archy/scripts/test-stability-72h.sh

#!/usr/bin/env bash
# FINAL-202: 72-Hour Stability Test
# Monitors a running Archipelago node for 72 hours, checking health every 5 minutes.
# Usage: bash test-stability-72h.sh <node-ip> [password]
# Logs results to /tmp/stability-test-<timestamp>.log

set -euo pipefail

NODE="${1:-192.168.1.228}"
BASE="http://${NODE}"
PASS="${2:-password123}"
DURATION_HOURS="${3:-72}"
CHECK_INTERVAL=300  # 5 minutes
COOKIE_JAR="/tmp/stability-cookies.txt"
TIMESTAMP=$(date +%Y%m%d-%H%M%S)
LOG_FILE="/tmp/stability-test-${TIMESTAMP}.log"
FAIL_LOG="/tmp/stability-failures-${TIMESTAMP}.log"

TOTAL_CHECKS=0
TOTAL_FAILURES=0
CONSECUTIVE_FAILURES=0
MAX_CONSECUTIVE=0
START_TIME=$(date +%s)
END_TIME=$((START_TIME + DURATION_HOURS * 3600))

log() { echo "[$(date '+%Y-%m-%d %H:%M:%S')] $*" | tee -a "$LOG_FILE"; }
fail_log() { echo "[$(date '+%Y-%m-%d %H:%M:%S')] FAIL: $*" | tee -a "$LOG_FILE" >> "$FAIL_LOG"; }

login() {
  curl -s -c "$COOKIE_JAR" -H "Content-Type: application/json" \
    -d "{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"auth.login\",\"params\":{\"password\":\"$PASS\"}}" \
    "${BASE}/rpc/" > /dev/null 2>&1
}

rpc() {
  curl -s -m 10 -b "$COOKIE_JAR" -c "$COOKIE_JAR" \
    -H "Content-Type: application/json" \
    -d "{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"$1\",\"params\":${2:-{}}}" \
    "${BASE}/rpc/" 2>/dev/null
}

check_health() {
  local failures=0

  # 1. Backend health
  local health_code
  health_code=$(curl -s -o /dev/null -w "%{http_code}" -m 10 "${BASE}/health" 2>/dev/null || echo "000")
  if [ "$health_code" != "200" ]; then
    fail_log "Backend health endpoint returned $health_code"
    failures=$((failures + 1))
  fi

  # 2. UI loads
  local ui_code
  ui_code=$(curl -s -o /dev/null -w "%{http_code}" -m 10 "${BASE}/" 2>/dev/null || echo "000")
  if [ "$ui_code" != "200" ] && [ "$ui_code" != "302" ]; then
    fail_log "Web UI returned $ui_code"
    failures=$((failures + 1))
  fi

  # 3. RPC responds
  local rpc_resp
  rpc_resp=$(rpc "system.info" 2>/dev/null)
  if ! echo "$rpc_resp" | grep -q '"result"'; then
    # Try re-login
    login
    rpc_resp=$(rpc "system.info" 2>/dev/null)
    if ! echo "$rpc_resp" | grep -q '"result"'; then
      fail_log "RPC system.info failed after re-login"
      failures=$((failures + 1))
    fi
  fi

  # 4. WebSocket endpoint
  local ws_code
  ws_code=$(curl -s -o /dev/null -w "%{http_code}" -m 5 -H "Upgrade: websocket" "${BASE}/ws/" 2>/dev/null || echo "000")
  if [ "$ws_code" = "000" ]; then
    fail_log "WebSocket endpoint unreachable"
    failures=$((failures + 1))
  fi

  # 5. Check containers via SSH (if accessible)
  local ssh_key="$HOME/.ssh/archipelago-deploy"
  if [ -f "$ssh_key" ]; then
    local crashed
    crashed=$(ssh -i "$ssh_key" -o ConnectTimeout=10 -o StrictHostKeyChecking=no "archipelago@${NODE}" \
      'sudo podman ps -a --format "{{.Names}} {{.Status}}" 2>/dev/null | grep -i "exited\|dead\|oom" | head -5' 2>/dev/null || echo "")
    if [ -n "$crashed" ]; then
      fail_log "Crashed/dead containers: $crashed"
      failures=$((failures + 1))
    fi

    # 6. Check memory usage
    local mem_info
    mem_info=$(ssh -i "$ssh_key" -o ConnectTimeout=10 -o StrictHostKeyChecking=no "archipelago@${NODE}" \
      'free -m | grep Mem | awk "{printf \"%d/%dMB (%.0f%%)\", \$3, \$2, \$3/\$2*100}"' 2>/dev/null || echo "unknown")
    log "  Memory: $mem_info"

    # 7. Check disk usage
    local disk_info
    disk_info=$(ssh -i "$ssh_key" -o ConnectTimeout=10 -o StrictHostKeyChecking=no "archipelago@${NODE}" \
      'df -h / | tail -1 | awk "{print \$3\"/\"\$2\" (\"\$5\" used)\"}"' 2>/dev/null || echo "unknown")
    log "  Disk: $disk_info"

    # 8. Check for OOM kills since start
    local oom_count
    oom_count=$(ssh -i "$ssh_key" -o ConnectTimeout=10 -o StrictHostKeyChecking=no "archipelago@${NODE}" \
      'dmesg 2>/dev/null | grep -c "Out of memory" || echo 0' 2>/dev/null || echo "unknown")
    if [ "$oom_count" != "0" ] && [ "$oom_count" != "unknown" ]; then
      fail_log "OOM kills detected: $oom_count"
      failures=$((failures + 1))
    fi

