]>
git.ipfire.org Git - location/libloc.git/log
Michael Tremer [Thu, 6 Mar 2025 18:10:56 +0000 (18:10 +0000)]
log: Perform formatting string sanitation when logging to stderr
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 18:08:46 +0000 (18:08 +0000)]
jenkins: Install perl(Test::More) on Fedora
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 18:00:45 +0000 (18:00 +0000)]
tests: Constify path to open in database test
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 17:59:03 +0000 (17:59 +0000)]
address: Never pass zero to __builtin_ctz
This is undefined behaviour which worked just fine in GCC, but in clang
the code is doing funny business.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 17:45:38 +0000 (17:45 +0000)]
configure: Enable more compiler warnings
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 17:43:17 +0000 (17:43 +0000)]
Fix all sorts of string formatting issues
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 16:53:44 +0000 (16:53 +0000)]
tests: Make bit length mismatch message clearer
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 16:43:29 +0000 (16:43 +0000)]
jenkins: Upload logs from tests in src/, too
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 16:37:09 +0000 (16:37 +0000)]
tree: Try harder to merge networks
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 16:24:46 +0000 (16:24 +0000)]
tree: Add network to the stack after we have tried again
If we add it before, we will try to merge the network with itself which
is not necessary as it will never result in anything good.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 15:48:55 +0000 (15:48 +0000)]
configure: Make Lua check work on Fedora, too
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 15:18:56 +0000 (15:18 +0000)]
jenkins: Install lua-unit wherever available
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 15:13:25 +0000 (15:13 +0000)]
configure: Don't run Lua tests if luaunit is not available
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 15:08:12 +0000 (15:08 +0000)]
debian: Don't build packages against Lua
This is a little bit more complicated, because we probably want to build
against multiple versions and I don't know how to do this.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 15:07:00 +0000 (15:07 +0000)]
jenkins: Actually install Lua when we want to build against it
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 15:03:51 +0000 (15:03 +0000)]
jenkins: Add all supported Debian architectures
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 15:00:58 +0000 (15:00 +0000)]
configure: Fail if Lua was enabled, but not found
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 14:44:42 +0000 (14:44 +0000)]
jenkins: Remove extra tests we currently don't support
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 14:42:15 +0000 (14:42 +0000)]
jenkins: Don't expect any tests to fail
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 14:42:05 +0000 (14:42 +0000)]
jenkins: Always Lua extension, too
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 14:39:01 +0000 (14:39 +0000)]
configure: Check syntax of Lua check
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 14:36:31 +0000 (14:36 +0000)]
configure: Don't automatically detect systemdunitdir
This is no longer possible on newer versions of systemd where the
variable has been removed from pkg-config.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 6 Mar 2025 11:42:50 +0000 (11:42 +0000)]
jenkins: Initial import
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Sat, 28 Sep 2024 12:51:56 +0000 (12:51 +0000)]
data: Import today's database
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Sat, 28 Sep 2024 12:48:20 +0000 (12:48 +0000)]
po: Update POTFILES.in
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Stefan Schantl [Sat, 28 Sep 2024 12:46:31 +0000 (14:46 +0200)]
perl: Return nothing in case invalid data has been passed to libloc
The old croach approch instantly will stop the script if invalid data
will be passed to libloc and also will fail to proper execute the
testsuite.
Signed-off-by: Stefan Schantl <stefan.schantl@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Fri, 10 Jan 2025 13:38:35 +0000 (13:38 +0000)]
importer: Don't import /4 or /10 networks from the routing table
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 6 Jan 2025 11:00:10 +0000 (11:00 +0000)]
importer: Ignore any sudden disconnects when we fetch a Geofeed
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Wed, 2 Oct 2024 18:33:24 +0000 (18:33 +0000)]
importer: Update a few AWS locations
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 26 Sep 2024 20:25:57 +0000 (20:25 +0000)]
perl: Fix a couple of NULL-pointer derefences in the module
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 26 Sep 2024 19:48:03 +0000 (19:48 +0000)]
libloc: Refactor summarizing IP address ranges
The algorithm seems to have had some bugs where it generated too many
unnecessary subnets which caused the system to run out of memory very
quickly when determining bogons.
