fix: do not keep broken IPs forever in redis

bin/dbinsert.py

@@ -74,6 +74,7 @@ class DBInsertManager(AbstractManager):
                 data = sanitized_data[i]
                 if not data:
                     self.logger.warning(f'No data for UUID {uuid}. This should not happen, but lets move on.')
+                    done.append(uuid)
                     continue
                 routing_info = responses['responses'][i]['response']  # our queries are on one single date, not a range
                 # Data gathered from IPASN History:
@@ -81,6 +82,7 @@ class DBInsertManager(AbstractManager):
                 # * AS number
                 if 'error' in routing_info:
                     self.logger.warning(f"Unable to find routing information for {data['ip']} - {data['datetime']}: {routing_info['error']}")
+                    done.append(uuid)
                     continue
                 # Single date query, getting from the object
                 datetime_routing = list(routing_info.keys())[0]
@@ -91,9 +93,11 @@ class DBInsertManager(AbstractManager):
                     continue
                 if 'asn' in entry and entry['asn'] in [None, '0']:
                     self.logger.warning(f"Unable to find the AS number associated to {data['ip']} - {data['datetime']} (got {entry['asn']}).")
+                    done.append(uuid)
                     continue
                 if 'prefix' in entry and entry['prefix'] in [None, '0.0.0.0/0', '::/0']:
                     self.logger.warning(f"Unable to find the prefix associated to {data['ip']} - {data['datetime']} (got {entry['prefix']}).")
+                    done.append(uuid)
                     continue
 
                 # Format: <YYYY-MM-DD>|sources -> set([<source>, ...])

poetry.lock

@@ -818,8 +818,8 @@ testing = ["flake8 (<5)", "func-timeout", "jaraco.functools", "jaraco.itertools"
 
 [metadata]
 lock-version = "1.1"
-python-versions = "^3.8"
+python-versions = "^3.8.1"
-content-hash = "a7b14e4e425b55589575ff0b0b2e8ec99f3d32ef32c8c3ac90dcf7ec2afb9fee"
+content-hash = "00557ce6a5aa68b4251296d5d45716a49467cecf5cde1a0cab3c00e2f1d721ff"
 
 [metadata.files]
 aiohttp = [

pyproject.toml

@@ -24,7 +24,7 @@ ssfetcher = "bin.ssfetcher:main"
 
 
 [tool.poetry.dependencies]
-python = "^3.8"
+python = "^3.8.1"
 redis = {version = "^4.4.0", extras = ["hiredis"]}
 flask-restx = "^0.5.1"
 gunicorn = "^20.1.0"

tools/clear_prepare_db.py

@@ -0,0 +1,16 @@
+#!/usr/bin/env python3
+
+import uuid
+
+from redis import Redis
+from bgpranking.default import get_socket_path
+
+redis_sanitized = Redis(unix_socket_path=get_socket_path('prepare'), db=0, decode_responses=True)
+
+while name := redis_sanitized.scan_iter(_type='HASH', count=10):
+    try:
+        uuid.uuid(name)
+    except Exception:
+        pass
+    if not redis_sanitized.sismember('to_insert'):
+        print(name)