[PATCH v2 1/3] location-importer.in: avoid violating NOT NULL constraints during JOIN

[PATCH v2 1/3] location-importer.in: avoid violating NOT NULL constraints during JOIN

[Peter Müller]

[-- Attachment #1: Type: text/plain, Size: 1148 bytes --]

Signed-off-by: Peter Müller <peter.mueller(a)ipfire.org>
---
 src/python/location-importer.in | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/python/location-importer.in b/src/python/location-importer.in
index f5ae4a9..77952f2 100644
--- a/src/python/location-importer.in
+++ b/src/python/location-importer.in
@@ -360,7 +360,7 @@ class CLI(object):
 					ON COMMIT DROP;
 				CREATE UNIQUE INDEX _autnums_number ON _autnums(number);
 
-				CREATE TEMPORARY TABLE _organizations(handle text, name text)
+				CREATE TEMPORARY TABLE _organizations(handle text, name text NOT NULL)
 					ON COMMIT DROP;
 				CREATE UNIQUE INDEX _organizations_handle ON _organizations(handle);
 			""")
@@ -373,7 +373,7 @@ class CLI(object):
 			self.db.execute("""
 				INSERT INTO autnums(number, name)
 					SELECT _autnums.number, _organizations.name FROM _autnums
-						LEFT JOIN _organizations ON _autnums.organization = _organizations.handle
+						JOIN _organizations ON _autnums.organization = _organizations.handle
 				ON CONFLICT (number) DO UPDATE SET name = excluded.name;
 			""")
 
-- 
2.26.2

[PATCH v2 2/3] importer: Import raw sources for inetnum's again

[Peter Müller]

[-- Attachment #1: Type: text/plain, Size: 5345 bytes --]

The extended feeds do not have enough detailed information
for us, so that we need to import inetnums from RIRs where
possible. Filtering private networks is necessary as RIR data
may contain 0.0.0.0/0 or similar entries for administrative
purposes or due to misfilings.

Special thanks goes to Michael for spending numerous hours
on this, setting up a testing environment and providing helpful
advice while debugging.

Partially fixes: #12458

Cc: Michael Tremer <michael.tremer(a)ipfire.org>
Signed-off-by: Peter Müller <peter.mueller(a)ipfire.org>
---
 src/python/importer.py          | 14 ++++----
 src/python/location-importer.in | 63 +++++++++++++++++++++++++++++++++
 2 files changed, 70 insertions(+), 7 deletions(-)

diff --git a/src/python/importer.py b/src/python/importer.py
index de20f37..f19db4b 100644
--- a/src/python/importer.py
+++ b/src/python/importer.py
@@ -30,8 +30,8 @@ WHOIS_SOURCES = (
 	"https://ftp.afrinic.net/pub/pub/dbase/afrinic.db.gz",
 
 	# Asia Pacific Network Information Centre
-	#"https://ftp.apnic.net/apnic/whois/apnic.db.inet6num.gz",
-	#"https://ftp.apnic.net/apnic/whois/apnic.db.inetnum.gz",
+	"https://ftp.apnic.net/apnic/whois/apnic.db.inet6num.gz",
+	"https://ftp.apnic.net/apnic/whois/apnic.db.inetnum.gz",
 	#"https://ftp.apnic.net/apnic/whois/apnic.db.route6.gz",
 	#"https://ftp.apnic.net/apnic/whois/apnic.db.route.gz",
 	"https://ftp.apnic.net/apnic/whois/apnic.db.aut-num.gz",
@@ -45,8 +45,8 @@ WHOIS_SOURCES = (
 	# XXX ???
 
 	# Réseaux IP Européens
-	#"https://ftp.ripe.net/ripe/dbase/split/ripe.db.inet6num.gz",
-	#"https://ftp.ripe.net/ripe/dbase/split/ripe.db.inetnum.gz",
+	"https://ftp.ripe.net/ripe/dbase/split/ripe.db.inet6num.gz",
+	"https://ftp.ripe.net/ripe/dbase/split/ripe.db.inetnum.gz",
 	#"https://ftp.ripe.net/ripe/dbase/split/ripe.db.route6.gz",
 	#"https://ftp.ripe.net/ripe/dbase/split/ripe.db.route.gz",
 	"https://ftp.ripe.net/ripe/dbase/split/ripe.db.aut-num.gz",
@@ -55,10 +55,10 @@ WHOIS_SOURCES = (
 
