File size: 4,307 Bytes
6993919 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 | #!/usr/bin/env python3
"""Qdrant collection audit + cleanup for test_col_* artifacts.
T15 (G14 FIX) — Qdrant accumulates test_col_* collections from integration
tests that forget to clean up. This script audits and cleans them.
Usage:
# Audit only (safe — no changes):
python scripts/ops/qdrant_audit.py
python scripts/ops/qdrant_audit.py --audit-only
# Audit + cleanup (drops test_col_*):
python scripts/ops/qdrant_audit.py --cleanup
# Custom Qdrant URL:
python scripts/ops/qdrant_audit.py --url http://qdrant.example:6333 --cleanup
# CI gate mode (exit 1 if any test_col_* found):
python scripts/ops/qdrant_audit.py --check
Cron suggestion (netcup):
# Daily audit, notify if anything leaked:
0 6 * * * python3 /root/scripts/qdrant_audit.py --audit-only | \\
curl -s --data-binary @- http://localhost:9080/rmi-info || true
"""
from __future__ import annotations
import argparse
import json
import sys
import urllib.request
def fetch_collections(base_url: str) -> list[dict]:
with urllib.request.urlopen(f"{base_url}/collections", timeout=10) as r:
return json.loads(r.read()).get("result", {}).get("collections", [])
def fetch_collection_info(base_url: str, name: str) -> dict:
try:
with urllib.request.urlopen(
f"{base_url}/collections/{name}", timeout=5
) as r:
return json.loads(r.read()).get("result", {})
except Exception as e:
return {"error": str(e)}
def delete_collection(base_url: str, name: str) -> bool:
req = urllib.request.Request(
f"{base_url}/collections/{name}", method="DELETE"
)
try:
with urllib.request.urlopen(req, timeout=10) as r:
return json.loads(r.read()).get("status") == "ok"
except Exception:
return False
def main() -> int:
p = argparse.ArgumentParser(description=__doc__)
p.add_argument(
"--url",
default="http://localhost:6333",
help="Qdrant base URL (default: http://localhost:6333)",
)
p.add_argument(
"--audit-only",
action="store_true",
help="Only audit — never delete anything",
)
p.add_argument(
"--cleanup",
action="store_true",
help="Audit + drop all test_col_* collections",
)
p.add_argument(
"--check",
action="store_true",
help="CI mode: exit 1 if any test_col_* found (no output unless found)",
)
args = p.parse_args()
cols = fetch_collections(args.url)
test_like = [c["name"] for c in cols if c["name"].startswith("test_col")]
production = [c["name"] for c in cols if not c["name"].startswith("test_col")]
if args.check:
if test_like:
print(
f"FAIL: {len(test_like)} test_col_* collection(s) found: "
f"{test_like}",
file=sys.stderr,
)
return 1
return 0
print(f"=== Qdrant Audit: {args.url} ===")
print(f"Total collections: {len(cols)}")
print(f" Production: {len(production)}")
print(f" Test artifacts (test_col_*): {len(test_like)}")
if not test_like:
print("\nCLEAN — zero test_col_* artifacts.")
return 0
print("\n--- Test artifacts detail ---")
for name in test_like:
info = fetch_collection_info(args.url, name)
points = info.get("points_count", "?")
vectors_size = (
info.get("config", {}).get("params", {}).get("vectors", {}).get("size", "?")
)
print(f" {name}: {points} points, {vectors_size}-dim vectors")
if args.audit_only:
print(f"\n(Use --cleanup to drop these {len(test_like)} collection(s))")
return 0
if args.cleanup:
print("\n--- Cleanup ---")
dropped = 0
for name in test_like:
if delete_collection(args.url, name):
print(f" ✓ dropped {name}")
dropped += 1
else:
print(f" ✗ FAILED to drop {name}")
print(f"\n{dropped}/{len(test_like)} dropped.")
return 0 if dropped == len(test_like) else 1
print("\n(Use --cleanup to drop these, or --audit-only to suppress this hint)")
return 0
if __name__ == "__main__":
raise SystemExit(main())
|