fix(backup): cross-check bundle coverage with dev prod repos
Some checks failed
CD Pipeline / workflow-shape (push) Successful in 1s
CD Pipeline / cancel-stale-cd (push) Has been skipped
CD Pipeline / tests (push) Successful in 59s
CD Pipeline / post-deploy-checks (push) Has been cancelled
CD Pipeline / build-and-deploy (push) Has been cancelled
AWOOOI Harbor 110 Local Repair / workflow-shape (push) Successful in 0s
AWOOOI Harbor 110 Local Repair / harbor-110-local-repair (push) Failing after 31s

This commit is contained in:
Your Name
2026-07-03 09:01:47 +08:00
parent 890c2e57a1
commit 7085a94eb6
7 changed files with 354 additions and 0 deletions

View File

@@ -3107,6 +3107,31 @@ def _enrich_from_current_readbacks(
reboot_service_data_backup.get("gitea_repo_bundle_active_blocker_count")
or gitea_bundle_backup.get("active_blocker_count")
)
state["gitea_repo_bundle_dev_prod_repo_truth_expected_count"] = _int(
reboot_service_data_backup.get(
"gitea_repo_bundle_dev_prod_repo_truth_expected_count"
)
or gitea_bundle_backup.get("dev_prod_repo_truth_expected_count")
)
state["gitea_repo_bundle_dev_prod_repo_truth_backup_covered_count"] = _int(
reboot_service_data_backup.get(
"gitea_repo_bundle_dev_prod_repo_truth_backup_covered_count"
)
or gitea_bundle_backup.get("dev_prod_repo_truth_backup_covered_count")
)
state["gitea_repo_bundle_missing_dev_prod_repo_backup_count"] = _int(
reboot_service_data_backup.get(
"gitea_repo_bundle_missing_dev_prod_repo_backup_count"
)
or gitea_bundle_backup.get("missing_dev_prod_repo_backup_count")
)
state["gitea_repo_bundle_all_dev_prod_repos_have_backup"] = bool(
reboot_service_data_backup.get(
"gitea_repo_bundle_all_dev_prod_repos_have_backup"
)
is True
or gitea_bundle_backup.get("all_dev_prod_repos_have_bundle_backup") is True
)
state["backup_core_green_false_but_gitea_repo_bundle_ready"] = bool(
reboot_service_data_backup.get(
"backup_core_green_false_but_gitea_repo_bundle_ready"
@@ -3352,6 +3377,20 @@ def _enrich_from_current_readbacks(
evidence["gitea_repo_bundle_active_blocker_count"] = _int(
state.get("gitea_repo_bundle_active_blocker_count")
)
evidence["gitea_repo_bundle_dev_prod_repo_truth_expected_count"] = _int(
state.get("gitea_repo_bundle_dev_prod_repo_truth_expected_count")
)
evidence[
"gitea_repo_bundle_dev_prod_repo_truth_backup_covered_count"
] = _int(
state.get("gitea_repo_bundle_dev_prod_repo_truth_backup_covered_count")
)
evidence["gitea_repo_bundle_missing_dev_prod_repo_backup_count"] = _int(
state.get("gitea_repo_bundle_missing_dev_prod_repo_backup_count")
)
evidence["gitea_repo_bundle_all_dev_prod_repos_have_backup"] = (
state.get("gitea_repo_bundle_all_dev_prod_repos_have_backup") is True
)
evidence["backup_core_green_false_but_gitea_repo_bundle_ready"] = (
state.get("backup_core_green_false_but_gitea_repo_bundle_ready") is True
)
@@ -3965,6 +4004,18 @@ def _set_rollups_and_summary(
"gitea_repo_bundle_active_blocker_count": _int(
state.get("gitea_repo_bundle_active_blocker_count")
),
"gitea_repo_bundle_dev_prod_repo_truth_expected_count": _int(
state.get("gitea_repo_bundle_dev_prod_repo_truth_expected_count")
),
"gitea_repo_bundle_dev_prod_repo_truth_backup_covered_count": _int(
state.get("gitea_repo_bundle_dev_prod_repo_truth_backup_covered_count")
),
"gitea_repo_bundle_missing_dev_prod_repo_backup_count": _int(
state.get("gitea_repo_bundle_missing_dev_prod_repo_backup_count")
),
"gitea_repo_bundle_all_dev_prod_repos_have_backup": (
state.get("gitea_repo_bundle_all_dev_prod_repos_have_backup") is True
),
"backup_core_green_false_but_gitea_repo_bundle_ready": (
state.get("backup_core_green_false_but_gitea_repo_bundle_ready") is True
),
@@ -4073,6 +4124,18 @@ def _set_rollups_and_summary(
"gitea_repo_bundle_active_blocker_count": _int(
state.get("gitea_repo_bundle_active_blocker_count")
),
"gitea_repo_bundle_dev_prod_repo_truth_expected_count": _int(
state.get("gitea_repo_bundle_dev_prod_repo_truth_expected_count")
),
"gitea_repo_bundle_dev_prod_repo_truth_backup_covered_count": _int(
state.get("gitea_repo_bundle_dev_prod_repo_truth_backup_covered_count")
),
"gitea_repo_bundle_missing_dev_prod_repo_backup_count": _int(
state.get("gitea_repo_bundle_missing_dev_prod_repo_backup_count")
),
"gitea_repo_bundle_all_dev_prod_repos_have_backup": (
state.get("gitea_repo_bundle_all_dev_prod_repos_have_backup") is True
),
"backup_core_green_false_but_gitea_repo_bundle_ready": (
state.get("backup_core_green_false_but_gitea_repo_bundle_ready") is True
),

