alt-mysql/scripts/worker.sh

#!/bin/bash
set -euo pipefail

PLUGIN_DIR="${PLUGIN_DIR:-$(cd "$(dirname "$0")/.." && pwd)}"
DATA_DIR="$PLUGIN_DIR/data"
JOB_DIR="$DATA_DIR/jobs/pending"
PROCESSING_DIR="$DATA_DIR/jobs/processing"
DONE_DIR="$DATA_DIR/jobs/done"
WORKER_LOCK_DIR="$DATA_DIR/worker.lock"
DB_LOCK_DIR="$DATA_DIR/lock"
PID_DIR="$DATA_DIR/pids"
CANCEL_DIR="$DATA_DIR/cancel"
WORKER_LOCK_STALE_SECONDS="${WORKER_LOCK_STALE_SECONDS:-300}"

mkdir -p "$JOB_DIR" "$PROCESSING_DIR" "$DONE_DIR" "$PID_DIR" "$CANCEL_DIR" "$DB_LOCK_DIR"

acquire_worker_lock() {
  if mkdir "$WORKER_LOCK_DIR" 2>/dev/null; then
    printf '%s' "$$" > "$WORKER_LOCK_DIR/pid" 2>/dev/null || true
    return 0
  fi

  local existing_pid="" lock_mtime=0 now lock_age=0
  if [ -r "$WORKER_LOCK_DIR/pid" ]; then
    existing_pid="$(cat "$WORKER_LOCK_DIR/pid" 2>/dev/null || true)"
  fi

  if [ -n "$existing_pid" ]; then
    if kill -0 "$existing_pid" 2>/dev/null; then
      echo "alt-mysql worker: lock held by running pid $existing_pid, exiting" >&2
      return 1
    fi
  else
    lock_mtime="$(stat -c '%Y' "$WORKER_LOCK_DIR" 2>/dev/null || stat -f '%m' "$WORKER_LOCK_DIR" 2>/dev/null || echo 0)"
    now="$(date +%s)"
    if [ "$lock_mtime" -gt 0 ]; then
      lock_age=$((now - lock_mtime))
    fi
    if [ "$lock_age" -lt "$WORKER_LOCK_STALE_SECONDS" ]; then
      echo "alt-mysql worker: lock present without a live pid but younger than ${WORKER_LOCK_STALE_SECONDS}s (age ${lock_age}s), exiting" >&2
      return 1
    fi
  fi

  echo "alt-mysql worker: reclaiming stale lock (pid=${existing_pid:-none}, age=${lock_age}s)" >&2
  rm -rf "$WORKER_LOCK_DIR"
  if mkdir "$WORKER_LOCK_DIR" 2>/dev/null; then
    printf '%s' "$$" > "$WORKER_LOCK_DIR/pid" 2>/dev/null || true
    return 0
  fi

  return 1
}

if ! acquire_worker_lock; then
  exit 0
fi

cleanup() {
  rm -rf "$WORKER_LOCK_DIR" 2>/dev/null || true
}
trap cleanup EXIT

db_lock_path_for_db() {
  local db_name="$1"
  local safe
  safe="$(printf '%s' "$db_name" | tr -c 'A-Za-z0-9_.-' '_')"
  printf '%s/%s.lock' "$DB_LOCK_DIR" "$safe"
}

release_db_lock_later() {
  local lock_path="$1"
  if [ -z "$lock_path" ]; then
    return 0
  fi
  (
    sleep 60
    rm -f "$lock_path" 2>/dev/null || true
  ) >/dev/null 2>&1 &
}

while true; do
  JOB_FILE="$(ls -1 "$JOB_DIR"/*.env 2>/dev/null | head -n 1 || true)"
  if [ -z "${JOB_FILE:-}" ]; then
    break
  fi

  BASE_NAME="$(basename "$JOB_FILE")"
  RUN_FILE="$PROCESSING_DIR/$BASE_NAME"
  mv "$JOB_FILE" "$RUN_FILE"
  JOB_ID="${BASE_NAME%.env}"
  PID_FILE="$PID_DIR/${JOB_ID}.pid"

  JOB_TYPE="restore"
  DB_NAME=""
  DB_LOCK_FILE=""
  # shellcheck disable=SC1090
  source "$RUN_FILE" || true
  JOB_TYPE="${JOB_TYPE:-restore}"
  DB_NAME="${DB_NAME:-}"
  DB_LOCK_FILE="${DB_LOCK_FILE:-}"

  LOCK_PATH=""
  if [ -n "$DB_LOCK_FILE" ]; then
    case "$DB_LOCK_FILE" in
      "$DB_LOCK_DIR"/*.lock) LOCK_PATH="$DB_LOCK_FILE" ;;
      *) LOCK_PATH="" ;;
    esac
  fi
  if [ -z "$LOCK_PATH" ] && [ -n "$DB_NAME" ]; then
    LOCK_PATH="$(db_lock_path_for_db "$DB_NAME")"
  fi

  RUNNER="$PLUGIN_DIR/scripts/restore_job.sh"
  if [ "$JOB_TYPE" = "backup" ]; then
    RUNNER="$PLUGIN_DIR/scripts/backup_job.sh"
  fi

  set +e
  "$RUNNER" "$RUN_FILE" &
  RUN_PID=$!
  echo "PID=${RUN_PID}" >> "$RUN_FILE"
  echo "${RUN_PID}" > "$PID_FILE"
  wait "$RUN_PID"
  EXIT_CODE=$?
  set -e

  rm -f "$PID_FILE"

  CANCEL_FLAG="${CANCEL_DIR}/${JOB_ID}.flag"
  if [ -f "$CANCEL_FLAG" ]; then
    rm -f "$CANCEL_FLAG"
    mv "$RUN_FILE" "$DONE_DIR/${JOB_ID}.cancel"
  elif [ "$EXIT_CODE" -eq 0 ]; then
    mv "$RUN_FILE" "$DONE_DIR/${JOB_ID}.ok"
  else
    mv "$RUN_FILE" "$DONE_DIR/${JOB_ID}.fail"
  fi

  release_db_lock_later "$LOCK_PATH"
done

exit 0