DocDoeAI / scripts /database_backup.py
asnannp's picture
Deploy backend cd4237ff: support routes + rate limit + exam_date nullable + upload 413 fix
7c6ffa6
Raw
History Blame Contribute Delete
8.19 kB
#!/usr/bin/env python3
"""Create and verify a PostgreSQL backup without exposing credentials.
The script is designed for a scheduler or managed job runner. It writes to a
temporary file, verifies the archive with pg_restore, records a checksum-only
manifest, then atomically promotes the backup. Credentials are passed through
PG* environment variables and never included in the command line or output.
Examples:
python scripts/database_backup.py --dry-run
python scripts/database_backup.py --output-dir D:\\docdoe-backups
"""
from __future__ import annotations
import argparse
import hashlib
import json
import os
import shutil
import subprocess
import sys
from dataclasses import dataclass
from datetime import datetime, timedelta, timezone
from pathlib import Path
from urllib.parse import parse_qs, unquote, urlsplit
from dotenv import load_dotenv
BACKEND_DIR = Path(__file__).resolve().parents[1]
DEFAULT_OUTPUT_DIR = BACKEND_DIR / "backups" / "database"
BACKUP_PREFIX = "docdoe-"
@dataclass(frozen=True)
class PostgresTarget:
host: str
port: int
database: str
username: str
password: str
sslmode: str | None
def process_environment(self) -> dict[str, str]:
env = os.environ.copy()
env.update(
{
"PGHOST": self.host,
"PGPORT": str(self.port),
"PGDATABASE": self.database,
"PGUSER": self.username,
"PGPASSWORD": self.password,
"PGCONNECT_TIMEOUT": env.get("PGCONNECT_TIMEOUT", "15"),
}
)
if self.sslmode:
env["PGSSLMODE"] = self.sslmode
return env
def parse_postgres_url(database_url: str) -> PostgresTarget:
raw = database_url.strip()
if not raw:
raise ValueError("DATABASE_URL is missing")
normalized = raw.replace("postgresql+psycopg://", "postgresql://", 1)
parsed = urlsplit(normalized)
if parsed.scheme not in {"postgres", "postgresql"}:
raise ValueError("Backups require a PostgreSQL DATABASE_URL")
if not parsed.hostname or not parsed.username:
raise ValueError("DATABASE_URL must include a host and username")
database = unquote(parsed.path.lstrip("/"))
if not database:
raise ValueError("DATABASE_URL must include a database name")
query = parse_qs(parsed.query)
sslmode = query.get("sslmode", [None])[0]
return PostgresTarget(
host=parsed.hostname,
port=parsed.port or 5432,
database=database,
username=unquote(parsed.username),
password=unquote(parsed.password or ""),
sslmode=sslmode,
)
def backup_filename(now: datetime) -> str:
stamp = now.astimezone(timezone.utc).strftime("%Y%m%dT%H%M%SZ")
return f"{BACKUP_PREFIX}{stamp}.dump"
def file_sha256(path: Path) -> str:
digest = hashlib.sha256()
with path.open("rb") as handle:
for chunk in iter(lambda: handle.read(1024 * 1024), b""):
digest.update(chunk)
return digest.hexdigest()
def prune_expired_backups(output_dir: Path, retention_days: int, now: datetime) -> list[Path]:
if retention_days < 1:
raise ValueError("Retention must be at least one day")
if not output_dir.exists():
return []
cutoff = now.astimezone(timezone.utc) - timedelta(days=retention_days)
removed: list[Path] = []
for backup in output_dir.glob(f"{BACKUP_PREFIX}*.dump"):
modified = datetime.fromtimestamp(backup.stat().st_mtime, tz=timezone.utc)
if modified >= cutoff:
continue
manifest = backup.with_suffix(".manifest.json")
backup.unlink()
removed.append(backup)
if manifest.exists():
manifest.unlink()
removed.append(manifest)
return removed