This version of the algorithm is now a little bit smarter and should be
significantly faster as well.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 26 Sep 2024 16:45:52 +0000 (16:45 +0000)]
tests: lua: Set a variable to true if we are checking for a boolean later
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 26 Sep 2024 14:52:38 +0000 (14:52 +0000)]
lua: Ensure that the testsuite is being executed with the correct version
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 26 Sep 2024 14:37:22 +0000 (14:37 +0000)]
lua: Check if we got returned something on some tests
Since Lua does not know about exceptions, functions usually return nil
instead of some value.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Wed, 19 Jun 2024 09:36:26 +0000 (09:36 +0000)]
importer: Ensure that we set timestamps in the announcements table
Otherwise aging out older announcements does not work.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Wed, 22 May 2024 18:12:04 +0000 (18:12 +0000)]
lua: Fix raising an exception if no network was found
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Wed, 22 May 2024 16:39:57 +0000 (16:39 +0000)]
database: Have the lookup function return 0 even if nothing was found
This is a potentially breaking change, but I think we need to do this as
the previous return code was not very descriptive.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Wed, 22 May 2024 15:35:28 +0000 (15:35 +0000)]
lua: Implement setting a log callback function
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Wed, 22 May 2024 15:30:31 +0000 (15:30 +0000)]
libloc: Allow passing a pointer to the log callback
This is basically re-implemnting the logic to pass log settings as the
previous functions were inappropriately named.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Tue, 21 May 2024 15:46:31 +0000 (15:46 +0000)]
Revert "database: Re-open the file handle in r+ mode"
This reverts commit
7654a5beb9b46758af613f15e01f8e0d0b3bd730 .
This change breaks the library on Linux.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Tue, 21 May 2024 15:34:57 +0000 (15:34 +0000)]
lua: Initialize the database object pointer
Unfortunately Lua does not initialize any new user data. Therefore it
could happen that the cleanup function will try to tidy up after a new
Database object could not be set up correctly and cause a segmentation
fault.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Tue, 21 May 2024 15:11:23 +0000 (15:11 +0000)]
configure: Scan for multiple Lua versions
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Sat, 11 May 2024 11:10:06 +0000 (12:10 +0100)]
database: Re-open the file handle in r+ mode
On Mac OS, we cannot re-read our temporary file from the testsuite unless we re-open in r+ mode.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Sat, 11 May 2024 10:56:26 +0000 (11:56 +0100)]
writer: Move the cursor back to end when finished writing
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Sat, 11 May 2024 10:55:46 +0000 (11:55 +0100)]
writer: Fail if the header could not be written successfully
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Fri, 10 May 2024 17:51:45 +0000 (18:51 +0100)]
Makefile: Don't try to use -export-symbols if --version-script= isn't available
libtool seems to mess up our symbol table by prepending an underscore character
to all lines which renders the file unparsable.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Fri, 10 May 2024 17:23:57 +0000 (18:23 +0100)]
configure: Use pkg-config to find OpenSSL
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 11 Apr 2024 17:45:18 +0000 (17:45 +0000)]
importer: Drop EDROP as it has been merged into DROP
http://www.spamhaus.org/resource-hub/network-security/spamhaus-drop-and-edrop-to-become-a-single-list/
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 8 Apr 2024 10:24:21 +0000 (10:24 +0000)]
lua: Add method to access database creation time
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Sat, 6 Apr 2024 11:11:28 +0000 (11:11 +0000)]
lua: Add function that returns subnets of a network
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Sat, 6 Apr 2024 10:49:41 +0000 (10:49 +0000)]
network: Add function to return a reverse pointer for networks
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Sat, 6 Apr 2024 10:33:33 +0000 (10:33 +0000)]
tests: Set LD_LIBRARY_PATH
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Sat, 6 Apr 2024 10:33:06 +0000 (10:33 +0000)]
address: Add functions to access a specific byte/nibble in an address
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Fri, 5 Apr 2024 15:58:53 +0000 (15:58 +0000)]
lua: Cleanup any database iterators
