railiance-platform/scripts/recovery_evidence.py
codex debf981097
All checks were successful
CI Smoke / host-smoke (push) Successful in 0s
CI Smoke / container-smoke (push) Successful in 3s
Close verified incident task and finish local workplan loose ends
Assistant: codex
Assistant-Model: gpt-6-astra
Assistant-Session: 01a0e3c3-621b-7350-9f77-50a8d3ee7657
2026-09-27 19:04:54 +02:00

147 lines
8 KiB
Python

"""Hash-pinned native recovery receipts, with original completion timestamps."""
import hashlib
import json
import re
from pathlib import Path
from service_assurance import timestamp
ROOT = Path(__file__).resolve().parents[1]
def pinned_receipt(entry, root):
path = (root / entry['path']).resolve()
if not path.is_relative_to((root / 'docs/evidence').resolve()):
raise ValueError('receipt outside evidence directory')
raw = path.read_bytes()
if hashlib.sha256(raw).hexdigest() != entry['sha256']:
raise ValueError('receipt drift')
return json.loads(raw)
def operation_times(receipt, now):
start, finish = (timestamp(receipt[key]) for key in ('started_at', 'finished_at'))
if not start <= finish <= now:
raise ValueError('invalid receipt chronology')
return start, finish
def primary_archive(receipt, now):
"""Accept only explicit full primary transfers, including a verified versioned GET."""
operation_times(receipt, now)
if (receipt['schema'] != 'platform.forgejo-primary-archive.v1'
or receipt['status'] != 'primary_fetched_pending_application_restore'
or receipt['stage'] != 'transfer_verified'
or receipt['archive_profile'] != 'full'
or not receipt['destination'].startswith(
's3://railiance-platform-pg-backup/platform-pg/application-archives/forgejo/')
or not re.fullmatch(r'[0-9a-f]{64}', receipt['ciphertext_sha256'])
or not all(receipt.get(key) is True for key in
('multipart_completed', 'version_pinned', 'download_hash_matches'))
or type(receipt['ciphertext_bytes']) is not int
or receipt['ciphertext_bytes'] <= 0
or any(type(receipt[key]) is not int or receipt[key] != receipt['ciphertext_bytes']
for key in ('uploaded_bytes', 'downloaded_bytes'))):
raise ValueError('primary archive not accepted')
def archive_signal(entry, receipt, now, root):
if entry['signal'] == 'offsite.upload':
primary_archive(receipt, now)
else:
start, _ = operation_times(receipt, now)
if (receipt['schema'] != 'platform.forgejo-isolated-restore.v1'
or receipt['status'] != 'restored'
or receipt['archive_profile'] != 'full'
or receipt['source_provider'] != 'Scaleway'
or receipt['stage'] != 'package_blob_recovery'
or not all(receipt.get(key) is True for key in
('database_import', 'application_health', 'cleanup'))
or not receipt['repositories_verified']
or type(receipt['package_blobs_verified']) is not int
or receipt['package_blobs_verified'] < 0
or type(receipt['database_counts']['package_blobs']) is not int
or receipt['package_blobs_verified'] != receipt['database_counts']['package_blobs']):
raise ValueError('full application recovery not accepted')
decryption = pinned_receipt(entry['decryption'], root)
transfer = pinned_receipt(entry['transfer'], root)
primary_archive(transfer, now)
decrypt_start, decrypt_finish = operation_times(decryption, now)
if (decryption['schema'] != 'platform.forgejo-primary-decryption.v1'
or decryption['status'] != 'primary_fetched_pending_application_restore'
or decryption['archive_profile'] != 'full'
or decryption['decrypted'] is not True
or decryption['download_hash_matches'] is not True
or not re.fullmatch(r'[0-9a-f]{64}', decryption['plaintext_sha256'])
or receipt['transfer_receipt_sha256'] != entry['decryption']['sha256']
or decryption['transfer_receipt_sha256'] != entry['transfer']['sha256']
or not timestamp(transfer['finished_at']) <= decrypt_start <= decrypt_finish <= start
or receipt['offsite_artifact'] != transfer['destination']
or decryption['destination'] != transfer['destination']
or decryption['ciphertext_bytes'] != transfer['ciphertext_bytes']
or any(r['ciphertext_sha256'] != transfer['ciphertext_sha256']
for r in (receipt, decryption))):
raise ValueError('recovery provenance mismatch')
return {'result': 'pass', 'observed_at': receipt['finished_at']}
def recovery_signals(now, root=ROOT):
index = json.loads((root / 'assurance/recovery-evidence.json').read_text())
if index['schema'] != 'railiance-platform.recovery-evidence.v1':
raise ValueError('unknown recovery index')
signals = {}
for entry in index['receipts']:
signal = entry['signal']
if signal in signals or signal not in ('apps-pg.restore', 'forgejo-db.restore',
'openbao.snapshot', 'offsite.upload', 'offsite.restore'):
raise ValueError('unexpected recovery signal')
sample = {'result': 'unavailable', 'observed_at': now.isoformat()}
try:
if signal.startswith('offsite.'):
signals[signal] = archive_signal(entry, pinned_receipt(entry, root), now, root)
continue
path = (root / entry['path']).resolve()
allowed = [root / 'docs/evidence']
if signal == 'openbao.snapshot':
allowed.append(root / 'reviews')
if not any(path.is_relative_to(directory.resolve()) for directory in allowed):
raise ValueError('receipt outside evidence directory')
raw = path.read_bytes()
if hashlib.sha256(raw).hexdigest() != entry['sha256']:
raise ValueError('receipt drift')
receipt = json.loads(raw)
if signal == 'openbao.snapshot':
required = ('snapshot_created', 'source_initialized', 'source_unsealed',
'snapshot_encrypted', 'encrypted_copy_off_host',
'encryption_verified', 'hash_verified', 'no_secret_material_recorded')
if (receipt.get('receipt_version') != 1
or receipt.get('source_cluster') != 'railiance01'
or receipt.get('source_namespace') != 'openbao'
or receipt.get('cluster_id') != 'fd28df5d-98ec-57dd-42ec-9b3e4f4e53bf'
or not all(receipt.get(key) is True for key in required)
or not receipt.get('encrypted_location_ref', '').startswith('offhost-custody:')):
raise ValueError('snapshot not accepted')
for key in ('snapshot_sha256', 'encrypted_snapshot_sha256'):
value = receipt.get(key, '')
if not re.fullmatch(r'sha256:[0-9a-f]{64}', value):
raise ValueError('snapshot hash missing')
if timestamp(receipt['created_at']) > now:
raise ValueError('future snapshot')
signals[signal] = {'result': 'pass', 'observed_at': receipt['created_at']}
continue
cell = signal.removesuffix('.restore')
if (receipt['schema'] != 'platform.scaleway-primary-restore.v1'
or receipt['primary_destination'] != f's3://railiance-platform-pg-backup/platform-pg/{cell}/'
or receipt['source'] != 'Scaleway Barman base backup and WAL'
or receipt['stage'] != 'database_acceptance'
or receipt['status'] != 'verified'
or receipt['cleanup'] is not True
or receipt['production_ready'] is not True):
raise ValueError('receipt not accepted')
completed = timestamp(receipt['finished_at'])
if not timestamp(receipt['started_at']) <= completed <= now:
raise ValueError('invalid receipt chronology')
sample = {'result': 'pass', 'observed_at': receipt['finished_at']}
except (OSError, ValueError, KeyError, TypeError, AttributeError):
pass
signals[signal] = sample
return signals