def calculate_effective_prefixes(self, start_interval=1451692800):
'''
This method calculates the number of effective /24
prefixes that lie underneath a specified route prefix
:param start_interval: the start interval of the data
stream
:return: number of effective prefixes under the route's
root prefix
'''
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('collector', self.collector_name)
stream.add_filter('record-type', 'ribs')
if isinstance(start_interval, datetime.datetime):
interval_start_utc = self.convert_interval_to_utc(start_interval)
stream.add_interval_filter(interval_start_utc - 300, interval_start_utc + 300)
else:
stream.add_interval_filter(start_interval - 300, start_interval + 300)
stream.start()
print "Starting routing table parsing"
while (stream.get_next_record(rec)):
elem = rec.get_next_elem()
while elem:
# Get the peer ASN and IP. We then construct a peer_id, since a collector
# can establish multiple connections with the same ASN.
peer_asn = elem.peer_asn
peer_asn_ip = elem.peer_address
# make this an unmodifiable tuple
peer_id = (peer_asn, peer_asn_ip)
peer_route_trie = self.peer_trie_handles[peer_id]
# Do a single level search of the route prefix to
# find the left and right prefix advertisements
single_level_prefixes = \
peer_route_trie.single_level_search(elem.fields['prefix'])
print single_level_prefixes
prefix_count = len(single_level_prefixes)
root_24_prefix_count = 0
lr_24_prefix_count1 = lr_24_prefix_count2 = 0
# The /24 prefixes below the advertised prefix are calculated
# as all the /24 prefixes served by the root - sum of the /24
# prefixes served by root's children
if prefix_count == 1:
root_24_prefix_count = \
2 ** (24 - int(str(single_level_prefixes[0]).lstrip('<')
.rstrip('>').split('/')[1]))
elif prefix_count == 2:
root_24_prefix_count = \
2 ** (24 - int(str(single_level_prefixes[0]).lstrip('<')
.rstrip('>').split('/')[1]))
lr_24_prefix_count1 = \
2 ** (24 - int(str(single_level_prefixes[1]).lstrip('<')
.rstrip('>').split('/')[1]))
else:
root_24_prefix_count = \
2 ** (24 - int(str(single_level_prefixes[0]).lstrip('<')
.rstrip('>').split('/')[1]))
lr_24_prefix_count1 = \
2 ** (24 - int(str(single_level_prefixes[1]).lstrip('<')
.rstrip('>').split('/')[1]))
lr_24_prefix_count2 = \
2 ** (24 - int(str(single_level_prefixes[2]).lstrip('<')
.rstrip('>').split('/')[1]))
effective_24_prefix_count = \
root_24_prefix_count - (lr_24_prefix_count1 +
lr_24_prefix_count2)
print "Effective Prefix Count : ", \
effective_24_prefix_count
for prefix in single_level_prefixes:
trie_node = peer_route_trie.search_exact(str(prefix))
as_path = trie_node.data['as-path'].split(" ")
as_path_headless = as_path[1:-1]
print "AS-Path : ", as_path
as_headless_len = len(as_path_headless)
if as_headless_len > 1:
for i in range(0, as_headless_len - 1):
print "Headless nodes : ", as_path_headless[i], \
as_path_headless[i + 1]
if as_path_headless[i] in self.aslink_datastore:
self.aslink_datastore[as_path_headless[i]] += 1
else:
self.aslink_datastore[as_path_headless[i]] = 1
elem = rec.get_next_elem()
def run_bgpstream(args):
(collector, start_time, end_time, data_type) = args
# initialize and configure BGPStream
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('collector', collector)
# NB: BGPStream uses inclusive/inclusive intervals, so subtract one off the
# end time since we are using inclusive/exclusive intervals
stream.add_interval_filter(start_time, end_time-1)
stream.add_filter('record-type', data_type)
stream.start()
# per-peer data
peers_data = {}
# loop over all records in the stream
while stream.get_next_record(rec):
elem = rec.get_next_elem()
# to track the peers that have elems in this record
peer_signatures = set()
# loop over all elems in the record
while elem:
# create a peer signature for this elem
sig = peer_signature(rec, elem)
peer_signatures.add(sig)
# if this is the first time we have ever seen this peer, create
# an empty result: (Pfxs_v4_set, Pfxs_v6_set, ASNs_v4_set, ASNs_v6_set)
if sig not in peers_data:
peers_data[sig] =[set(),set(),
set(),set()]
if('prefix' in elem.fields):
pfx=elem.fields['prefix'];
if(":" in pfx):
peers_data[sig][1].add(pfx)
else:
peers_data[sig][0].add(pfx)
if('as-path' in elem.fields):
path_split=elem.fields['as-path'].split()
if(":" in pfx):
for ASN in path_split:
peers_data[sig][3].add(ASN)
else:
for ASN in path_split:
peers_data[sig][2].add(ASN)
elem = rec.get_next_elem()
# one peer that was present in this record (allows a true, per-collector
# count of records since each record can contain elems for many peers)
# the time in the output row is truncated down to a multiple of
# RESULT_GRANULARITY so that slices can be merged correctly
start_time = \
int(math.floor(start_time/RESULT_GRANULARITY) * RESULT_GRANULARITY)
# for each peer that we processed data for, create an output row
return [((start_time, collector, p), (peers_data[p])) for p in peers_data]
def recv_bgpstream_rib(begin, until, collector):
"""
Receive and parse BGP RIB records from a given bgpstream collector.
"""
logging.info("CALL recv_bgpstream_rib")
# Create bgpstream
stream = BGPStream()
rec = BGPRecord()
# set filtering
stream.add_filter('collector', collector)
stream.add_filter('record-type', 'updates')
stream.add_interval_filter(begin, until)
# Start the stream
stream.start()
while (stream.get_next_record(rec)):
if rec.status == 'valid':
elem = rec.get_next_elem()
else:
logging.warn("stream record invalid, skipping.")
continue
bgp_message = None
while (elem):
if (elem.type.upper() == 'A') or (elem.type.upper() == 'R'):
bgp_message = BGPmessage(elem.time, 'update')
aspath = elem.fields['as-path'].split()
for a in aspath:
if not '{' in a: # ignore AS-SETs
bgp_message.add_as_to_path(a)
bgp_message.add_announce(elem.fields['prefix'])
output(bgp_message)
elem = rec.get_next_elem()
# end while (elem)
# end while (stream...)
output('FLUSH')
def run_collector(stream, output_file):
# Start the stream
stream.start()
# Create record
rec = BGPRecord()
# Collect some general stream stats
total_records = 0
total_updates = 0
# Write to output file
with open(output_file, 'w') as f:
# Read in each record at a time
while (stream.get_next_record(rec)):
# Print the record information only if it is not a valid record
if rec.status != "valid":
print('Recieved invalid record from BGPStream collector {}'.
format(rec.collector))
record_members = []
for i in inspect.getmembers(rec):
# Ignores anything starting with underscore
# (that is, private and protected attributes)
if not i[0].startswith('_'):
# Ignores methods
if not inspect.ismethod(i[1]):
record_members.append(i)
print('Invalid record: {}'.format(record_members))
else:
elem = rec.get_next_elem()
while (elem):
# Aggregate metadata about new record
metadata = dict()
metadata['rec_dump_time'] = rec.dump_time
metadata['project'] = rec.project
metadata['collector'] = rec.collector
metadata['rec_type'] = rec.type
metadata['rec_time'] = rec.time
metadata['status'] = rec.status
metadata['elem_type'] = elem.type
metadata['elem_time'] = elem.time
metadata['peer_address'] = elem.peer_address
metadata['peer_asn'] = elem.peer_asn
# Update and log stats
total_records += 1
if elem.type == 'A':
total_updates += 1
if total_records % 100 == 0:
print(
'Collected {} total records and {} total announcements...'
.format(total_records, total_updates))
# Build full record and save to file
full_record = merge_dicts(metadata, elem.fields)
f.write(json.dumps(full_record) + '\n')
# Get the next element
elem = rec.get_next_elem()
def create_trie_from_bgpstream_info(self, interval_start=1451692800):
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('collector', self.collector_name)
stream.add_filter('record-type', 'ribs')
if isinstance(interval_start, datetime.datetime):
interval_start_utc = self.convert_interval_to_utc(interval_start)
stream.add_interval_filter(interval_start_utc - 300, interval_start_utc + 300)
else:
stream.add_interval_filter(interval_start - 300, interval_start + 300)
stream.start()
while (stream.get_next_record(rec)):
elem = rec.get_next_elem()
while elem:
# Get the peer ASN and IP. We then construct a peer_id, since a collector
# can establish multiple connections with the same ASN.
peer_asn = elem.peer_asn
peer_asn_ip = elem.peer_address
# make this an unmodifiable tuple
peer_id = (peer_asn, peer_asn_ip)
peer_route_trie = self.peer_trie_handles[peer_id]
trie_node = peer_route_trie.add(elem.fields['prefix'])
trie_node.data['as-path'] = elem.fields['as-path']
elem = rec.get_next_elem()
def _get_data(self, prefix, datetime):
"""
output example: [['15547', '8220', '1853', '1205'],[..another AS path..]]