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Fri, 5 Apr 2024 15:36:14 +0000 (15:36 +0000)]
lua: Create a simple iterator for all networks
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Fri, 5 Apr 2024 15:31:31 +0000 (15:31 +0000)]
tests: Fix bit length tests
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Sun, 31 Mar 2024 16:33:34 +0000 (16:33 +0000)]
lua: Don't raise an error if a network cannot be found
The lookup function now returns nil which is easier to handle than
catching a Lua error.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Sun, 31 Mar 2024 14:16:14 +0000 (14:16 +0000)]
lua: Add compatibility function to compile with Lua >= 5.1
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Fri, 22 Mar 2024 14:45:41 +0000 (14:45 +0000)]
tests: Add tests for #13236
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Fri, 22 Mar 2024 14:42:30 +0000 (14:42 +0000)]
network: Fix handling bit length on merge
The check was kind of useless because of incorrect values from the bit
length function.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Fri, 22 Mar 2024 14:41:44 +0000 (14:41 +0000)]
address: Fix bit length calculation
I have no idea what the previous version computed, but it wasn't the
right thing.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Tue, 19 Mar 2024 10:45:41 +0000 (10:45 +0000)]
importer: Permit Geofeeds for everything instead of ignoring
This seems to become the default and so we should avoid making the
overrides file too verbose.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 18 Mar 2024 17:12:57 +0000 (17:12 +0000)]
importer: Drop even more indexes
We have gone a little bit wild on this but should not need them.
This patch also migrates any GIST indexes to SP-GIST.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 18 Mar 2024 17:05:36 +0000 (17:05 +0000)]
importer: Drop any indexes we no longer need
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 18 Mar 2024 16:56:50 +0000 (16:56 +0000)]
importer: Make the export 200x faster
The export became a huge problem here as it was very slow. One large
query produced one very large output which was difficult to test and
validate as the total export could easily take 17-20 hrs.
This patch fixes this so that the export now takes around ~5min. Sadly
we are still not using more than one processor core, because PostgreSQL
thinks its faster with just the one, but we trade a lot of CPU time for
a little bit of extra memory instead.
We create a new temporary table that holds all networks that we are
interested in. We then run a couple of broken down queries that perform
one thing instead of multiple at the same time.
First, we add all AS numbers, then all countries which are the longest
queries taking now around 20-30s. Then we apply any data from the
external feeds and our own overrides.
The final export requires about 1.7 GiB of memory because we don't
perform any deduplication in the database any more. That has been a lot
better implemented in our own code and I don't think that < 2 GiB is a
reason for concern.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 18 Mar 2024 16:56:08 +0000 (16:56 +0000)]
importer: Replace all GIST indexes with SP-GIST
They are a lot faster for our application.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 18 Mar 2024 16:52:26 +0000 (16:52 +0000)]
importer: ANALYZE all tables before we are running the export
This should help the query planner to make better decisions when
planning the large queries.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Sat, 16 Mar 2024 11:34:21 +0000 (11:34 +0000)]
importer: Remove more traces of the Geofeed overrides table
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Tue, 12 Mar 2024 11:24:47 +0000 (11:24 +0000)]
importer: Convert networks back to string
psycopg3 returns inet types as IP networks.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Tue, 12 Mar 2024 11:17:50 +0000 (11:17 +0000)]
importer: Allow storing multiple Geofeeds for the same network
This is very likely not required except for a couple of overrides.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Tue, 12 Mar 2024 11:14:13 +0000 (11:14 +0000)]
importer: Currently update the source when encountering a conflict
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Tue, 12 Mar 2024 11:12:01 +0000 (11:12 +0000)]
importer: Unify the way we check Geofeed URLs
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Tue, 12 Mar 2024 11:06:22 +0000 (11:06 +0000)]
importer: Import Geofeed overrides with other Geofeeds
The previous approach neglegted that we check if a Geofeed is permitted
to change the location for a subnet so that Geofeeds cannot overwrite
any subnets they don't belong to.