def _resolve_binary(name: str, dry_run: bool) -> str:
resolved = shutil.which(name)
if resolved:
return resolved
if dry_run:
return name
raise RuntimeError(f"{name} is not installed or not available on PATH")
def _run(command: list[str], env: dict[str, str], *, stdout=None) -> None:
completed = subprocess.run(
command,
env=env,
stdout=stdout,
stderr=subprocess.PIPE,
text=True,
check=False,
)
if completed.returncode != 0:
detail = (completed.stderr or "unknown error").strip().splitlines()[-1]
raise RuntimeError(f"Database backup command failed: {detail[:240]}")
def create_backup(
*,
database_url: str,
output_dir: Path,
retention_days: int,
verify: bool,
dry_run: bool,
now: datetime | None = None,
) -> Path | None:
target = parse_postgres_url(database_url)
moment = now or datetime.now(timezone.utc)
output_dir = output_dir.expanduser().resolve()
final_path = output_dir / backup_filename(moment)
partial_path = final_path.with_suffix(".dump.partial")
pg_dump = _resolve_binary("pg_dump", dry_run)
pg_restore = _resolve_binary("pg_restore", dry_run) if verify else None
if dry_run:
print(
f"[DRY RUN] PostgreSQL backup: {target.host}:{target.port}/{target.database} "
f"-> {final_path} (verify={verify}, retention={retention_days}d)"
)
return None
output_dir.mkdir(parents=True, exist_ok=True)
partial_path.unlink(missing_ok=True)
env = target.process_environment()
try:
_run(
[
pg_dump,
"--format=custom",
"--compress=6",
"--no-owner",
"--no-privileges",
"--file",
str(partial_path),
],
env,
)
if not partial_path.exists() or partial_path.stat().st_size == 0:
raise RuntimeError("pg_dump completed without creating a non-empty archive")
if verify and pg_restore:
_run([pg_restore, "--list", str(partial_path)], env, stdout=subprocess.DEVNULL)
partial_path.replace(final_path)
manifest = {
"version": 1,
"created_at_utc": moment.astimezone(timezone.utc).isoformat(),
"database_host": target.host,
"database_name": target.database,
"format": "postgresql_custom",
"backup_file": final_path.name,
"bytes": final_path.stat().st_size,
"sha256": file_sha256(final_path),
"verified_with_pg_restore": verify,
}
final_path.with_suffix(".manifest.json").write_text(
json.dumps(manifest, indent=2) + "\n",
encoding="utf-8",
)
removed = prune_expired_backups(output_dir, retention_days, moment)
print(f"[OK] Backup verified: {final_path} ({manifest['bytes']} bytes)")
if removed:
print(f"[OK] Removed {len(removed)} expired backup file(s)")
return final_path
except Exception:
partial_path.unlink(missing_ok=True)
raise
def main(argv: list[str] | None = None) -> int:
load_dotenv(BACKEND_DIR / ".env", override=False)
parser = argparse.ArgumentParser(description="Create and verify a DocDoe PostgreSQL backup.")
parser.add_argument("--database-url", default=os.getenv("DATABASE_URL"))
parser.add_argument(
"--output-dir",
type=Path,
default=Path(os.getenv("DATABASE_BACKUP_DIR", str(DEFAULT_OUTPUT_DIR))),
)
parser.add_argument(
"--retention-days",
type=int,
default=int(os.getenv("DATABASE_BACKUP_RETENTION_DAYS", "14")),
)
parser.add_argument("--skip-verify", action="store_true")
parser.add_argument("--dry-run", action="store_true")
args = parser.parse_args(argv)
try:
create_backup(
database_url=args.database_url or "",
output_dir=args.output_dir,
retention_days=args.retention_days,
verify=not args.skip_verify,
dry_run=args.dry_run,
)
except (OSError, RuntimeError, ValueError) as exc:
print(f"[ERROR] {exc}", file=sys.stderr)
return 1
return 0
if __name__ == "__main__":
raise SystemExit(main())