:param prefix:
:param datetime: end interval
:return: list of AS paths
"""
print('[*] ris.py: _get_data() called')
print('[*] ris.py: _get_data() prefix: {}'.format(prefix))
start = int(datetime) - 20000 # 20000 second seems to be the shortest interval to get data from BGPstream
stop = int(datetime)
result = []
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('prefix', prefix)
stream.add_filter('record-type', 'ribs')
stream.add_filter('project', 'ris')
stream.add_interval_filter(start, stop)
stream.start()
while stream.get_next_record(rec):
if rec.status == "valid":
elem = rec.get_next_elem()
while elem:
as_path = elem.fields['as-path'].split()
as_path.append(' ') # for tree creation
result.append(as_path)
elem = rec.get_next_elem()
print('[*] ris.py: _get_data() finished.')
return result
def downloader(start_date, duration):
"""Download BGP paths from Routeviews and RIPE NCC from a start date for a certain duration."""
# Start of UNIX time
base = int(datetime.datetime.strptime(start_date, '%m/%d/%Y').strftime('%s'))
# Create a new bgpstream instance and a reusable bgprecord instance
stream = BGPStream()
rec = BGPRecord()
# Consider 1-day interval:
stream.add_interval_filter(base, base + int(duration))
stream.add_filter('record-type', 'ribs')
stream.start()
path_set = set()
f = open('rib.txt', 'w')
while(stream.get_next_record(rec)):
if rec.status != "valid":
continue
else:
elem = rec.get_next_elem()
while(elem):
path = elem.fields['as-path']
if '{' in path or '(' in path:
elem = rec.get_next_elem()
continue
prefix = elem.fields['prefix']
# Focus on IPv4 prefixes
if ":" not in prefix and path not in path_set:
f.write(path.replace(' ', '|') + '\n')
path_set.add(path)
elem = rec.get_next_elem()
f.close()
def sendMessageToKafka(self, col_name, col_data):
print "lalalala"
stream = BGPStream()
record = BGPRecord()
time_start = int(col_data.get('ribs').get('latestDumpTime'))
time_end = time_start + int(col_data.get('ribs').get('dumpPeriod'))
stream.add_filter('collector', col_name)
stream.add_filter('record-type', 'ribs')
stream.add_interval_filter(time_start, time_end)
print "Before Start"
stream.start()
print "After Start"
while stream.get_next_record(record):
if record.status == "valid":
elem = record.get_next_elem()
while elem:
# print "Record:{}".format(elem)
producer = Kafka_producer()
producer.send_data(col_name, json.dumps(elem))
elem = record.get_next_elem()
else:
print "## Current record not valid!"
print "One Collector Finished"
def next(self):
rec = BGPRecord()
print('get next rec')
if stream.get_next_record(rec):
if rec.status == 'valid':
elem = rec.get_next_elem()
timestamp = rec.time
nlri = []
withdraw = []
attr = {}
while (elem):
peer_address = elem.peer_address
peer_asn = elem.peer_asn
if elem.type == 'A' or elem.type == 'R':
nlri.append(elem.fields['prefix'])
attr['as_path'] = elem.fields['as-path']
attr['nexthop'] = elem.fields['next-hop']
attr['community'] = elem.fields['communities']
elif elem.type == 'W':
withdraw.append(elem.fields['prefix'])
elem = rec.get_next_elem()
return (timestamp, attr, nlri, withdraw)
else:
return self.next(rec)
else:
StopIteration
def get_data_graph(self):
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('record-type', 'ribs')
stream.add_interval_filter(self.start, self.end)
stream.start()
as_graph = nx.Graph()
while stream.get_next_record(rec):
if rec.status == "valid":
elem = rec.get_next_elem()
while elem:
# the list is a list, which stores an as-path
as_path = [
k
for k, g in groupby(elem.fields['as-path'].split(" "))
]
peer = str(elem.peer_asn)
# judge whether the as-path is legal
if len(as_path) > 1 and as_path[0] == peer:
if re.match(self.pattern, elem.fields['prefix']):
self.as_prefix[as_path[-1]].add(
elem.fields['prefix'])
# add edges to the graph
for i in range(0, len(as_path) - 1):
as_graph.add_edge(as_path[i], as_path[i + 1])
elem = rec.get_next_elem()
as_graph = self.add_geo_loc(as_graph)
return as_graph
def __init__(self):
# Create a new bgpstream instance
# and a reusable bgprecord instances
self.stream = BGPStream()
self.rec = BGPRecord()
self.origin_ases = set()
self.bgp_lens = defaultdict(lambda: defaultdict(lambda: None))
def build_sql_db():
conn = sqlite3.connect('bgp_stage.db')
c = conn.cursor()
# Create a new bgpstream instance and a reusable bgprecord instance
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('collector','rrc11')
stream.add_interval_filter(1438417216,1438417316)
stream.start()
while(stream.get_next_record(rec)):
if rec.status != "valid":
continue
else:
elem = rec.get_next_elem()
while(elem):
if elem.type == "AB":
prefix = elem.fields["prefix"]
as_path = elem.fields["as-path"].split(" ")
origin = as_path[-1]
time = elem.time
#IP Prefix database
ip_min, ip_max = calculate_min_max(prefix)
c.execute("SELECT ip_min FROM prefix_as WHERE ip_min = (?) AND ip_max = (?) AND as_o = (?)", (ip_min, ip_max, origin))
row = c.fetchone()
if len(row) != 0:
c.execute("UPDATE prefix_as SET count = count + 1 WHERE ip_min = (?) AND ip_max = (?) AND as_o = (?)", (ip_min, ip_max, origin))
else:
c.execute("INSERT INTO prefix_as VALUES(?,?,?,?,?)", (ip_min, ip_max, origin, 1, time))
#AS link database
for as1,as2 in zip(as_path, as_path[1:]) :
c.execute("SELECT as_o FROM as_link WHERE as_o = (?) AND as_n = (?)",(as1,as2))
row = c.fetchone()
if len(row) != 0:
c.execute("UPDATE as_link SET count = count + 1 WHERE as_o = (?) AND as_n = (?)",
(as1, as2))
else:
c.execute("INSERT INTO as_link VALUES(?,?,?,?)", (as1, as2, 1, 0))
elif elem.type == "WA":
prefix = elem.fields["prefix"]
time = elem.time
#Needs research
print(rec.project, rec.collector, rec.type, rec.time, rec.status,
elem.type, elem.peer_address, elem.peer_asn, elem.fields)
print(prefix,elem.time, "W")
print(rec.project, rec.collector, rec.type, rec.time, rec.status,
elem.type, elem.peer_address, elem.peer_asn, elem.fields)
elem = rec.get_next_elem()
conn.commit()
conn.close()
def dump2file(type, name, latestDumpTime, dumpDuration, dumpPeriod):
print type + ' of ' + name + ': '
csv_header = ['type', 'addr', 'as', 'prefix', 'next_hop', 'as_path']
_file = None
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('collector', c_name)
if type == 'ribs':
stream.add_filter('record-type', 'ribs')
_file = open(c_name + '_ribs.csv', 'w+')
elif type == 'updates':
stream.add_filter('record-type', 'updates')
_file = open(c_name + '_updates.csv', 'w+')
stream.add_interval_filter(latestDumpTime, latestDumpTime + dumpPeriod)
stream.start()
count = 0
useless_c = 0
writer = csv.writer(_file)
writer.writerow(csv_header)
# Get next record
while (stream.get_next_record(rec)):
# Print the record information only if it is not a valid record
if rec.status != "valid":
# print rec.project, rec.collector, rec.type, rec.time, rec.status
print 'current rec not valid.'
else:
elem = rec.get_next_elem()
while (elem):
useless_c += 1
if useless_c % 1000 == 0:
print 'Got ' + str(useless_c) + ' elem totally.'
# Print record and elem information
if isIPV6(elem): # ipv6 packet ignored
elem = rec.get_next_elem()
continue
count += 1
# print rec.project, rec.collector, rec.type, rec.time, rec.status,
# print elem.type, elem.peer_address, elem.peer_asn
# elem.fields contains four column: communities, next-hop, prefix, as-path
field = elem.fields
prefix = field['prefix'] if 'prefix' in field.keys() else ''
next_hop = field['next-hop'] if 'next-hop' in field.keys(
) else ''
as_path = field['as-path'] if 'as-path' in field.keys() else ''
as_path = as_path.replace(' ', '|')
writer.writerow([
elem.type, elem.peer_address, elem.peer_asn, prefix,
next_hop, as_path
])
elem = rec.get_next_elem()
_file.close()
print 'count: ' + str(count)
return count
def main():
(options, args) = getopts()
start = options.start_time
end = options.end_time
target_prefs = Set()
with open('./../../atlas/anchor_prefix.txt', 'rb') as br:
for l in br:
target_prefs.add(l.strip())
# Create a new bgpstream instance and a reusable bgprecord instance
stream = BGPStream()
rec = BGPRecord()
with open('./data/stream_{0}'.format(start), 'wb') as bw:
#for pref in target_prefs:
for ptmp in target_prefs:
stream.add_filter('prefix', ptmp)
# stream.add_filter('prefix','0.0.0.0/0')
# Consider RIPE RRC 10 only
stream.add_filter('record-type', 'updates')
stream.add_filter('collector', 'rrc00')
# Consider this time interval:
# Sat Aug 1 08:20:11 UTC 2015
# stream.add_interval_filter(1438417216,1438417216)
# stream.add_interval_filter(1451606400,1454785264
stream.add_interval_filter(start, end)
# Start the stream
stream.start()
# Get next record
cnt = 0
while stream.get_next_record(rec):
# Print the record information only if it is not a valid record
if rec.status != "valid":
pass
# print '*', rec.project, rec.collector, rec.type, rec.time, rec.status
else:
cnt += 1
elem = rec.get_next_elem()
while elem:
if elem.type == 'S':
continue
# Print record and elem information
# print rec.project, rec.collector, rec.type, rec.time, rec.status,
# print elem.type, elem.peer_address, elem.peer_asn, elem.fields, elem.pref
bw.write(
'{0}\t{1}\t{2}\t{3}\t{4}\t{5}\t{6}\t{7}\t{8}\t{9}\n'.