View File

@@ -9,17 +9,21 @@ from __future__ import annotations
import json
from datetime import datetime
from pathlib import Path
from typing import Any
from urllib.parse import urlencode
from urllib.request import Request, urlopen
from zoneinfo import ZoneInfo
from src.core.config import settings
from src.services.snapshot_paths import default_operations_dir
SCHEMA_VERSION = "gitea_repo_bundle_backup_readback_v1"
TZ_TAIPEI = ZoneInfo("Asia/Taipei")
DEFAULT_HOST = "188"
PROMETHEUS_TIMEOUT_SECONDS = 4
_DEFAULT_OPERATIONS_DIR = default_operations_dir(Path(__file__))
_DEV_PROD_REPO_READBACK_FILE = "gitea-all-product-dev-prod-repo-readback.snapshot.json"
def build_gitea_repo_bundle_backup_readback(
@@ -28,6 +32,9 @@ def build_gitea_repo_bundle_backup_readback(
generated_at: str | None = None,
host: str = DEFAULT_HOST,
metrics_error: str = "",
dev_prod_repo_refs: list[str] | None = None,
dev_prod_repo_truth_ref: str = "",
dev_prod_repo_truth_error: str = "",
) -> dict[str, Any]:
"""Build the readback payload from Prometheus vector samples."""
now = generated_at or datetime.now(TZ_TAIPEI).isoformat(timespec="seconds")
@@ -39,6 +46,12 @@ def build_gitea_repo_bundle_backup_readback(
metrics_present = bool(filtered) and not metrics_error
scalar = _scalar_index(filtered)
repo_rows = _repo_rows(filtered)
dev_prod_coverage = _dev_prod_repo_truth_coverage(
repo_rows=repo_rows,
expected_repo_refs=dev_prod_repo_refs,
source_ref=dev_prod_repo_truth_ref,
source_error=dev_prod_repo_truth_error,
)
expected_repo_count = int(_first_scalar(scalar, "awoooi_gitea_bundle_expected_repo_count", 0))
missing_count = int(_first_scalar(scalar, "awoooi_gitea_bundle_expected_repo_missing_count", 0))
@@ -72,6 +85,13 @@ def build_gitea_repo_bundle_backup_readback(
blockers.append("gitea_bundle_sample_restore_dry_run_missing")
elif metrics_present and not sample_restore_ok:
blockers.append("gitea_bundle_sample_restore_dry_run_failed")
if dev_prod_coverage["source_error"]:
blockers.append("gitea_bundle_dev_prod_repo_truth_unavailable")
elif (
dev_prod_coverage["evaluated"]
and not dev_prod_coverage["all_dev_prod_repos_have_bundle_backup"]
):
blockers.append("gitea_bundle_dev_prod_repo_truth_missing")
ready = not blockers
if ready:
@@ -86,6 +106,12 @@ def build_gitea_repo_bundle_backup_readback(
elif "gitea_bundle_sample_restore_dry_run_missing" in blockers:
status_value = "blocked_gitea_bundle_restore_dry_run_missing"
safe_next_step = "run_gitea_bundle_sample_restore_dry_run_no_secret_no_prod_restore"
elif "gitea_bundle_dev_prod_repo_truth_missing" in blockers:
status_value = "blocked_gitea_bundle_dev_prod_repo_truth_gap"
safe_next_step = "run_gitea_bundle_backup_until_all_dev_prod_repos_are_covered"
elif "gitea_bundle_dev_prod_repo_truth_unavailable" in blockers:
status_value = "blocked_gitea_bundle_dev_prod_repo_truth_unavailable"
safe_next_step = "restore_committed_dev_prod_repo_truth_snapshot_then_rerun_readback"
else:
status_value = "blocked_gitea_bundle_integrity_gap"
safe_next_step = "inspect_awoooi_gitea_bundle_metrics_before_any_restore_or_repo_change"
@@ -120,8 +146,21 @@ def build_gitea_repo_bundle_backup_readback(
"all_expected_ok": all_expected_ok,
"sample_restore_dry_run_present": sample_restore_present,
"sample_restore_dry_run_ok": sample_restore_ok,
"dev_prod_repo_truth_expected_count": dev_prod_coverage[
"expected_repo_count"
],
"dev_prod_repo_truth_backup_covered_count": dev_prod_coverage[
"backup_covered_repo_count"
