chg: Reduce the queries against the ranking db

pull/12/head
Raphaël Vinot 2018-04-10 14:36:08 +02:00
parent f65228a2df
commit ce96ad9825
1 changed files with 7 additions and 14 deletions

View File

@ -26,28 +26,21 @@ class Ranking():
self.logger.info('Start ranking') self.logger.info('Start ranking')
set_running(self.__class__.__name__) set_running(self.__class__.__name__)
today = date.today().isoformat() today = date.today().isoformat()
v4_last = self.asn_meta.get('v4|last') v4_last, v6_last = self.asn_meta.mget('v4|last', 'v6|last')
v6_last = self.asn_meta.get('v6|last')
if not v4_last or not v6_last: if not v4_last or not v6_last:
'''Failsafe if asn_meta has not been populated yet''' '''Failsafe if asn_meta has not been populated yet'''
unset_running(self.__class__.__name__) unset_running(self.__class__.__name__)
return return
asns_aggregation_key_v4 = f'{today}|asns|v4' asns_aggregation_key_v4 = f'{today}|asns|v4'
asns_aggregation_key_v6 = f'{today}|asns|v6' asns_aggregation_key_v6 = f'{today}|asns|v6'
self.ranking.delete(asns_aggregation_key_v4, asns_aggregation_key_v6) to_delete = set([asns_aggregation_key_v4, asns_aggregation_key_v6])
cleaned_up = set() r_pipeline = self.ranking.pipeline()
for source in self.storage.smembers(f'{today}|sources'): for source in self.storage.smembers(f'{today}|sources'):
self.logger.info(f'{today} - Ranking source: {source}') self.logger.info(f'{today} - Ranking source: {source}')
r_pipeline = self.ranking.pipeline()
for asn in self.storage.smembers(f'{today}|{source}'): for asn in self.storage.smembers(f'{today}|{source}'):
prefixes_aggregation_key_v4 = f'{today}|{asn}|v4' prefixes_aggregation_key_v4 = f'{today}|{asn}|v4'
prefixes_aggregation_key_v6 = f'{today}|{asn}|v6' prefixes_aggregation_key_v6 = f'{today}|{asn}|v6'
if prefixes_aggregation_key_v4 not in cleaned_up: to_delete.update([prefixes_aggregation_key_v4, prefixes_aggregation_key_v6])
self.ranking.delete(prefixes_aggregation_key_v4)
cleaned_up.add(prefixes_aggregation_key_v4)
if prefixes_aggregation_key_v6 not in cleaned_up:
self.ranking.delete(prefixes_aggregation_key_v6)
cleaned_up.add(prefixes_aggregation_key_v6)
if asn == '0': if asn == '0':
# Default ASN when no matches. Probably spoofed. # Default ASN when no matches. Probably spoofed.
continue continue
@ -66,8 +59,7 @@ class Ranking():
else: else:
asn_rank_v6 += len(ips) * self.config_files[source]['impact'] asn_rank_v6 += len(ips) * self.config_files[source]['impact']
r_pipeline.zincrby(prefixes_aggregation_key_v6, prefix, prefix_rank * self.config_files[source]['impact']) r_pipeline.zincrby(prefixes_aggregation_key_v6, prefix, prefix_rank * self.config_files[source]['impact'])
v4count = self.asn_meta.get(f'{v4_last}|{asn}|v4|ipcount') v4count, v6count = self.asn_meta.mget(f'{v4_last}|{asn}|v4|ipcount', f'{v6_last}|{asn}|v6|ipcount')
v6count = self.asn_meta.get(f'{v6_last}|{asn}|v6|ipcount')
if v4count: if v4count:
asn_rank_v4 /= float(v4count) asn_rank_v4 /= float(v4count)
if asn_rank_v4: if asn_rank_v4:
@ -78,7 +70,8 @@ class Ranking():
if asn_rank_v6: if asn_rank_v6:
r_pipeline.set(f'{today}|{source}|{asn}|rankv6', asn_rank_v6) r_pipeline.set(f'{today}|{source}|{asn}|rankv6', asn_rank_v6)
r_pipeline.zincrby(asns_aggregation_key_v6, asn, asn_rank_v6) r_pipeline.zincrby(asns_aggregation_key_v6, asn, asn_rank_v6)
r_pipeline.execute() self.ranking.delete(*to_delete)
r_pipeline.execute()
unset_running(self.__class__.__name__) unset_running(self.__class__.__name__)
self.logger.info('Ranking done.') self.logger.info('Ranking done.')