format(rec.project, rec.collector, rec.type, rec.time,
rec.status, elem.type, elem.fields['prefix'],
elem.peer_address, elem.peer_asn, elem.fields))
bw.flush()
elem = rec.get_next_elem()
print 'Successful termination; Start time: {0}'.format(start)
def iterate_stream(stream, collector):
rec = BGPRecord()
while stream.get_next_record(rec):
elem = rec.get_next_elem()
while elem:
for item in bgpstream_format(collector, elem):
yield item
elem = rec.get_next_elem()
def iterate_stream(stream, collector):
rec = BGPRecord()
while stream.get_next_record(rec):
elem = rec.get_next_elem()
while elem:
for item in bgpstream_format(collector, elem):
yield item
elem = rec.get_next_elem()
def run_bgpstream(args):
(collector, start_time, end_time, data_type) = args
# initialize and configure BGPStream
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('collector', collector)
# NB: BGPStream uses inclusive/inclusive intervals, so subtract one off the
# end time since we are using inclusive/exclusive intervals
stream.add_interval_filter(start_time, end_time-1)
stream.add_filter('record-type', data_type)
stream.start()
# per-peer data
peers_data = {}
# loop over all records in the stream
while stream.get_next_record(rec):
elem = rec.get_next_elem()
# loop over all elems in the record
while elem:
# create a peer signature for this elem
sig = peer_signature(rec, elem)
# if this is the first time we have ever seen this peer, create
# an empty result: (dict(Pfx_v4)=Pfx_origins, dict(Pfx_v6)=Pfx_origins))
if sig not in peers_data:
peers_data[sig] =[{},{}]
if('prefix' in elem.fields):
pfx=elem.fields['prefix'];
origin=""
if('as-path' in elem.fields):
path_split=elem.fields['as-path'].split()
if(len(path_split)!=0):
origin=path_split[len(path_split)-1]
if(":" in pfx):
if(pfx not in peers_data[sig][1]):
peers_data[sig][1][pfx]=set()
#discard as origin: AS sets, and ASN=23456 [AS_TRANS]
if(origin!="" and origin!="23456" and "{" not in origin): peers_data[sig][1][pfx].add(origin)
else:
if(pfx not in peers_data[sig][0]):
peers_data[sig][0][pfx]=set()
#discard as origin: AS sets, and ASN=23456 [AS_TRANS]
if(origin!="" and origin!="23456" and "{" not in origin): peers_data[sig][0][pfx].add(origin)
elem = rec.get_next_elem()
# the time in the output row is truncated down to a multiple of
# RESULT_GRANULARITY so that slices can be merged correctly
start_time = \
int(math.floor(start_time/RESULT_GRANULARITY) * RESULT_GRANULARITY)
# for each peer that we processed data for, create an output row
return [((start_time, collector, p), (peers_data[p])) for p in peers_data]
def path_finder(self):
logging.info(
f"[ZombieRecordFinder-{self.collector}] starting path_finder()")
result_path = self.config['DEFAULT']['Result']
dump_path = dict()
stream = self.get_stream()
stream.start()
rec = BGPRecord()
try:
while stream and stream.get_next_record(rec):
if rec.status != "valid":
continue
if rec.type == "unknown":
continue
recordTimeStamp = int(rec.time)
while self.changing and self.changing[-1][0] < recordTimeStamp:
_, prefix = self.changing.pop()
self.watching_prefix.add(prefix)
while self.zombies and self.zombies[-1][0] < recordTimeStamp:
ts, prefix = self.zombies.pop()
dump_path[f"{prefix}|{ts}"] = copy.deepcopy(
self.path[prefix])
if prefix in self.watching_prefix:
self.watching_prefix.remove(prefix)
f = open(
f"{result_path}/{self.year}-{self.month}-changing-{self.collector}.txt",
"a+")
data = ",".join(self.record[prefix])
f.write(f"{prefix} {ts} ? {data} \n")
f.close()
del data
del self.record[prefix]
else:
logging.warning(
f"[ZombieRecordFinder-{self.collector}] trying to remove unwatched prefix {prefix}"
)
elem = rec.get_next_elem()
while (elem):
self.analyze_element(elem, recordTimeStamp)
elem = rec.get_next_elem()
except Exception as e:
logging.error(
f"[ZombieRecordFinder-{self.collector}] exit with error : {e}")
finally:
with open(
f"{result_path}/{self.year}-{self.month}-zombie-record-finder-{self.collector}.json",
'w') as fp:
json.dump(dump_path, fp)
def main():
(options, args) = getopts()
start = options.start_time
end = options.end_time
target_prefs = Set()
with open('./../../atlas/anchor_prefix.txt', 'rb') as br:
for l in br:
target_prefs.add(l.strip())
# Create a new bgpstream instance and a reusable bgprecord instance
stream = BGPStream()
rec = BGPRecord()
with open('./data/stream_{0}'.format(start), 'wb') as bw:
#for pref in target_prefs:
for ptmp in target_prefs:
stream.add_filter('prefix', ptmp)
# stream.add_filter('prefix','0.0.0.0/0')
# Consider RIPE RRC 10 only
stream.add_filter('record-type', 'updates')
stream.add_filter('collector', 'rrc00')
# Consider this time interval:
# Sat Aug 1 08:20:11 UTC 2015
# stream.add_interval_filter(1438417216,1438417216)
# stream.add_interval_filter(1451606400,1454785264
stream.add_interval_filter(start, end)
# Start the stream
stream.start()
# Get next record
cnt = 0
while stream.get_next_record(rec):
# Print the record information only if it is not a valid record
if rec.status != "valid":
pass
# print '*', rec.project, rec.collector, rec.type, rec.time, rec.status
else:
cnt += 1
elem = rec.get_next_elem()
while elem:
if elem.type == 'S':
continue
# Print record and elem information
# print rec.project, rec.collector, rec.type, rec.time, rec.status,
# print elem.type, elem.peer_address, elem.peer_asn, elem.fields, elem.pref
bw.write('{0}\t{1}\t{2}\t{3}\t{4}\t{5}\t{6}\t{7}\t{8}\t{9}\n'.format(
rec.project, rec.collector, rec.type, rec.time, rec.status,
elem.type, elem.fields['prefix'], elem.peer_address, elem.peer_asn, elem.fields))
bw.flush()
elem = rec.get_next_elem()
print 'Successful termination; Start time: {0}'.format(start)
def __init__(self, route_collector="rrc00", rpki_validator="rpki-validator.realmv6.org:8282"):
self.rc = route_collector
rpki = rpki_validator.split(":")
self.mgr = RTRManager(rpki[0], rpki[1])
# self._start_rtr_manager()
self.stream = BGPStream()
self.rec = BGPRecord()
def main():
(options, args) = getopts()
limit = options.limit
start = options.start_time
end = options.end_time
# Create a new bgpstream instance and a reusable bgprecord instance
stream = BGPStream()
rec = BGPRecord()
# Consider RIPE RRC 10 only
stream.add_filter('record-type', 'updates')
stream.add_filter('collector', 'rrc00')
stream.add_filter('prefix', '0.0.0.0/0')
# Consider this time interval:
# Sat Aug 1 08:20:11 UTC 2015
# stream.add_interval_filter(1438417216,1438417216)
# stream.add_interval_filter(1451606400,1454785264)
#stream.add_interval_filter(1454630400, 1454716800)
# 1 hour
#1454284800 - 1454288400
stream.add_interval_filter(start, end)
# Start the stream
stream.start()
# Get next record
prefixes_update = defaultdict(int)
prefixes_withdraw = defaultdict(int)
while stream.get_next_record(rec):
# Print the record information only if it is not a valid record
if rec.status != "valid":
pass
# print '*', rec.project, rec.collector, rec.type, rec.time, rec.status
else:
elem = rec.get_next_elem()
while elem:
if elem.type == 'A':
#print elem.fields['as-path']
prefixes_update[elem.fields['prefix']] += 1
if elem.type == 'W':
prefixes_withdraw[elem.fields['prefix']] += 1
#print rec.project, rec.collector, rec.type, rec.time, rec.status,
#print elem.type, elem.peer_address, elem.peer_asn, elem.fields
elem = rec.get_next_elem()
for k in prefixes_update:
if prefixes_update[k] >= limit:
print k + "\t" + str(prefixes_update[k]) + "\t" + str(
prefixes_withdraw[k])
def main():
(options, args) = getopts()
limit = options.limit
start = options.start_time
end = options.end_time
# Create a new bgpstream instance and a reusable bgprecord instance
stream = BGPStream()
rec = BGPRecord()
# Consider RIPE RRC 10 only
stream.add_filter('record-type', 'updates')
stream.add_filter('collector', 'rrc00')
stream.add_filter('prefix', '0.0.0.0/0')
# Consider this time interval:
# Sat Aug 1 08:20:11 UTC 2015
# stream.add_interval_filter(1438417216,1438417216)
# stream.add_interval_filter(1451606400,1454785264)
#stream.add_interval_filter(1454630400, 1454716800)
# 1 hour
#1454284800 - 1454288400
stream.add_interval_filter(start, end)
# Start the stream
stream.start()