 EXTENDED_SOURCES = (
 	# African Network Information Centre
-	"https://ftp.afrinic.net/pub/stats/afrinic/delegated-afrinic-extended-latest",
+	#"https://ftp.afrinic.net/pub/stats/afrinic/delegated-afrinic-extended-latest",
 
 	# Asia Pacific Network Information Centre
-	"https://ftp.apnic.net/apnic/stats/apnic/delegated-apnic-extended-latest",
+	#"https://ftp.apnic.net/apnic/stats/apnic/delegated-apnic-extended-latest",
 
 	# American Registry for Internet Numbers
 	"https://ftp.arin.net/pub/stats/arin/delegated-arin-extended-latest",
@@ -67,7 +67,7 @@ EXTENDED_SOURCES = (
 	"http://ftp.lacnic.net/pub/stats/lacnic/delegated-lacnic-extended-latest",
 
 	# Réseaux IP Européens
-	"https://ftp.ripe.net/pub/stats/ripencc/delegated-ripencc-extended-latest",
+	#"https://ftp.ripe.net/pub/stats/ripencc/delegated-ripencc-extended-latest",
 )
 
 class Downloader(object):
diff --git a/src/python/location-importer.in b/src/python/location-importer.in
index 77952f2..e3a07a0 100644
--- a/src/python/location-importer.in
+++ b/src/python/location-importer.in
@@ -393,6 +393,10 @@ class CLI(object):
 		if line.startswith("aut-num:"):
 			return self._parse_autnum_block(block)
 
+		# inetnum
+		if line.startswith("inet6num:") or line.startswith("inetnum:"):
+			return self._parse_inetnum_block(block)
+
 		# organisation
 		elif line.startswith("organisation:"):
 			return self._parse_org_block(block)
@@ -422,6 +426,65 @@ class CLI(object):
 			autnum.get("asn"), autnum.get("org"),
 		)
 
+	def _parse_inetnum_block(self, block):
+		logging.debug("Parsing inetnum block:")
+
+		inetnum = {}
+		for line in block:
+			logging.debug(line)
+
+			# Split line
+			key, val = split_line(line)
+
+			if key == "inetnum":
+				start_address, delim, end_address = val.partition("-")
+
+				# Strip any excess space
+				start_address, end_address = start_address.rstrip(), end_address.strip()
+
+				# Convert to IP address
+				try:
+					start_address = ipaddress.ip_address(start_address)
+					end_address   = ipaddress.ip_address(end_address)
+				except ValueError:
+					logging.warning("Could not parse line: %s" % line)
+					return
+
+				# Set prefix to default
+				prefix = 32
+
+				# Count number of addresses in this subnet
+				num_addresses = int(end_address) - int(start_address)
+				if num_addresses:
+					prefix -= math.log(num_addresses, 2)
+
+				inetnum["inetnum"] = "%s/%.0f" % (start_address, prefix)
+
+			elif key == "inet6num":
+				inetnum[key] = val
+
+			elif key == "country":
+				if val == "UNITED STATES":
+					val = "US"
+
+				inetnum[key] = val.upper()
+
+		# Skip empty objects
+		if not inetnum:
+			return
+
+		network = ipaddress.ip_network(inetnum.get("inet6num") or inetnum.get("inetnum"), strict=False)
+
+		# Bail out in case we have processed a non-public IP network
+		if network.is_private:
+			logging.warning("Skipping non-globally routable network: %s" % network)
+			return
+
+		self.db.execute("INSERT INTO networks(network, country) \
+			VALUES(%s, %s) ON CONFLICT (network) DO UPDATE SET country = excluded.country",
+			"%s" % network, inetnum.get("country"),
+		)
+
 	def _parse_org_block(self, block):
 		org = {}
 		for line in block:
-- 
2.26.2

[PATCH v2 3/3] importer: Purge any redundant entries

[Peter Müller]

[-- Attachment #1: Type: text/plain, Size: 2241 bytes --]

When importing inetnums, we might import various small networks
which are not relevant for us as long as they do not have a
different country code than their parent network.

Therefore we delete all these entries to keep the database
smaller without losing any information. The second version of this
patch introduces a SQL statement parallelised across all CPUs
available, while the DELETE-statement of the first version literally
took ages to complete.

However, cleaning up those data still takes about 26 hours (!) on
our location02 testing machine, making daily updates of the location
database impossible to the current knowledge.

real    1521m30.620s
user    38m45.521s
sys     9m6.027s

Special thanks goes to Michael for spending numerous hours
on this, setting up a testing environment, doing PostgreSQL magic
and providing helpful advice while debugging.