],
"missing_dev_prod_repo_backup_count": dev_prod_coverage[
"missing_backup_repo_count"
],
"all_dev_prod_repos_have_bundle_backup": dev_prod_coverage[
"all_dev_prod_repos_have_bundle_backup"
],
},
"repos": repo_rows,
"dev_prod_repo_truth_coverage": dev_prod_coverage,
"metrics_error": metrics_error,
"restore_dry_run": {
"metric_name": "awoooi_gitea_bundle_sample_restore_dry_run_ok",
@@ -140,6 +179,7 @@ def load_latest_gitea_repo_bundle_backup_readback(
*,
host: str = DEFAULT_HOST,
prometheus_url: str | None = None,
operations_dir: Path | None = None,
) -> dict[str, Any]:
"""Load live Prometheus readback, failing closed when metrics are unavailable."""
url = (prometheus_url or settings.PROMETHEUS_URL).rstrip("/")
@@ -161,6 +201,16 @@ def load_latest_gitea_repo_bundle_backup_readback(
]
samples: list[dict[str, Any]] = []
errors: list[str] = []
dev_prod_refs: list[str] = []
dev_prod_ref = f"docs/operations/{_DEV_PROD_REPO_READBACK_FILE}"
dev_prod_error = ""
try:
dev_prod_refs = _load_dev_prod_repo_refs(
(operations_dir or _DEFAULT_OPERATIONS_DIR)
/ _DEV_PROD_REPO_READBACK_FILE
)
except Exception as exc: # noqa: BLE001 - fail closed with redacted error class
dev_prod_error = exc.__class__.__name__
for query in queries:
try:
samples.extend(_query_prometheus_vector(url, query))
@@ -170,6 +220,9 @@ def load_latest_gitea_repo_bundle_backup_readback(
metric_samples=samples,
host=host,
metrics_error=",".join(errors),
dev_prod_repo_refs=dev_prod_refs,
dev_prod_repo_truth_ref=dev_prod_ref,
dev_prod_repo_truth_error=dev_prod_error,
)
@@ -247,3 +300,58 @@ def _repo_rows(samples: list[dict[str, Any]]) -> list[dict[str, Any]]:
elif name == "awoooi_gitea_bundle_checksum_digest_present":
row["checksum_digest_present"] = value == 1
return [repos[key] for key in sorted(repos)]
def _load_dev_prod_repo_refs(path: Path) -> list[str]:
payload = json.loads(path.read_text(encoding="utf-8"))
products = payload.get("products") if isinstance(payload, dict) else None
if not isinstance(products, list):
raise ValueError("dev_prod_products_not_list")
refs = [
str(product.get("gitea_repo") or "")
for product in products
if isinstance(product, dict) and product.get("gitea_repo")
]
if not refs:
raise ValueError("dev_prod_repo_refs_empty")
return sorted(set(refs))
def _dev_prod_repo_truth_coverage(
*,
repo_rows: list[dict[str, Any]],
expected_repo_refs: list[str] | None,
source_ref: str,
source_error: str,
) -> dict[str, Any]:
expected = sorted(set(expected_repo_refs or []))
covered = sorted(
str(row.get("repo"))
for row in repo_rows
if row.get("repo")
and row.get("present") is True
and row.get("ok") is True
and row.get("checksum_digest_present") is True
)
covered_set = set(covered)
expected_set = set(expected)
missing = sorted(expected_set - covered_set)
unexpected = sorted(covered_set - expected_set) if expected else []
evaluated = bool(expected) or bool(source_error)
return {
"schema_version": "gitea_bundle_dev_prod_repo_truth_coverage_v1",
"source_ref": source_ref,
"source_error": source_error,
"evaluated": evaluated,
"expected_repo_count": len(expected),
"backup_covered_repo_count": len(expected_set & covered_set),
"missing_backup_repo_count": len(missing),
"unexpected_backup_repo_count": len(unexpected),
"all_dev_prod_repos_have_bundle_backup": evaluated
and not source_error
and bool(expected)
and not missing,
"expected_repos": expected,
"missing_backup_repos": missing,
"unexpected_backup_repos": unexpected,
}