# Get next record
prefixes_update = defaultdict(int)
prefixes_withdraw = defaultdict(int)
while stream.get_next_record(rec):
# Print the record information only if it is not a valid record
if rec.status != "valid":
pass
# print '*', rec.project, rec.collector, rec.type, rec.time, rec.status
else:
elem = rec.get_next_elem()
while elem:
if elem.type == 'A':
#print elem.fields['as-path']
prefixes_update[elem.fields['prefix']] += 1
if elem.type == 'W':
prefixes_withdraw[elem.fields['prefix']] += 1
#print rec.project, rec.collector, rec.type, rec.time, rec.status,
#print elem.type, elem.peer_address, elem.peer_asn, elem.fields
elem = rec.get_next_elem()
for k in prefixes_update:
if prefixes_update[k] >= limit:
print k + "\t" + str(prefixes_update[k]) + "\t" + str(prefixes_withdraw[k])
def run(self):
stream = BGPStream()
rec = BGPRecord()
if self.upd_file is None:
stream.add_filter('collector', self.collector)
stream.add_filter('record-type', self.record_type)
stream.add_interval_filter(self.from_date, self.to_date)
stream.set_live_mode()
else:
stream.set_data_interface('singlefile')
if self.upd_file:
stream.set_data_interface_option('singlefile', 'upd-file',
self.upd_file)
if self.rib_file:
stream.set_data_interface_option('singlefile', 'rib-file',
self.rib_file)
if self.prefix_filter is not None:
for prefix in self.prefix_filter:
stream.add_filter('prefix', prefix)
if self.peer_as_filter:
for asn in self.peer_as_filter:
stream.add_filter('peer-asn', str(asn))
if self.communities_filter:
for community in self.communities_filter:
stream.add_filter('community', community)
stream.start()
stream.get_next_record(rec)
prev = rec.time
while (stream.get_next_record(rec)):
now = rec.time
if rec.status == 'valid':
elem = rec.get_next_elem()
while (elem):
statement = None
peer_address = elem.peer_address
peer_asn = elem.peer_asn
if peer_asn in self.asn_to_nexthop:
if elem.type == 'A' or elem.type == 'R':
prefix = elem.fields['prefix']
as_path = elem.fields['as-path']
nexthop = elem.fields['next-hop']
if peer_asn in self.asn_to_nexthop:
nexthop = self.asn_to_nexthop[peer_asn]
statement = 'announce route %s next-hop %s as-path' \
' [ %s ]' % (prefix, nexthop, as_path)
elif elem.type == 'W':
prefix = elem.fields['prefix']
statement = 'withdraw route %s' % prefix
if statement:
sys.stdout.write("%s\n" % statement)
sys.stdout.flush()
elem = rec.get_next_elem()
time.sleep(self.delay + now - prev)
prev = now
def download_data():
peer_state = defaultdict(dict)
results = defaultdict(defaultdict_list)
current_bin = 0
# create a new bgpstream instance
stream = BGPStream()
# create a reusable bgprecord instance
rec = BGPRecord()
bgprFilter = "type updates"
bgprFilter += " and project ris "
for prefix in prefixes:
bgprFilter += " and prefix more %s " % prefix
logging.info("Connecting to BGPstream... (%s)" % bgprFilter)
logging.info("Timestamps: %s, %s" % (startts, endts))
stream.parse_filter_string(bgprFilter)
stream.add_interval_filter(startts, endts)
stream.start()
while (stream.get_next_record(rec)):
if rec.status != "valid":
print(rec.project, rec.collector, rec.type, rec.time, rec.status)
# from IPython import embed
# embed()
if current_bin == 0:
current_bin = rec.time
# slide the time window:
if current_bin + bin_size < rec.time:
timebins = range(current_bin, rec.time, bin_size)
for i in timebins[:-1]:
results["other"]["timebin"].append(i)
for pfx, p_s in peer_state.items():
for peeras, state in p_s.items():
results[pfx][peeras].append(state)
current_bin = timebins[-1]
elem = rec.get_next_elem()
while (elem):
# peerip g= elem.peer_address
peeras = elem.peer_asn
prefix = elem.fields["prefix"]
peer_state[prefix][peeras] = elem.type
elem = rec.get_next_elem()
return results
def main():
parser = argparse.ArgumentParser()
parser.formatter_class = argparse.RawDescriptionHelpFormatter
parser.description = textwrap.dedent('''\
a proof-of-concept utility for watching updates from BGPstream
and then printing out if an unexpected update is heard
''')
parser.epilog = textwrap.dedent('''\
Example: watch these route announcements
%(prog)s -f routes.yaml ''')
required = parser.add_argument_group('required arguments')
required.add_argument("-f",
"--file",
required=True,
help="yaml file of prefixes to origin asn")
parser.add_argument("-d",
"--debug",
action='store_true',
help="print out all updates containing these prefixes")
args = parser.parse_args()
routes = pytricia.PyTricia(48) # longest reasonable pfx in dfz
with open(args.file, 'r') as f:
routesfile = yaml.safe_load(f)
for pfx in routesfile:
routes[pfx] = routesfile[pfx]
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('record-type', 'updates')
stream.add_interval_filter(int(time.time()), 0)
stream.set_live_mode()
stream.start()
while (stream.get_next_record(rec)):
if rec.status == 'valid':
elem = rec.get_next_elem()
while (elem):
if 'as-path' in elem.fields:
path = elem.fields['as-path'].split()
prefix = elem.fields['prefix']
if prefix in routes and (routes[prefix] != path[-1]
or args.debug):
print('Heard prefix:', elem.fields['prefix'],
'AS-PATH:', elem.fields['as-path'],
' Found by project:', rec.project, 'collector:',
rec.collector, 'type:', rec.type, 'at time:',
rec.time, 'Type:', elem.type, 'Peer:',
elem.peer_address, 'AS', elem.peer_asn)
elem = rec.get_next_elem()
def run_bgpstream(args):
(collector, start_time, end_time, data_type) = args
# initialize and configure BGPStream
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('collector', collector)
# NB: BGPStream uses inclusive/inclusive intervals, so subtract one off the
# end time since we are using inclusive/exclusive intervals
stream.add_interval_filter(start_time, end_time - 1)
stream.add_filter('record-type', data_type)
stream.start()
# per-peer data
peers_data = {}
# loop over all records in the stream
while stream.get_next_record(rec):
elem = rec.get_next_elem()
# to track the peers that have elems in this record
peer_signatures = set()
# loop over all elems in the record
while elem:
# create a peer signature for this elem
sig = peer_signature(rec, elem)
peer_signatures.add(sig)
# if this is the first time we have ever seen this peer, create
# an empty result: (elem_cnt, peer_record_cnt, coll_record_cnt)
if sig not in peers_data:
peers_data[sig] = [0, 0, 0]
peers_data[sig][0] += 1 # increment elem cnt for this peer
elem = rec.get_next_elem()
# done with elems, increment the 'coll_record_cnt' field for just
# one peer that was present in this record (allows a true, per-collector
# count of records since each record can contain elems for many peers)
if len(peer_signatures):
first = True
for sig in peer_signatures: # increment peer_record_cnt for all
if first:
peers_data[sig][2] += 1 # increment the coll_record_cnt
first = False
peers_data[sig][1] += 1
# the time in the output row is truncated down to a multiple of
# RESULT_GRANULARITY so that slices can be merged correctly
start_time = \
int(math.floor(start_time/RESULT_GRANULARITY) * RESULT_GRANULARITY)
# for each peer that we processed data for, create an output row
return [((start_time, collector, p), (peers_data[p])) for p in peers_data]
def MOAS_prefixes():
# Create a new bgpstream instance and a reusable bgprecord instance
stream = BGPStream()
rec = BGPRecord()
# Consider Route Views Singapore only
stream.add_filter('collector', 'route-views.sg')
# Consider RIBs dumps only
stream.add_filter('record-type', 'ribs')
# Consider this time interval:
# Sat, 01 Aug 2015 7:50:00 GMT - 08:10:00 GMT
stream.add_interval_filter(1438415400, 1438416600)
# Start the stream
stream.start()
# <prefix, origin-ASns-set > dictionary
prefix_origin = defaultdict(set)
prefix_origin_dict = defaultdict(list)
# Get next record
while (stream.get_next_record(rec)):
elem = rec.get_next_elem()
while (elem):
# Get the prefix
pfx = elem.fields['prefix']
# Get the list of ASes in the AS path
ases = elem.fields['as-path'].split(" ")
if len(ases) > 0:
# Get the origin ASn (rightmost)
origin = ases[-1]
if 262857 == origin: #AS262857 - UFRN: AS262857 177.20.128.0/19 UNIVERSIDADE FEDERAL DO RIO GRANDE DO NORTE
'''
