fix: bug: edge-control add_route targets non-existent Caddy server edge — registration succeeds in registry but traffic never routes (#789)

- install.sh: use Caddy `servers { name edge }` global option so the
  emitted Caddyfile produces a predictably-named server
- lib/caddy.sh: add `_discover_server_name` that queries the admin API
  for the first server listening on :80/:443 — add_route and remove_route
  use dynamic discovery instead of hardcoding `/servers/edge/`
- lib/caddy.sh: add_route, remove_route, and reload_caddy now check HTTP
  status codes (≥400 → return 1 with error message) instead of only
  checking curl exit code

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

tools/edge-control/install.sh

@@ -225,13 +225,19 @@ EOF
 chmod 600 "$GANDI_ENV"
 
 # Create Caddyfile with admin API and wildcard cert
+# The "servers" global option names the auto-generated server "edge" so that
+# lib/caddy.sh (which discovers the server dynamically) finds a predictable
+# name — defense-in-depth alongside the dynamic discovery in add_route.
 CADDYFILE="/etc/caddy/Caddyfile"
-cat > "$CADDYFILE" <<EOF
+cat > "$CADDYFILE" <<'CADDYEOF'
 # Caddy configuration for edge control plane
 # Admin API enabled on 127.0.0.1:2019
 
 {
   admin localhost:2019
+  servers {
+    name edge
+  }
 }
 
 # Default site (reverse proxy for edge tunnels will be added dynamically)
@@ -240,7 +246,7 @@ cat > "$CADDYFILE" <<EOF
     dns gandi {env.GANDI_API_KEY}
   }
 }
-EOF
+CADDYEOF
 
 # Start Caddy
 systemctl restart caddy 2>/dev/null || {

tools/edge-control/lib/caddy.sh

@@ -19,6 +19,24 @@ CADDY_ADMIN_URL="${CADDY_ADMIN_URL:-http://127.0.0.1:2019}"
 # Domain suffix for projects
 DOMAIN_SUFFIX="${DOMAIN_SUFFIX:-disinto.ai}"
 
+# Discover the Caddy server name that listens on :80/:443
+# Usage: _discover_server_name
+_discover_server_name() {
+  local server_name
+  server_name=$(curl -sS "${CADDY_ADMIN_URL}/config/apps/http/servers" \
+    | jq -r 'to_entries | map(select(.value.listen[]? | test(":(80|443)$"))) | .[0].key // empty') || {
+    echo "Error: could not query Caddy admin API for servers" >&2
+    return 1
+  }
+
+  if [ -z "$server_name" ]; then
+    echo "Error: could not find a Caddy server listening on :80/:443" >&2
+    return 1
+  fi
+
+  echo "$server_name"
+}
+
 # Add a route for a project
 # Usage: add_route <project> <port>
 add_route() {
@@ -26,6 +44,9 @@ add_route() {
   local port="$2"
   local fqdn="${project}.${DOMAIN_SUFFIX}"
 
+  local server_name
+  server_name=$(_discover_server_name) || return 1
+
   # Build the route configuration (partial config)
   local route_config
   route_config=$(cat <<EOF
@@ -58,16 +79,21 @@ add_route() {
 EOF
 )
 
-  # Append route using POST /config/apps/http/servers/edge/routes
+  # Append route via admin API, checking HTTP status
-  local response
+  local response status body
-  response=$(curl -s -X POST \
+  response=$(curl -sS -w '\n%{http_code}' -X POST \
-    "${CADDY_ADMIN_URL}/config/apps/http/servers/edge/routes" \
+    "${CADDY_ADMIN_URL}/config/apps/http/servers/${server_name}/routes" \
     -H "Content-Type: application/json" \
-    -d "$route_config" 2>&1) || {
+    -d "$route_config") || {
     echo "Error: failed to add route for ${fqdn}" >&2
-    echo "Response: ${response}" >&2
     return 1
   }
+  status=$(echo "$response" | tail -n1)
+  body=$(echo "$response" | sed '$d')
+  if [ "$status" -ge 400 ]; then
+    echo "Error: Caddy admin API returned ${status}: ${body}" >&2
+    return 1
+  fi
 
   echo "Added route: ${fqdn} → 127.0.0.1:${port}" >&2
 }
@@ -78,31 +104,45 @@ remove_route() {
   local project="$1"
   local fqdn="${project}.${DOMAIN_SUFFIX}"
 
-  # First, get current routes
+  local server_name
-  local routes_json
+  server_name=$(_discover_server_name) || return 1
-  routes_json=$(curl -s "${CADDY_ADMIN_URL}/config/apps/http/servers/edge/routes" 2>&1) || {
+
+  # First, get current routes, checking HTTP status
+  local response status body
+  response=$(curl -sS -w '\n%{http_code}' \
+    "${CADDY_ADMIN_URL}/config/apps/http/servers/${server_name}/routes") || {
     echo "Error: failed to get current routes" >&2
     return 1
   }
+  status=$(echo "$response" | tail -n1)
+  body=$(echo "$response" | sed '$d')
+  if [ "$status" -ge 400 ]; then
+    echo "Error: Caddy admin API returned ${status}: ${body}" >&2
+    return 1
+  fi
 
   # Find the route index that matches our fqdn using jq
   local route_index
-  route_index=$(echo "$routes_json" | jq -r "to_entries[] | select(.value.match[]?.host[]? == \"${fqdn}\") | .key" 2>/dev/null | head -1)
+  route_index=$(echo "$body" | jq -r "to_entries[] | select(.value.match[]?.host[]? == \"${fqdn}\") | .key" 2>/dev/null | head -1)
 
   if [ -z "$route_index" ] || [ "$route_index" = "null" ]; then
     echo "Warning: route for ${fqdn} not found" >&2
     return 0
   fi
 
-  # Delete the route at the found index
+  # Delete the route at the found index, checking HTTP status
-  local response
+  response=$(curl -sS -w '\n%{http_code}' -X DELETE \
-  response=$(curl -s -X DELETE \
+    "${CADDY_ADMIN_URL}/config/apps/http/servers/${server_name}/routes/${route_index}" \
-    "${CADDY_ADMIN_URL}/config/apps/http/servers/edge/routes/${route_index}" \
+    -H "Content-Type: application/json") || {
-    -H "Content-Type: application/json" 2>&1) || {
     echo "Error: failed to remove route for ${fqdn}" >&2
-    echo "Response: ${response}" >&2
     return 1
   }
+  status=$(echo "$response" | tail -n1)
+  body=$(echo "$response" | sed '$d')
+  if [ "$status" -ge 400 ]; then
+    echo "Error: Caddy admin API returned ${status}: ${body}" >&2
+    return 1
+  fi
 
   echo "Removed route: ${fqdn}" >&2
 }
@@ -110,13 +150,18 @@ remove_route() {
 # Reload Caddy to apply configuration changes
 # Usage: reload_caddy
 reload_caddy() {
-  local response
+  local response status body
-  response=$(curl -s -X POST \
+  response=$(curl -sS -w '\n%{http_code}' -X POST \
-    "${CADDY_ADMIN_URL}/reload" 2>&1) || {
+    "${CADDY_ADMIN_URL}/reload") || {
     echo "Error: failed to reload Caddy" >&2
-    echo "Response: ${response}" >&2
     return 1
   }
+  status=$(echo "$response" | tail -n1)
+  body=$(echo "$response" | sed '$d')
+  if [ "$status" -ge 400 ]; then
+    echo "Error: Caddy reload returned ${status}: ${body}" >&2
+    return 1
+  fi
 
   echo "Caddy reloaded" >&2
 }