23 Commits
1.0 ... 1.4

Author SHA1 Message Date
Job Snijders
b1074667e8 Add missing closure in bird aspath output 2021-08-20 10:58:07 +00:00
Job Snijders
bdc455d72e Update information on SOURCES 2021-08-20 10:22:19 +00:00
Job Snijders
8cc7ba2751 Bump version 2021-08-20 10:08:59 +00:00
Job Snijders
06b430b420 align bgpq4_print_cisco_aspath and bgpq4_print_cisco_oaspath 2021-08-20 03:51:51 +00:00
Job Snijders
8c51669da9 Clarify what options are mutually exclusive
This fixes #20
2021-08-20 03:06:26 +00:00
Job Snijders
dc1270f10f Improve mikrotik output
This fixes #32
2021-08-20 02:44:37 +00:00
Job Snijders
362f747f0f Let OS decide on buffer size 2021-08-20 02:10:31 +00:00
Job Snijders
6c8c6cfec1 Reduce memory usage 2021-08-20 01:58:38 +00:00
Job Snijders
fb29cd54e1 make local functions static 2021-08-20 01:12:43 +00:00
Job Snijders
903704ef3b Update CHANGES 2021-08-20 01:03:18 +00:00
Job Snijders
b481111cf6 Replace ASN storage facility with Red Black tree macros (RB_) 2021-08-20 00:50:56 +00:00
Job Snijders
80b7dcf4de KNF & renaming 2021-08-19 20:21:25 +00:00
Job Snijders
5ce7d7bfea Untangle function and KNF 2021-08-19 19:59:21 +00:00
Job Snijders
8ec260f03f chop off first and last line 2021-08-19 16:08:37 +00:00
Job Snijders
675155356d Make note of Ben's extra aggregation trick 2021-08-19 16:02:03 +00:00
Job Snijders
da750ed48f KNF 2021-08-19 15:39:19 +00:00
Job Snijders
7fdb481130 bump version 2021-08-19 15:31:25 +00:00
Job Snijders
fa84d146b3 Also put manpage in distribution tarball 2021-08-19 15:30:02 +00:00
Job Snijders
c90c59dba8 Update manpage and readme 2021-08-19 15:20:53 +00:00
Job Snijders
5730525f11 Fix manpage escaping 2021-08-19 15:04:07 +00:00
Job Snijders
78f57867cc Improve manpage 2021-08-19 14:59:58 +00:00
Job Snijders
2391fdb317 Typo 2021-08-19 14:06:01 +00:00
Job Snijders
9f8d82db21 Ensure all queries end with a newline 2021-08-19 13:19:05 +00:00
12 changed files with 1131 additions and 1307 deletions

12
CHANGES
View File

@@ -1,3 +1,15 @@
1.4 (2021-08-20)
- Fix BIRD aspath output
1.3 (2021-08-20)
- Change versioning from X.Y.Z to Y.Z
- The repository file hierachy has been reorganized (compat/ include/)
- Man page has been extended
- Large portions of code have been reformatted to adhere to OpenBSD's source
file style guide to improve readability.
- Refactor: replace two-dimensional array for ASN storage with Red-Black tree
- Reduced memory usage
0.0.9 (2021-08-18)
- Fix various memory errors

29
IDEAS Normal file
View File

@@ -0,0 +1,29 @@
Ben Maddison taught me another aggregation trick:
route-set: AS37271:RS-EXAMPLE
mp-members: 192.0.2.0/27
mp-members: 192.0.2.32/27
mp-members: 192.0.2.64/27
mp-members: 192.0.2.96/27
mp-members: 192.0.2.128/26
mp-members: 192.0.2.128/27
mp-members: 192.0.2.160/27
mp-members: 192.0.2.192/27
mp-members: 192.0.2.224/27
descr: Example route-set
mnt-by: MAINT-AS37271
changed: benm@workonline.africa 20210819
source: RADB
BGPQ4 produces the following:
$ bgpq4 -A AS37271:RS-EXAMPLE
no ip prefix-list NN
ip prefix-list NN permit 192.0.2.0/25 ge 27 le 27
ip prefix-list NN permit 192.0.2.128/26 le 27
ip prefix-list NN permit 192.0.2.192/26 ge 27 le 27
But the following aggregation also is valid, and shorter:
ip prefix-list NN permit 192.0.2.0/24 ge 27 le 27
ip prefix-list NN permit 192.0.2.128/26