View File

@@ -1368,6 +1368,18 @@ def apply_gitea_repo_bundle_backup_readback(
"sample_restore_dry_run_ok": (
summary.get("sample_restore_dry_run_ok") is True
),
"dev_prod_repo_truth_expected_count": _int(
summary.get("dev_prod_repo_truth_expected_count")
),
"dev_prod_repo_truth_backup_covered_count": _int(
summary.get("dev_prod_repo_truth_backup_covered_count")
),
"missing_dev_prod_repo_backup_count": _int(
summary.get("missing_dev_prod_repo_backup_count")
),
"all_dev_prod_repos_have_bundle_backup": (
summary.get("all_dev_prod_repos_have_bundle_backup") is True
),
"production_restore_performed": (
_dict(readback.get("restore_dry_run")).get("production_restore_performed")
is True
@@ -1391,6 +1403,18 @@ def apply_gitea_repo_bundle_backup_readback(
]
service_backup["gitea_repo_bundle_active_blockers"] = active_blockers
service_backup["gitea_repo_bundle_active_blocker_count"] = len(active_blockers)
service_backup["gitea_repo_bundle_dev_prod_repo_truth_expected_count"] = (
bundle_summary["dev_prod_repo_truth_expected_count"]
)
service_backup["gitea_repo_bundle_dev_prod_repo_truth_backup_covered_count"] = (
bundle_summary["dev_prod_repo_truth_backup_covered_count"]
)
service_backup["gitea_repo_bundle_missing_dev_prod_repo_backup_count"] = (
bundle_summary["missing_dev_prod_repo_backup_count"]
)
service_backup["gitea_repo_bundle_all_dev_prod_repos_have_backup"] = (
bundle_summary["all_dev_prod_repos_have_bundle_backup"]
)
service_backup["backup_core_green_false_but_gitea_repo_bundle_ready"] = (
service_backup.get("backup_core_green") is False and ready
)
@@ -1415,6 +1439,18 @@ def apply_gitea_repo_bundle_backup_readback(
]
rollups["gitea_repo_bundle_repo_row_count"] = bundle_summary["repo_row_count"]
rollups["gitea_repo_bundle_active_blocker_count"] = len(active_blockers)
rollups["gitea_repo_bundle_dev_prod_repo_truth_expected_count"] = (
bundle_summary["dev_prod_repo_truth_expected_count"]
)
rollups["gitea_repo_bundle_dev_prod_repo_truth_backup_covered_count"] = (
bundle_summary["dev_prod_repo_truth_backup_covered_count"]
)
rollups["gitea_repo_bundle_missing_dev_prod_repo_backup_count"] = (
bundle_summary["missing_dev_prod_repo_backup_count"]
)
rollups["gitea_repo_bundle_all_dev_prod_repos_have_backup"] = (
bundle_summary["all_dev_prod_repos_have_bundle_backup"]
)
rollups["backup_core_green_false_but_gitea_repo_bundle_ready"] = (
service_backup["backup_core_green_false_but_gitea_repo_bundle_ready"]
)