BI 177.20.128.0/19 198.32.125.84 280 100 0 1916, 262857 IGP
E 177.20.128.0/19 213.248.67.117 0 70 0 1299, 2914, 1916, 262857 IGP
E 177.20.128.0/19 213.248.98.93 0 70 0 1299, 2914, 1916, 262857 IGP
'''
print "Achou UFRN"
# Insert the origin ASn in the set of
# origins for the prefix
prefix_origin[pfx].add(origin)
prefix_origin_dict[pfx].append(ases)
elem = rec.get_next_elem()
# Print the list of MOAS prefix and their origin ASns
for pfx in prefix_origin:
if len(prefix_origin[pfx]) > 1:
pass #print pfx, ",".join(prefix_origin[pfx])
def pushRIBData(AF, collector, startts, endts):
stream = getBGPStream("ribs", AF, [collector], startts, endts)
topicName = "ihr_bgp_" + collector + "_rib"
admin_client = KafkaAdminClient(
bootstrap_servers=['kafka1:9092', 'kafka2:9092', 'kafka3:9092'],
client_id='bgp_producer_admin')
try:
topic_list = [
NewTopic(name=topicName, num_partitions=1, replication_factor=1)
]
admin_client.create_topics(new_topics=topic_list, validate_only=False)
except:
pass
admin_client.close()
stream.start()
producer = KafkaProducer(
bootstrap_servers=['kafka1:9092', 'kafka2:9092', 'kafka3:9092'],
# acks=0,
value_serializer=lambda v: msgpack.packb(v, use_bin_type=True),
linger_ms=1000,
request_timeout_ms=300000,
compression_type='snappy')
rec = BGPRecord()
while stream and stream.get_next_record(rec):
completeRecord = {}
completeRecord["rec"] = getRecordDict(rec)
completeRecord["elements"] = []
recordTimeStamp = rec.time
recordTimeStamp = int(recordTimeStamp) * 1000
elem = rec.get_next_elem()
while (elem):
elementDict = getElementDict(elem)
completeRecord["elements"].append(elementDict)
elem = rec.get_next_elem()
producer.send(topicName, completeRecord, timestamp_ms=recordTimeStamp)
producer.close()
def getting_BGP_update():
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('collector', 'rrc11')
stream.add_interval_filter(1438417216, 1438417216)
# Start the stream
stream.start()
while (stream.get_next_record(rec)):
if rec.status != "valid":
print rec.project, rec.collector, rec.type, rec.time, rec.status
else:
elem = rec.get_next_elem()
while (elem):
print rec.project, rec.collector, rec.type, rec.time, rec.status,
print elem.type, elem.peer_address, elem.peer_asn, elem.fields
elem = rec.get_next_elem()
def generate_stream():
bs = BGPStream()
rec = BGPRecord()
#initialize MySql
a = MySqlDAL()
a.start()
#initialize the publisher in port number 12345
publisher = ZmqPublisher(12345)
bs.add_interval_filter(calendar.timegm(time.gmtime()) - delay, 0)
# bs.add_filter('collector', 'route-views.sg')
bs.add_filter('record-type', 'updates')
bs.start()
print('Beginning to read from stream')
input_id = 0
while bs.get_next_record(rec):
elem = rec.get_next_elem()
while elem is not None:
# sleep until it is time to send this record
'''
now = calendar.timegm(time.gmtime())
sim_time = now - delay
if elem.time > sim_time:
time.sleep(elem.time - sim_time)
'''
if elem.type not in ['A', 'W']:
continue
input_id += 1
msg = elem2bgplay(rec, elem)
msg['type'] = 'A'
msg['id'] = input_id
print(msg)
# Publish the message
publisher.publish(msg)
# Write it to DB
if elem.type == 'A':
a.add(msg)
elif elem.type == 'W':
a.remove(msg)
else:
print "Error: Unknown type: " + elem.type
elem = rec.get_next_elem()
def generate_stream():
bs = BGPStream()
rec = BGPRecord()
#initialize MySql
db_writer = MysqlWriter()
db_writer.start()
#initialize the publisher in port number 12345
publisher = ZmqPublisher(12345)
bs.add_interval_filter(calendar.timegm(time.gmtime()) - delay, 0)
# bs.add_filter('collector', 'route-views.sg')
bs.add_filter('record-type', 'updates')
bs.start()
print('Beginning to read from stream')
input_id = 0
while bs.get_next_record(rec):
elem = rec.get_next_elem()
while elem is not None:
# sleep until it is time to send this record
'''
now = calendar.timegm(time.gmtime())
sim_time = now - delay
if elem.time > sim_time:
time.sleep(elem.time - sim_time)
'''
if elem.type not in ['A', 'W']:
continue
input_id += 1
msg = elem2bgplay(rec, elem)
msg['type'] = 'A'
msg['id'] = input_id
print(msg)
# Publish the message
publisher.publish(msg)
# Write it to DB
if elem.type == 'A':
db_writer.add(msg)
elif elem.type == 'W':
db_writer.remove(msg)
else:
print "Error: Unknown type: " + elem.type
elem = rec.get_next_elem()
def get_ribs(collector, start, period):
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('collector', collector)
stream.add_interval_filter(
int(start.timestamp()) - period, int(start.timestamp()))
stream.add_filter('record-type', 'ribs')
stream.start()
while (stream.get_next_record(rec)):
if rec.status == "valid":
elem = rec.get_next_elem()
while (elem):
yield (rec.project, rec.collector,
datetime.utcfromtimestamp(rec.time).isoformat(),
elem.peer_address, elem.peer_asn, elem.fields['prefix'],
elem.fields['as-path'])
elem = rec.get_next_elem()
def initUpdates(seedingCheckOverride, lastBGPUpdateProcessed): # Consider passing in last BGP update processed because of new system.
global cursor, startTime, endTime, seeded, stream, rec
startTime = lastBGPUpdateProcessed
#cursor.execute("""SELECT intVal AS lastBGPUpdateProcessed FROM metadata WHERE name = 'lastBGPUpdateProcessed'""")
#result = cursor.fetchone()
#if result == None:
# print "ERROR: NO Start Time Found in DB (aka lastBGPUpdateProcessed). Recommended snd time is 1489224750 - int(2592000 * 4) (which is 1478856750) because this is a 4 month backlog."
# exit()
#cursor.execute("""INSERT INTO metadata (name, intVal) VALUES ('lastBGPUpdateProcessed', {0})""".format(1489224750 - int(2592000 * 4)))
#conn.commit()
#else:
# (startTime,) = result
cursor.execute("""SELECT intVal AS endTime FROM metadata WHERE name = 'endTime'""")
result = cursor.fetchone()
if result == None:
print "ERROR: NO End Time Found in DB. Recommended end time is 1489224749 because this is the timestamp of the first cert."
print "ERROR: Other recommended end time is 1491775893 which is when all certs have been covered."
exit()
else:
(endTime,) = result
if not seedingCheckOverride:
cursor.execute("""SELECT stringVal AS seeded FROM metadata WHERE name = 'seeded'""")
result = cursor.fetchone()
if result == None:
# We will assume that the DB is not seeded if there is no entry and not throw any errors in this case.
seeded = False
print "line 59 DB not seeded upon call to initUpdates. DB should be seeded with RIBs before updates can be processed. Call initSeeding first. Exiting."
exit()
else:
strValSeeded = ""
(strValSeeded,) = result
if strValSeeded == "false":
seeded = False
print "line 66 DB not seeded upon call to initUpdates. DB should be seeded with RIBs before updates can be processed. Call initSeeding first. Exiting."
exit()
elif strValSeeded == "true":
seeded = True
else:
print "Invalid value for seeded metadata entry. Exiting."
exit()
else:
seeded = True
# create a new bgpstream instance
stream = BGPStream()
# create a reusable bgprecord instance
rec = BGPRecord()
stream.add_filter('record-type','updates') # here we collect updates. This could be changed to ribs to instead acquire periodic snapshots of the RIBs.
# configure the stream to retrieve Updates from the route-views listener.
stream.add_filter('collector', 'route-views2')
# getting updates only from one peer gives us only the perferred route of this peer and no rejected routes.
# only get updates from level3. Level3 is a provider to ViaWest so their choice is a likely choice of ViaWest.
stream.add_filter('peer-asn', '3356')
stream.add_interval_filter(startTime, endTime)
stream.start()
def recv_bgpstream_updates(begin, until, collector, output_queue):
logging.info ("CALL recv_bgpstream_updates")
# wait for first RIB table dump to complete
while (rib_ts < 0):
time.sleep(RIB_TS_WAIT/10)
time.sleep(RIB_TS_WAIT)
# Create bgpstream
stream = BGPStream()
rec = BGPRecord()
# set filtering
stream.add_filter('collector', collector)
stream.add_filter('record-type','updates')
stream.add_interval_filter(rib_ts,until)
# Start the stream
stream.start()
while (stream.get_next_record(rec)):
if rec.status == 'valid':
elem = rec.get_next_elem()
else:
logging.warn("stream record invalid, skipping ...")
