#!/bin/bash
# cdp-latency-check.sh — measure CDP relay latency per node.
#
# Probes each node's host-side relay (PEER_IP:CDP_PORT from netvm-names.sh,
# which pins the registry ports) at /json/version and prints one line per
# node:
#   name:latency_ms:code
# On failure (timeout / connection refused / non-200-ish transport error):
#   name:FAIL:000
#
# Self-contained: only needs netvm-names.sh in the same directory and curl.
set -u

BIN_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
# shellcheck disable=SC1091
source "$BIN_DIR/netvm-names.sh"

TIMEOUT_S=10

# Registry-driven node list (was hardcoded 4 nodes; def/dev had no
# cdp-latency coverage — 2026-10-06).
watched_nodes() {
  "$BIN_DIR/netvm-registry.py" 2>/dev/null | cut -d: -f1
}

# Allow sourcing for tests without running checks.
if [ "${CDP_LATENCY_CHECK_LIB_ONLY:-}" = "1" ]; then
  return 0 2>/dev/null || exit 0
fi

NODES="$(watched_nodes)"
if [ -z "$NODES" ]; then
  echo "node registry empty/unreadable" >&2
  exit 1
fi

# shellcheck disable=SC2086 (intended word splitting: one node per word)
for node in $NODES; do
  netvm_names "$node"
  url="http://${PEER_IP}:${CDP_PORT}/json/version"
  probe=$(curl -s -m "$TIMEOUT_S" -o /dev/null -w "%{time_total} %{http_code}" "$url" 2>/dev/null)
  rc=$?
  secs=$(printf '%s' "$probe" | awk '{print $1}')
  code=$(printf '%s' "$probe" | awk '{print $2}')
  if [ "$rc" -ne 0 ] || [ -z "$code" ] || [ "$code" = "000" ]; then
    echo "${node}:FAIL:000"
    continue
  fi
  ms=$(awk -v s="$secs" 'BEGIN { printf "%d", (s + 0) * 1000 }')
  echo "${node}:${ms}:${code}"
done