That means that we will have to add those networks to the Geofeeds as
well.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 7 Mar 2024 14:03:50 +0000 (14:03 +0000)]
importer: Remove a debugging line
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 7 Mar 2024 14:03:16 +0000 (14:03 +0000)]
importer: Convert the file handle to text before passing to the CSV parser
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 7 Mar 2024 13:56:36 +0000 (13:56 +0000)]
importer: Fix another variable error in CSV parser
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 7 Mar 2024 13:56:20 +0000 (13:56 +0000)]
importer: Skip ASN lines in extended format
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 7 Mar 2024 13:52:01 +0000 (13:52 +0000)]
importer: Fix another incorrect variable name
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 7 Mar 2024 13:49:58 +0000 (13:49 +0000)]
importer: Fix incorrect variable name
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 7 Mar 2024 12:39:09 +0000 (12:39 +0000)]
importer: Use database pipelining when parsing feeds
Pipelining should allow us to parse feeds faster since we no longer wait
for a response from the database for each row that we are inserting.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 7 Mar 2024 12:24:18 +0000 (12:24 +0000)]
importer: Timeout if fetching a Geofeed takes longer than 5 seconds
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 7 Mar 2024 12:18:14 +0000 (12:18 +0000)]
importer: Wrap everything into asyncio
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Thu, 7 Mar 2024 12:16:30 +0000 (12:16 +0000)]
database: Create a connection pool for async operation
This is very useful when we want to run multiple tasks concurrently
using asyncio.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Wed, 6 Mar 2024 23:47:25 +0000 (23:47 +0000)]
importer: Ignore comments in Geofeeds
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Wed, 6 Mar 2024 23:22:30 +0000 (23:22 +0000)]
importer: Check imported Geofeed override URLs
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Wed, 6 Mar 2024 23:16:44 +0000 (23:16 +0000)]
importer: Add structure to add Geofeed overrides
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Wed, 6 Mar 2024 22:46:44 +0000 (22:46 +0000)]
tree: Don't check if we need to fill the stack
If the stack is empty, the for loop won't run and we will straight go to
where we push the network onto the stack.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Wed, 6 Mar 2024 22:45:28 +0000 (22:45 +0000)]
tree: Be smarter when removing networks from the stack
Since we are working through a sorted tree, we will either only see
subnets of a network or we won't. Once we see a network that isn't a
subnet, we remove the supernet until nothing is left on the stack.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Wed, 6 Mar 2024 22:44:29 +0000 (22:44 +0000)]
tree: Fix memory leak in dedup code
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 4 Mar 2024 12:39:23 +0000 (12:39 +0000)]
Add README.md
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 4 Mar 2024 12:21:23 +0000 (12:21 +0000)]
Update translations
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 4 Mar 2024 12:20:10 +0000 (12:20 +0000)]
importer: Merge the downloader into our main downloader
I don't know why we ended up with duplicated code here, but there seems
to be no reason whatsoever for this.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 4 Mar 2024 12:10:37 +0000 (12:10 +0000)]
importer: Move the split functions into the main importer
This has no functional changes, but moved the functions into a different
file.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 4 Mar 2024 12:07:21 +0000 (12:07 +0000)]
database: Migrate to psycopg3
Fixes: #12953
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 4 Mar 2024 12:06:33 +0000 (12:06 +0000)]
importer: No longer import Geofeeds concurrently
This is kind of a problem since many servers have gone away, respond
very slowly, etc. But since we are moving towards psycopg3, we cannot do
this the old way any more.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 4 Mar 2024 11:47:10 +0000 (11:47 +0000)]
importer: Drop the geofeed sources when updating RIR data
This is a cleaner way to drop any previously imported content and should
also be faster because we can drop a lot of DELETE staments for objects
without Geofeeds.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 4 Mar 2024 11:42:04 +0000 (11:42 +0000)]
importer: Create a better structure to import RIRs
All information about all RIRs will now be imported in one large
database transaction per RIR which should bring us better integrity and
help us to phase out any stale data.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 4 Mar 2024 10:29:38 +0000 (10:29 +0000)]
importer: Add option to only import specific RIRs
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 4 Mar 2024 10:23:49 +0000 (10:23 +0000)]
importer: Also import networks that are smaller than /48 or /24
These cannot appear in the global routing table, but that does not mean
that there isn't any value in importing them into our database.
In case they are just noise, we will filter them out later on.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Michael Tremer [Mon, 4 Mar 2024 10:22:22 +0000 (10:22 +0000)]
importer: Improve checks for unspecified networks
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>