    # 9. Check archipelago service
    local svc_status
    svc_status=$(ssh -i "$ssh_key" -o ConnectTimeout=10 -o StrictHostKeyChecking=no "archipelago@${NODE}" \
      'systemctl is-active archipelago 2>/dev/null || echo inactive' 2>/dev/null || echo "unknown")
    if [ "$svc_status" != "active" ]; then
      fail_log "Archipelago service status: $svc_status"
      failures=$((failures + 1))
    fi
  fi

  # 10. Check Tor services
  local tor_resp
  tor_resp=$(rpc "tor.list-services" 2>/dev/null)
  if echo "$tor_resp" | grep -q '"result"'; then
    local tor_count
    tor_count=$(echo "$tor_resp" | python3 -c "import sys,json; r=json.load(sys.stdin); print(len(r.get('result',{}).get('services',[])))" 2>/dev/null || echo "0")
    log "  Tor services: $tor_count"
  fi

  # 11. Check peer connections
  local peers_resp
  peers_resp=$(rpc "network.list-peers" 2>/dev/null)
  if echo "$peers_resp" | grep -q '"result"'; then
    local peer_count
    peer_count=$(echo "$peers_resp" | python3 -c "import sys,json; r=json.load(sys.stdin); print(len(r.get('result',{}).get('peers',[])))" 2>/dev/null || echo "0")
    log "  Connected peers: $peer_count"
  fi

  # 12. Ecash wallet balance check
  local ecash_resp
  ecash_resp=$(rpc "wallet.ecash-balance" 2>/dev/null)
  if echo "$ecash_resp" | grep -q '"result"'; then
    local balance
    balance=$(echo "$ecash_resp" | python3 -c "import sys,json; r=json.load(sys.stdin); print(r.get('result',{}).get('balance',0))" 2>/dev/null || echo "0")
    log "  Ecash balance: $balance sats"
  fi

  return $failures
}

# ─── Main Loop ────────────────────────────────────────────────────
log "╔════════════════════════════════════════════════════════════════╗"
log "║  72-Hour Stability Test — Archipelago                        ║"
log "╚════════════════════════════════════════════════════════════════╝"
log "Target: $NODE"
log "Duration: ${DURATION_HOURS}h (until $(date -r $END_TIME '+%Y-%m-%d %H:%M:%S' 2>/dev/null || date -d @$END_TIME '+%Y-%m-%d %H:%M:%S' 2>/dev/null || echo 'unknown'))"
log "Check interval: ${CHECK_INTERVAL}s"
log "Log file: $LOG_FILE"
log "Failure log: $FAIL_LOG"
log ""

# Initial login
login
log "Authenticated to node"

while [ "$(date +%s)" -lt "$END_TIME" ]; do
  TOTAL_CHECKS=$((TOTAL_CHECKS + 1))
  ELAPSED_H=$(( ($(date +%s) - START_TIME) / 3600 ))
  ELAPSED_M=$(( (($(date +%s) - START_TIME) % 3600) / 60 ))

  log "Check #${TOTAL_CHECKS} (${ELAPSED_H}h${ELAPSED_M}m elapsed)"

  if check_health; then
    CONSECUTIVE_FAILURES=0
    log "  Status: OK"
  else
    FAIL_RESULT=$?
    TOTAL_FAILURES=$((TOTAL_FAILURES + FAIL_RESULT))
    CONSECUTIVE_FAILURES=$((CONSECUTIVE_FAILURES + 1))
    if [ "$CONSECUTIVE_FAILURES" -gt "$MAX_CONSECUTIVE" ]; then
      MAX_CONSECUTIVE=$CONSECUTIVE_FAILURES
    fi
    log "  Status: $FAIL_RESULT failure(s) this check"

    if [ "$CONSECUTIVE_FAILURES" -ge 5 ]; then
      log "WARNING: 5 consecutive check failures — node may be down!"
    fi
  fi

  sleep "$CHECK_INTERVAL"
done

# ─── Final Report ─────────────────────────────────────────────────
log ""
log "╔════════════════════════════════════════════════════════════════╗"
log "║  72-Hour Stability Test — COMPLETE                           ║"
log "╚════════════════════════════════════════════════════════════════╝"
log ""
log "Duration: ${DURATION_HOURS}h"
log "Total checks: $TOTAL_CHECKS"
log "Total failures: $TOTAL_FAILURES"
log "Max consecutive failures: $MAX_CONSECUTIVE"
log ""

UPTIME_PCT=0
if [ "$TOTAL_CHECKS" -gt 0 ]; then
  PASSED=$((TOTAL_CHECKS - TOTAL_FAILURES))
  UPTIME_PCT=$(python3 -c "print(f'{${PASSED}/${TOTAL_CHECKS}*100:.1f}')" 2>/dev/null || echo "?")
fi
log "Uptime: ${UPTIME_PCT}%"

if [ "$TOTAL_FAILURES" -eq 0 ]; then
  log "RESULT: PASS — Zero failures over ${DURATION_HOURS}h"
  exit 0
else
  log "RESULT: FAIL — $TOTAL_FAILURES failures detected"
  log "See failure details: $FAIL_LOG"
  exit 1
fi