key.txt"
create_secret "${EMAIL_HOST_PASSWORD_SECRET}" "${REMOTE_DIR}/secrets/email_host_password.txt"
create_secret "${FCM_SERVER_KEY_SECRET}" "${REMOTE_DIR}/secrets/fcm_server_key.txt"
create_secret "${APNS_AUTH_KEY_SECRET}" "${REMOTE_DIR}/secrets/apns_auth_key.p8"
rm -f "${REMOTE_DIR}/secrets/postgres_password.txt"
rm -f "${REMOTE_DIR}/secrets/secret_key.txt"
rm -f "${REMOTE_DIR}/secrets/email_host_password.txt"
rm -f "${REMOTE_DIR}/secrets/fcm_server_key.txt"
rm -f "${REMOTE_DIR}/secrets/apns_auth_key.p8"
set -a
# shellcheck disable=SC1090
source "${REMOTE_DIR}/prod.env"
# shellcheck disable=SC1090
source "${REMOTE_DIR}/runtime.env"
set +a
docker stack deploy --with-registry-auth -c "${REMOTE_DIR}/swarm-stack.prod.yml" "${STACK_NAME}"
EOF
log "Waiting for stack convergence (${DEPLOY_WAIT_SECONDS}s max)"
start_epoch="$(date +%s)"
while true; do
services="$(ssh "${SSH_OPTS[@]}" "${SSH_TARGET}" "docker stack services '${DEPLOY_STACK_NAME}' --format '{{.Name}} {{.Replicas}}'" 2>/dev/null || true)"
if [[ -n "${services}" ]]; then
all_ready=1
while IFS=' ' read -r svc replicas; do
[[ -z "${svc}" ]] && continue
current="${replicas%%/*}"
desired="${replicas##*/}"
if [[ "${desired}" == "0" ]]; then
continue
fi
if [[ "${current}" != "${desired}" ]]; then
all_ready=0
fi
done <<< "${services}"
if [[ "${all_ready}" -eq 1 ]]; then
break
fi
fi
now_epoch="$(date +%s)"
elapsed=$((now_epoch - start_epoch))
if (( elapsed >= DEPLOY_WAIT_SECONDS )); then
die "Timed out waiting for stack to converge. Check: ssh ${SSH_TARGET} docker stack services ${DEPLOY_STACK_NAME}"
fi
sleep 10
done
if [[ "${SKIP_HEALTHCHECK}" != "1" && -n "${DEPLOY_HEALTHCHECK_URL:-}" ]]; then
log "Running health check: ${DEPLOY_HEALTHCHECK_URL}"
curl -fsS --max-time 20 "${DEPLOY_HEALTHCHECK_URL}" >/dev/null
fi
log "Deploy completed successfully."
log "Stack: ${DEPLOY_STACK_NAME}"
log "Images:"
log " ${API_IMAGE}"
log " ${WORKER_IMAGE}"
log " ${ADMIN_IMAGE}"