continue
logging.info("Record TS: "+str(rec.time))
while (elem):
logging.info(" -- Record Element Type: " + elem.type + ", TS: " + str(elem.time))
bgp_message = BGPmessage(elem.time, 'update')
src_peer = dict()
src_addr = elem.peer_address
src_asn = elem.peer_asn
src_peer['addr'] = src_addr
src_peer['port'] = 0
src_peer['asn'] = src_asn
bgp_message.set_source(src_peer)
if elem.type.upper() == 'A':
bgp_message.add_announce(elem.fields['prefix'])
bgp_message.set_nexthop(elem.fields['next-hop'])
aspath = elem.fields['as-path'].split()
for a in aspath:
if not '{' in a: # ignore AS-SETs
bgp_message.add_as_to_path(a)
output_queue.put(bgp_message)
elif elem.type.upper() == 'W':
bgp_message.add_withdraw(elem.fields['prefix'])
output_queue.put(bgp_message)
elem = rec.get_next_elem()
def main(rib, target_prefs):
# Create a new bgpstream instance and a reusable bgprecord instance
stream = BGPStream()
rec = BGPRecord()
with open('./data/stream_{0}'.format(start), 'wb') as bw:
for ptmp in target_prefs:
stream.add_filter('prefix', ptmp)
# Consider RIPE RRC 10 only
stream.add_filter('record-type', 'updates')
stream.add_filter('record-type', 'ribs')
#stream.add_filter('collector', 'rrc04')
stream.add_filter('project', 'ris')
stream.add_filter('project', 'routeviews')
stream.add_interval_filter(start-60*60*8, start)
stream.add_rib_period_filter(10000000000000)
# Start the stream
stream.start()
while stream.get_next_record(rec):
# Print the record information only if it is not a valid record
if rec.status != "valid":
continue
#if rec.time < start:
elem = rec.get_next_elem()
while elem:
if elem.type == 'A' or elem.type == 'R':
rib.add_to_rib(rec.collector, elem.peer_address, elem.fields['prefix'], elem.time, elem.fields['as-path'])
elem = rec.get_next_elem()
#else:
rib.flush()
print 'Successful termination; Start time: {0}'.format(start)
def recv_bgpstream_rib(begin, until, collector, output_queue):
logging.info ("CALL recv_bgpstream_rib")
# Create bgpstream
stream = BGPStream()
rec = BGPRecord()
# set filtering
stream.add_filter('collector', collector)
stream.add_filter('record-type','updates')
stream.add_interval_filter(begin,until)
# Start the stream
stream.start()
while (stream.get_next_record(rec)):
global rib_ts
if rec.status == 'valid':
elem = rec.get_next_elem()
else:
logging.warn("stream record invalid, skipping.")
continue
if (rib_ts > 0) and (rec.time > (rib_ts + RIB_TS_INTERVAL/2)):
logging.info("received full RIB table dump.")
break
bgp_message = None
while (elem):
if (elem.type.upper() == 'A') or (elem.type.upper() == 'R'):
rib_ts = elem.time
bgp_message = BGPmessage(elem.time, 'update')
bgp_message.set_nexthop(elem.fields['next-hop'])
src_peer = dict()
src_addr = elem.peer_address
src_asn = elem.peer_asn
src_peer['addr'] = src_addr
src_peer['port'] = 0
src_peer['asn'] = src_asn
bgp_message.set_source(src_peer)
aspath = elem.fields['as-path'].split()
for a in aspath:
if not '{' in a: # ignore AS-SETs
bgp_message.add_as_to_path(a)
bgp_message.add_announce(elem.fields['prefix'])
output_queue.put(bgp_message)
elem = rec.get_next_elem()
def init_stream(config_files, start_time, end_time):
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('project', 'ris')
stream.add_filter('project', 'routeviews')
stream.add_filter('record-type', 'ribs')
for exp_id in config_files:
config_file = config_files[exp_id]
stream.add_filter('prefix', config_file['superprefix'])
stream.add_interval_filter(start_time, end_time)
return stream, rec
def count_valid_withdrawals_per_second(t0,num_seconds):
# div0 in progress tracking if <100 seconds
assert num_seconds>100;
# Create a new bgpstream instance
stream = BGPStream();
# use Routeviews Saopaulo records
stream.add_filter('project','routeviews');
stream.add_filter('collector','route-views.saopaulo');
stream.add_filter('record-type','updates');
# time interval
stream.add_interval_filter(t0,t0+tt-1);
# progress tracking
prog = t0;
# Create a reusable bgprecord instance
rec = BGPRecord();
# list of zeroes
ctr = [0]*tt;
# Start the stream
stream.start();
# iterate over records
while(stream.get_next_record(rec)):
if rec.status == "valid":
elem = rec.get_next_elem();
while(elem):
if elem.type == 'W':
ctr[rec.time-t0] += 1;
elem = rec.get_next_elem();
# el-cheapo progress indication with dots. comment out if you don't want the dots.
# """
if rec.time > prog and (rec.time - t0) % floor(num_seconds/100) == 0:
sys.stdout.write('.');
sys.stdout.flush();
prog = rec.time;
# """
# print newline after the dots
print "";
return ctr;
def stream_ribfirst(stream, filter_peer):
rec = BGPRecord()
rib_started = False
while (stream.get_next_record(rec)):
if rec.status != "valid":
print rec.project, rec.collector, rec.type, rec.time, rec.status
else:
if rec.type == 'rib':
rib_started = True
if rib_started:
elem = rec.get_next_elem()
while (elem):
if elem.peer_address in filter_peer or len(
filter_peer) == 0:
if elem.type == 'W':
bgp_message = 'BGPSTREAM|'+str(rec.collector) \
+'|'+str(elem.type)+'|'+str(elem.peer_address) \
+'|'+str(elem.peer_asn)+'|'+str(elem.time)+'|'+str(elem.fields['prefix'])
elif 'as-path' in elem.fields:
as_path = elem.fields['as-path'].split(
'{')[0].split(',')[0].rstrip(' ')
bgp_message = 'BGPSTREAM|'+str(rec.collector) \
+'|'+str(elem.type)+'|'+str(elem.peer_address) \
+'|'+str(elem.peer_asn)+'|'+str(elem.time)+'|'+str(elem.fields['prefix']) \
+'|'+as_path
else:
bgp_message = None
#print 'YES'+'\t'+str(rec.project)+'\t'+str(rec.collector)+'\t'+str(rec.time)+'\t'+str(rec.type)+'\t'+str(rec.status)
if rec.collector not in peer_set:
peer_set[rec.collector] = set()
peer_set[rec.collector].add(elem.peer_address)
if bgp_message is not None:
yield bgp_message
elem = rec.get_next_elem()
def start_collecting(self, start_timestamp, end_timestamp=0):
self.stream.add_interval_filter(start_timestamp, end_timestamp)
print("Start BGPStream:", start_timestamp, end_timestamp)
self.stream.start()
rec = BGPRecord()
act_dump = "unknown"
while (self.stream.get_next_record(rec)):
self.i += 1
if self.i % 10000 == 0:
print(self.i)
if rec.status == "valid":
if (act_dump != rec.dump_position):
act_dump = rec.dump_position
print('Dump Position:', rec.dump_position)
elem = rec.get_next_elem()
while (elem):
self.counter.update(elem.type)
elem = rec.get_next_elem()
print(self.counter)
def get_data(timestamp):
graph = Graph(password="******")
stream = BGPStream()
rec = BGPRecord()
rec_time = None
# IPv4
stream.add_filter('prefix', '198.41.0.0/24') # A-root
stream.add_filter('prefix', '192.33.4.0/24') # C-root
stream.add_filter('prefix', '199.7.91.0/24') # D-root
stream.add_filter('prefix', '192.203.230.0/24') # E-root, IPv4 only
stream.add_filter('prefix', '192.5.5.0/24') # F-root
stream.add_filter('prefix', '192.112.36.0/24') # G-root, IPv4 only
stream.add_filter('prefix', '198.97.190.0/24') # H-root
stream.add_filter('prefix', '192.36.148.0/24') # I-root
stream.add_filter('prefix', '192.58.128.0/24') # J-root
stream.add_filter('prefix', '193.0.14.0/24') # K-root
stream.add_filter('prefix', '199.7.83.0/24') # L-root
stream.add_filter('prefix', '202.12.27.0/24') # M-root
# IPv6
stream.add_filter('prefix', '2001:503:ba3e::/48') # A
stream.add_filter('prefix', '2001:500:2::/48') # C
stream.add_filter('prefix', '2001:500:2d::/48') # D
stream.add_filter('prefix', '2001:500:2f::/48') # F
stream.add_filter('prefix', '2001:500:1::/48') # H
stream.add_filter('prefix', '2001:7fe::/33') # I
stream.add_filter('prefix', '2001:503:c27::/48') # J
stream.add_filter('prefix', '2001:7fd::/48') # K
stream.add_filter('prefix', '2001:500:9f::/48') # L
stream.add_filter('prefix', '2001:dc3::/32') # M
stream.add_filter('record-type', 'ribs')
# stream.add_filter('collector', 'route-views.soxrs')
stream.add_filter('project', 'routeviews')
stream.add_interval_filter(timestamp, timestamp)
stream.start()
result = {}
while stream.get_next_record(rec):
rec_time = rec.time
if rec.status == "valid":
elem = rec.get_next_elem()
while elem:
print('{} {} {} {} {}'.format(rec.collector, elem.type, elem.peer_address, elem.peer_asn, elem.fields))
as_path = elem.fields['as-path'].split()
as_path.reverse()
prefix = elem.fields['prefix']
if prefix not in result:
result[prefix] = []
result[prefix].append(as_path)
elem = rec.get_next_elem()
# get only unique lists in result
for prefix in result:
result[prefix] = [list(x) for x in set(tuple(x) for x in result[prefix])]
print('timestamp {} ==> result: {}'.format(rec_time, result))
for prefix in result:
for path in result[prefix]:
print('path: {}'.format(path))
cur_node = None
prev_node = None
counter_as_prepend = 0
for index, asn in enumerate(path):
cur_node = asn
graph.run('MERGE(s:asn{{name:"{0}", label:"{0}"}})'.format(asn)) # create new node if not exist.