Partially fixes: #12458

Cc: Michael Tremer <michael.tremer(a)ipfire.org>
Signed-off-by: Peter Müller <peter.mueller(a)ipfire.org>
---
 src/python/location-importer.in | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)

diff --git a/src/python/location-importer.in b/src/python/location-importer.in
index e3a07a0..1467923 100644
--- a/src/python/location-importer.in
+++ b/src/python/location-importer.in
@@ -374,7 +374,27 @@ class CLI(object):
 				INSERT INTO autnums(number, name)
 					SELECT _autnums.number, _organizations.name FROM _autnums
 						JOIN _organizations ON _autnums.organization = _organizations.handle
-				ON CONFLICT (number) DO UPDATE SET name = excluded.name;
+				ON CONFLICT (number) DO UPDATE SET name = excluded.name
+			""")
+
+			self.db.execute("""
+				--- Purge any redundant entries
+				CREATE TEMPORARY TABLE _garbage ON COMMIT DROP
+				AS
+				SELECT network FROM networks candidates
+				WHERE EXISTS (
+					SELECT FROM networks
+					WHERE
+						networks.network << candidates.network
+					AND
+						networks.country = candidates.country
+				);
+
+				CREATE UNIQUE INDEX _garbage_search ON _garbage USING BTREE(network);
+
+				DELETE FROM networks WHERE EXISTS (
+					SELECT FROM _garbage WHERE networks.network = _garbage.network
+				);
 			""")
 
 		# Download all extended sources
-- 
2.26.2

Re: [PATCH v2 3/3] importer: Purge any redundant entries

[Michael Tremer]

[-- Attachment #1: Type: text/plain, Size: 2903 bytes --]

Hello,

I had to revert this patch and the previous one.

For some reason, it is impossible to generate a new database although I made more processor cores and memory available to the system.

Please review this and try to find another approach that has less of an impact on the database system and that enables us to run a database update within about an hour.

Best,
-Michael

> On 20 Sep 2020, at 20:21, Peter Müller <peter.mueller(a)ipfire.org> wrote:
> 
> When importing inetnums, we might import various small networks
> which are not relevant for us as long as they do not have a
> different country code than their parent network.
> 
> Therefore we delete all these entries to keep the database
> smaller without losing any information. The second version of this
> patch introduces a SQL statement parallelised across all CPUs
> available, while the DELETE-statement of the first version literally
> took ages to complete.
> 
> However, cleaning up those data still takes about 26 hours (!) on
> our location02 testing machine, making daily updates of the location
> database impossible to the current knowledge.
> 
> real    1521m30.620s
> user    38m45.521s
> sys     9m6.027s
> 
> Special thanks goes to Michael for spending numerous hours
> on this, setting up a testing environment, doing PostgreSQL magic
> and providing helpful advice while debugging.
> 
> Partially fixes: #12458
> 
> Cc: Michael Tremer <michael.tremer(a)ipfire.org>
> Signed-off-by: Peter Müller <peter.mueller(a)ipfire.org>
> ---
> src/python/location-importer.in | 22 +++++++++++++++++++++-
> 1 file changed, 21 insertions(+), 1 deletion(-)
> 
> diff --git a/src/python/location-importer.in b/src/python/location-importer.in
> index e3a07a0..1467923 100644
> --- a/src/python/location-importer.in
> +++ b/src/python/location-importer.in
> @@ -374,7 +374,27 @@ class CLI(object):
> 				INSERT INTO autnums(number, name)
> 					SELECT _autnums.number, _organizations.name FROM _autnums
> 						JOIN _organizations ON _autnums.organization = _organizations.handle
> -				ON CONFLICT (number) DO UPDATE SET name = excluded.name;
> +				ON CONFLICT (number) DO UPDATE SET name = excluded.name
> +			""")
> +
> +			self.db.execute("""
> +				--- Purge any redundant entries
> +				CREATE TEMPORARY TABLE _garbage ON COMMIT DROP
> +				AS
> +				SELECT network FROM networks candidates
> +				WHERE EXISTS (
> +					SELECT FROM networks
> +					WHERE
> +						networks.network << candidates.network
> +					AND
> +						networks.country = candidates.country
> +				);
> +
> +				CREATE UNIQUE INDEX _garbage_search ON _garbage USING BTREE(network);
> +
> +				DELETE FROM networks WHERE EXISTS (
> +					SELECT FROM _garbage WHERE networks.network = _garbage.network
> +				);
> 			""")
> 
> 		# Download all extended sources
> -- 
> 2.26.2