if index > 0:
if cur_node != prev_node:
query = 'MATCH (s:asn),(d:asn) ' \
'WHERE s.name="{0}" AND d.name="{1}" ' \
'MERGE (s)-[r:TO {{prefix: "{3}", time: {2}, prepended: {4}}}]->(d)'\
.format(cur_node, prev_node, rec_time, prefix, counter_as_prepend)
graph.run(query)
if counter_as_prepend > 0:
counter_as_prepend = 0 # reset
else: # AS prepending
counter_as_prepend += 1
# else: # origin AS
# graph.run('MATCH (s:asn)'
# 'WHERE s.name="{0}"'
# 'SET s.origin="{1}_{2}"'
# 'RETURN s'.format(asn, prefix, rec_time))
prev_node = cur_node
def get_data(timestamp):
graph = Graph(password="******")
stream = BGPStream()
rec = BGPRecord()
rec_time = None
# stream.add_filter('prefix', '198.41.0.0/24') # A-root
# stream.add_filter('prefix', '192.228.79.0/24') # B-root, only 1 site
# stream.add_filter('prefix', '192.33.4.0/24') # C-root
# stream.add_filter('prefix', '199.7.91.0/24') # D-root
# stream.add_filter('prefix', '192.203.230.0/24') # E-root, IPv4 only
# stream.add_filter('prefix', '192.5.5.0/24') # F-root
# stream.add_filter('prefix', '192.112.36.0/24') # G-root, IPv4 only
# stream.add_filter('prefix', '198.97.190.0/24') # H-root
# stream.add_filter('prefix', '192.36.148.0/24') # I-root
# stream.add_filter('prefix', '192.58.128.0/24') # J-root
stream.add_filter('prefix', '193.0.14.0/24') # K-root
# stream.add_filter('prefix', '199.7.83.0/24') # L-root
# stream.add_filter('prefix', '202.12.27.0/24') # M-root
# IPv6
# stream.add_filter('prefix', '2001:503:ba3e::/48') # A
## stream.add_filter('prefix', '2001:500:84::/48') # B, only 1 site
# stream.add_filter('prefix', '2001:500:2::/48') # C
# stream.add_filter('prefix', '2001:500:2d::/48') # D
# stream.add_filter('prefix', '2001:500:2f::/48') # F
# stream.add_filter('prefix', '2001:500:1::/48') # H
# stream.add_filter('prefix', '2001:7fe::/33') # I
# stream.add_filter('prefix', '2001:503:c27::/48') # J
# stream.add_filter('prefix', '2001:7fd::/48') # K
# stream.add_filter('prefix', '2001:500:9f::/48') # L
# stream.add_filter('prefix', '2001:dc3::/32') # M
stream.add_filter('record-type', 'ribs')
# stream.add_filter('collector', 'rrc01')
stream.add_filter('project', 'routeviews')
stream.add_interval_filter(timestamp, timestamp)
stream.start()
result = {}
while stream.get_next_record(rec):
rec_time = rec.time
if rec.status == "valid":
elem = rec.get_next_elem()
while elem:
print rec.collector, elem.type, elem.peer_address, elem.peer_asn, elem.fields
as_path = elem.fields['as-path'].split()
as_path.reverse()
prefix = elem.fields['prefix']
if prefix not in result:
result[prefix] = []
result[prefix].append(as_path)
elem = rec.get_next_elem()
# get only unique lists in result
for prefix in result:
result[prefix] = [list(x) for x in set(tuple(x) for x in result[prefix])]
print('timestamp {} ==> result: {}'.format(rec_time, result))
for prefix in result:
for path in result[prefix]:
print('path: {}'.format(path))
cur_node = None
prev_node = None
counter_as_prepend = 0
for index, asn in enumerate(path):
searched_node = graph.find('asn', property_key='label', property_value=asn)
try:
cur_node = searched_node.next() # see if the AS node is already in the db or not. If yes, cur_node == prev_node
except StopIteration:
cur_node = Node('asn', label=str(asn)) # if not exists, then create a new one
if index > 0:
if index == len(path) - 1:
cur_node['path'] = path # attach AS path to the last ASN
if cur_node != prev_node:
if counter_as_prepend > 0:
cur_node['prepended'] = counter_as_prepend
counter_as_prepend = 0 # reset
peering = Relationship(cur_node, 'TO', prev_node, time=rec_time, prefix=prefix)
# peering['time'] = rec_time
# peering['prefix'] = prefix
graph.create(peering)
else: # AS prepending
counter_as_prepend += 1
prev_node = cur_node
def main():
parser = argparse.ArgumentParser(description='', epilog='')
parser.add_argument('-b', '--begin',
help='Begin date (inclusive), format: yyyy-mm-dd HH:MM',
type=valid_date, required=True)
parser.add_argument('-u', '--until',
help='Until date (exclusive), format: yyyy-mm-dd HH:MM',
type=valid_date, required=True)
parser.add_argument('-c', '--collector',
help='Route collector from RIPE RIS or Route-Views project.',
type=str, required=True)
parser.add_argument('-m', '--mongodb',
help='MongoDB connection parameters.',
type=str, default=None)
parser.add_argument('-l', '--loglevel',
help='Set loglevel [DEBUG,INFO,WARNING,ERROR,CRITICAL].',
type=str, default='WARNING')
args = vars(parser.parse_args())
numeric_level = getattr(logging, args['loglevel'].upper(), None)
if not isinstance(numeric_level, int):
raise ValueError('Invalid log level: %s' % loglevel)
logging.basicConfig(level=numeric_level,
format='%(asctime)s : %(levelname)s : %(message)s')
ts_begin = int((args['begin'] - datetime(1970, 1, 1)).total_seconds())
ts_until = int((args['until'] - datetime(1970, 1, 1)).total_seconds())
mongodbstr = None
if args['mongodb']:
mongodbstr = args['mongodb'].strip()
# BEGIN
logging.info("START")
# Create bgpstream
stream = BGPStream()
rec = BGPRecord()
# set filtering
stream.add_filter('collector',args['collector'])
stream.add_filter('record-type','ribs')
stream.add_interval_filter(ts_begin,ts_until)
# Start the stream
stream.start()
rib_ts = 0
rib_origins = dict()
while(stream.get_next_record(rec)):
if rec.status == 'valid':
elem = rec.get_next_elem()
else:
logging.warn("stream record invalid, skipping ...")
continue
if rec.time > (rib_ts + RIB_TS_INTERVAL):
rib_ts = rec.time
if mongodbstr:
store_rib_origins(rib_ts, rib_origins, mongodbstr)
else:
print_rib_origins(rib_ts, rib_origins)
rib_origins = dict()
while(elem):
prefix = elem.fields['prefix']
aspath = elem.fields['as-path'].split()
for a in aspath: # remove AS-SETs
if '{' in a:
aspath.remove(a)
origin = aspath[-1]
if prefix not in rib_origins:
rib_origins[prefix] = list()
if origin not in rib_origins[prefix]:
rib_origins[prefix].append(origin)
elem = rec.get_next_elem()
import time
from _pybgpstream import BGPStream, BGPRecord, BGPElem
start_time = time.time()
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('prefix','140.78.0.0/16')
stream.add_filter('record-type','ribs')
stream.add_filter('collector','rrc11')
# stream.add_filter('project','ris')
stream.add_interval_filter(1464681000, 1464682200) # 1464682200 ==> 05/31/2016 @ 8:10am (UTC)
stream.start()
#
# while(stream.get_next_record(rec)):
# # Print the record information only if it is not a valid record
# if rec.status == "valid":
# elem = rec.get_next_elem()
# while elem:
# # Print record and elem information
# print rec.collector, elem.peer_address, elem.peer_asn, elem.fields['as-path']
# elem = rec.get_next_elem()
result = []
while stream.get_next_record(rec):
if rec.status == "valid":
elem = rec.get_next_elem()
while elem:
print rec.collector, elem.type, elem.peer_address, elem.peer_asn, elem.fields
def load_data(start, stop, collectors, window, threshold):
peers = {}
# collectors is a list of the collectors we want to include
# Start and stop define the interval we are looking in the data
# Create a new BGPStream instance and a reusable BGPRecord instance
stream = BGPStream()
rec = BGPRecord()
# Add filter for each collector.
# If no collector is mentioned, it will consider 16 of them
if collectors:
for collector in collectors:
print collector
stream.add_filter('collector', collector)
else:
for i in range(0, 10):
stream.add_filter('collector', 'rrc0' + str(i))
for i in range(10, 16):
stream.add_filter('collector', 'rrc' + str(i))
stream.add_filter('record-type', 'updates')
# Consider the interval from "start" to "stop" in seconds since epoch
stream.add_interval_filter(start, stop)
# Start the stream
stream.start()
# For each record (one record = one second, can have multiple elements for the same second) we handle its updates
while stream.get_next_record(rec):
timestamp = rec.time
if rec.status != "valid":
print rec.project, rec.collector, rec.type, timestamp, rec.status
else:
# Go through all elements of the record
elem = rec.get_next_elem()
while elem:
# Consider only the A and W updates
if elem.type not in ['A', 'W']:
elem = rec.get_next_elem()
continue
peer = elem.peer_address
updatetype = elem.type
prefix = elem.fields['prefix']
if peer not in peers:
peers[peer] = {
'A': [],
'W': []
}
update = {'tst': timestamp, 'prefix': prefix}
if updatetype == 'A':
handleUpdate(peers[peer]['A'], burst2writeA, update, peer, updatetype, timestamp, window, threshold)
saveGraphPoint(peers[peer]['A'], updatetype, peer, timestamp, collectors, threshold)
else:
handleUpdate(peers[peer]['W'], burst2writeW, update, peer, updatetype, timestamp, window, threshold)
saveGraphPoint(peers[peer]['W'], updatetype, peer, timestamp, collectors, threshold)
elem = rec.get_next_elem()
# After processing all records, we write the graph json files with the graph points recorded for each peer
for peer in graph_points:
peer_file_name = peer.replace(':', '_')
if not os.path.exists(peer_file_name):
os.makedirs(peer_file_name)
with open(peer_file_name+'/'+peer_file_name + '-graph.json', 'w') as outfile:
json.dump(graph_points[peer], outfile, indent=2)
# Write the last burst of A updates if there is one left
if burst2writeA:
for peer in burst2writeA:
if burst2writeA[peer]:
for timestamp in burst2writeA[peer]:
writeBurst(peer, burst2writeA, 'A', timestamp)
# Write the last burst of W updates if there is one left
if burst2writeW:
for peer in burst2writeW:
if burst2writeW[peer]:
for timestamp in burst2writeW[peer]:
writeBurst(peer, burst2writeW, 'W', timestamp)
# transform csv names in json file to use getJSON in plotGrap
# step to CSV is used to avoid appending to the end of a json file directly as appending
# to a json file overwrite the whole file
jsonlist = []
with open('csv_peernames-'+'-'.join(collectors)+'.csv', 'rb') as f:
reader = csv.reader(f)
for row in reader:
jsonlist.append(row[0])
jsondata = json.dumps(jsonlist, indent=2)
fd = open('json_file_names-' + '-'.join(collectors) + '.json', 'w')
fd.write(jsondata)
fd.close()
def main():
parser = argparse.ArgumentParser(description='', epilog='')
parser.add_argument('-b', '--begin',
help='Begin date (inclusive), format: yyyy-mm-dd HH:MM',
type=valid_date, required=True)
parser.add_argument('-u', '--until',
help='Until date (exclusive), format: yyyy-mm-dd HH:MM',
type=valid_date, required=True)
parser.add_argument('-c', '--collector',
help='Route collector from RIPE RIS or Route-Views project.',
type=str, required=True)
parser.add_argument('-m', '--mongodb',
help='MongoDB connection parameters.',
type=str, default=None)
parser.add_argument('-k', '--keepsnapshots',
help='Keep all snapshots, works only with -s.',
action='store_true')
parser.add_argument('-s', '--snapshot',
help='Enable snapshoting.',
action='store_true')
parser.add_argument('-l', '--loglevel',
help='Set loglevel [DEBUG,INFO,WARNING,ERROR,CRITICAL].',
type=str, default='WARNING')
args = vars(parser.parse_args())
numeric_level = getattr(logging, args['loglevel'].upper(), None)
if not isinstance(numeric_level, int):
raise ValueError('Invalid log level: %s' % loglevel)
logging.basicConfig(level=numeric_level,
format='#> %(asctime)s : %(levelname)s : %(message)s')
ts_begin = int((args['begin'] - datetime(1970, 1, 1)).total_seconds())
ts_until = int((args['until'] - datetime(1970, 1, 1)).total_seconds())
mongodbstr = None
if args['mongodb']:
mongodbstr = args['mongodb'].strip()
rib_ts = 0
rib_origins = dict()
origins_lt = list()
if args['snapshot']:
rib_ts, rib_origins = load_snapshot(mongodbstr)
if rib_ts > ts_begin:
logging.info ("SKIP, found snapshot with newer ts")
ts_begin = rib_ts - RIB_TS_THRESHOLD
# BEGIN
logging.info("START")
# Create bgpstream
stream = BGPStream()
rec = BGPRecord()
# set filtering
stream.add_filter('collector',args['collector'])
stream.add_filter('record-type','ribs')
stream.add_interval_filter(ts_begin,ts_until)
# Start the stream
stream.start()
while(stream.get_next_record(rec)):
if rec.status == 'valid':
elem = rec.get_next_elem()
else:
logging.warn("stream record invalid, skipping ...")
continue
#end if
if rec.time > (rib_ts + RIB_TS_THRESHOLD):
for p in rib_origins:
for o in rib_origins[p]:
if rib_origins[p][o][1] < (rib_ts - RIB_TS_THRESHOLD):
origins_lt.append( (p,o,rib_origins[p][o][0],rib_origins[p][o][1]) )
#end if
#end for
#end for
if args['snapshot'] and (len(rib_origins.keys()) > 0):
store_snapshot(rec.time, rib_origins, mongodbstr)
if not args['keepsnapshots']:
remove_snapshot(rib_ts, mongodbstr)
# end if keepsnapshots
# end if snapshot
rib_ts = rec.time
logging.info("ts: "+str(rib_ts))
if len(origins_lt) > 0:
if mongodbstr:
store_origins_lt(rib_ts,origins_lt, mongodbstr)
else:
print_origins_lt(rib_ts,origins_lt)
#end if
for l in origins_lt:
del rib_origins[l[0]][l[1]]
#end for
origins_lt = list()
# end if
#end if
while(elem):
prefix = elem.fields['prefix']
aspath = elem.fields['as-path'].split()
for a in aspath: # remove AS-SETs
if '{' in a:
aspath.remove(a)
#end if
#end for
origin = aspath[-1]
if prefix not in rib_origins:
rib_origins[prefix] = dict()
#end if
if origin not in rib_origins[prefix]:
rib_origins[prefix][origin] = (rib_ts,rib_ts)
else:
rib_origins[prefix][origin] = (rib_origins[prefix][origin][0],rib_ts)
#end if
elem = rec.get_next_elem()
#end while
#end while
if args['snapshot']:
print "NOTE: remaining origin lifetimes are stored in latest snapshot (%d)!\n" % rib_ts
if (len(rib_origins.keys()) > 0):
store_snapshot(rib_ts, rib_origins, mongodbstr)
# end if
else:
print "NOTE: output remaining origin lifetimes with current ts (%d)\n" % rib_ts
origins_lt = list()
for p in rib_origins:
for o in rib_origins[p]:
origins_lt.append( (p,o,rib_origins[p][o][0],rib_ts) )
if mongodbstr:
store_origins_lt(rib_ts,origins_lt, mongodbstr)
else:
print_origins_lt(rib_ts,origins_lt)
def run_bgpstream(args):
(collector, start_time, end_time, data_type) = args
# initialize and configure BGPStream
stream = BGPStream()
rec = BGPRecord()
stream.add_filter('collector', collector)
# NB: BGPStream uses inclusive/inclusive intervals, so subtract one off the
# end time since we are using inclusive/exclusive intervals
stream.add_interval_filter(start_time, end_time-1)
stream.add_filter('record-type', data_type)
stream.start()
# per-peer data
peers_data = {}
# loop over all records in the stream
while stream.get_next_record(rec):
elem = rec.get_next_elem()
# loop over all elems in the record
while elem:
# create a peer signature for this elem
sig = peer_signature(rec, elem)
# if this is the first time we have ever seen this peer, create
# an empty result: (Pfxs_v4_set , Pfxs_v6_set,
# dict(Transit_ASN)=Pfxs_v4_set , dict(Transit_ASN)=Pfxs_v6_set)
if sig not in peers_data:
peers_data[sig] =[set(),set(),{},{}]
pfx=""
if('prefix' in elem.fields):
pfx=elem.fields['prefix'];
if(":" in pfx):
peers_data[sig][1].add(pfx)
else:
peers_data[sig][0].add(pfx)
if('as-path' in elem.fields):
#Squash the AS Path to keep only distinct ASNs, i.e., remove prepending
path_split = [k for k, g in groupby(elem.fields['as-path'].split(" "))]
if(len(path_split)!=0):
for i in range(1,len(path_split)-1):
transit=path_split[i]
if(":" in pfx):
if(transit not in peers_data[sig][3]):
peers_data[sig][3][transit]=set()
peers_data[sig][3][transit].add(pfx)
elif(pfx!=""):
if(transit not in peers_data[sig][2]):
peers_data[sig][2][transit]=set()
peers_data[sig][2][transit].add(pfx)
elem = rec.get_next_elem()
# the time in the output row is truncated down to a multiple of
# RESULT_GRANULARITY so that slices can be merged correctly
start_time = \
int(math.floor(start_time/RESULT_GRANULARITY) * RESULT_GRANULARITY)
# for each peer that we processed data for, create an output row
return [((start_time, collector, p), (peers_data[p])) for p in peers_data]