520
README.md
View File

@@ -1,240 +1,232 @@
NAME
----
# NAME
`bgpq4` - bgp filtering automation tool
**bgpq4** - bgp filtering automation tool
The `bgpq4` utility queries IRRd and then generates IRR and/or RPKI based
filters formatted for a wide assortment of BGP implementations.
# SYNOPSIS
SYNOPSIS
--------
**bgpq4**
\[**-h** *host\[:port]*]
\[**-S** *sources*]
\[**-EPz**]
\[**-f** *asn* |
**-F** *fmt* |
**-G** *asn*
**-t**]
\[**-46ABbDdJjNnsXU**]
\[**-a** *asn*]
\[**-r** *len*]
\[**-R** *len*]
\[**-m** *max*]
\[**-W** *len*]
*OBJECTS*
\[...]
\[EXCEPT OBJECTS]
```
bgpq4 [-h host[:port]] [-S sources] [-Ez] [-f asn | -F fmt | -G asn | -t] [-46ABbDdeJjNnpsUX] [-a asn] [-r len] [-R len] [-m max] [-W len] OBJECTS [...] EXCEPT OBJECTS
```
# DESCRIPTION
DESCRIPTION
-----------
The bgpq4 utility used to generate configuration (prefix-lists,
extended access-lists, policy-statement terms and as-path lists)
based on IRR routing data.
The
**bgpq4**
utility used to generate configurations (prefix-lists, extended
access-lists, policy-statement terms and as-path lists) based on RADB data.
The options are as follows:
#### -4
**-4**
Generate IPv4 prefix/access-lists (default).
> generate IPv4 prefix/access-lists (default).
#### -6
**-6**
Generate IPv6 prefix/access-lists (IPv4 by default).
> generate IPv6 prefix/access-lists (IPv4 by default).
#### -A
**-A**
Try to aggregate generated filters as much as possible (not all output formats
supported).
> try to aggregate prefix-lists as much as possible (not all output
> formats supported).
#### -a asn
**-a** *asn*
Specify ASN that shall be denied in case of empty prefix-list (OpenBGPD).
> specify what asn shall be denied in case of empty prefix-list (OpenBGPD)
#### -B
**-B**
Generate output in OpenBGPD format.
> generate output in OpenBGPD format (default: Cisco)
#### -b
**-b**
Generate output in BIRD format.
> generate output in BIRD format (default: Cisco).
#### -d
**-d**
Enable some debugging output.
> enable some debugging output.
#### -E
**-e**
Generate extended access-list (Cisco) or policy-statement term using
route-filters (Juniper), [ip|ipv6]-prefix-list (Nokia) or prefix-filter
(OpenBGPD)
> generate output in Arista EOS format (default: Cisco).
#### -e
**-E**
Generate output in Arista EOS format.
> generate extended access-list (Cisco), policy-statement term using
> route-filters (Juniper), \[ip|ipv6]-prefix-list (Nokia) or prefix-sets
> (OpenBGPd).
#### -f `AS number`
**-f** *number*
Generate input as-path access-list for adjacent as `AS number`.
> generate input as-path access-list.
#### -F `fmt`
**-F** *fmt*
Generate output in user-defined format.
> generate output in user-defined format.
#### -G `number`
**-G** *number*
Generate output as-path access-list.
> generate output as-path access-list.
#### -h `host[:port]`
**-h** *host\[:port]*
Host running IRRD database (default: `rr.ntt.net`).
> host running IRRD database (default: rr.ntt.net).
#### -J
**-J**
Generate config for Juniper.
> generate config for Juniper (default: Cisco).
#### -j
**-j**
Generate output in JSON format.
> generate output in JSON format (default: Cisco).
#### -K
**-K**
Generate config for MikroTik.
> generate config for Mikrotik (default: Cisco).
#### -m `length`
**-l** *name*
Maximum length of accepted prefixes (default: `32` for IPv4, `128` for IPv6).
> name of generated entry.
#### -M `match`
**-L** *limit*
Extra match conditions for Juniper route-filters. See the examples section.
> limit recursion depth when expanding as-sets.
#### -n
**-m** *len*
Generate config for Nokia SR OS (former Alcatel-Lucent) MD-CLI.
> maximum prefix-length of accepted prefixes (default: 32 for IPv4 and
> 128 for IPv6).
#### -N
**-M** *match*
Generate config for Nokia SR OS (former Alcatel-Lucent) classic CLI.
> extra match conditions for Juniper route-filters.
#### -l `name`
**-n**
`Name` of generated configuration stanza.
> generate config for Nokia SR OS MD-CLI (Cisco IOS by default)
#### -L `limit`
**-N**
Limit recursion depth when expanding. This slows `bgpq4` a bit, but sometimes
is a useful feature to prevent generated filters from growing too big.
> generate config for Nokia SR OS classic CLI (Cisco IOS by default).
#### -p
**-p**
Enable use of private ASNs and ASNs used for documentation purpose only
(default: disabled).
> accept routes registered for private ASNs (default: disabled)
#### -r `length`
**-P**
Allow more-specific routes with masklen starting with specified length.
> generate prefix-list (default, backward compatibility).
#### -R `length`
**-r** *len*
Allow more-specific routes up to specified masklen too. (Please, note: objects
with prefix-length greater than specified length will be always allowed.)
> allow more specific routes starting with specified masklen too.
#### -s
**-R** *len*
Generate sequence numbers in IOS-style prefix-lists.
> allow more specific routes up to specified masklen too.
#### -S `sources`
**-s**
Use specified sources only (recommended: RADB,RIPE,APNIC).
> generate sequence numbers in IOS-style prefix-lists.
#### -t
**-S** *sources*
Generate as-sets for OpenBGPD (OpenBSD 6.4+), BIRD and JSON formats.
> use specified sources only (recommended: RADB,RIPE,APNIC).
#### -T
**-t**
Disable pipelining (not recommended).
> generate as-sets for OpenBGPd, BIRD and JSON formats.
#### -U
**-T**
Generate output in Huawei format.
> disable pipelining (not recommended).
#### -W `length`
**-W** *len*
Generate as-path strings of a given length maximum (0 for infinity).
> generate as-path strings of no more than len items (use 0 for inifinity).
#### -X
**-U**
Generate output in Cisco IOS XR format.
> generate config for Huawei devices (Cisco IOS by default)
#### -z
**-X**
Generate Juniper route-filter-list (JunOS 16.2+).
> generate config for Cisco IOS XR devices (plain IOS by default).
#### `OBJECTS`
**-z**
`OBJECTS` means networks (in prefix format), autonomous systems, as-sets and
route-sets. If multiple objects are specified they will be merged.
> generate route-filter-lists (JunOS 16.2+).
#### `EXCEPT OBJECTS`
*OBJECTS*
You can exclude autonomous sets, as-sets and route-sets found during
expansion from future expansion.
> means networks (in prefix format), autonomous systems, as-sets and route-sets.
EXAMPLES
--------
Generating prefix filter for MikroTik for `AS20597`:
*EXCEPT OBJECTS*
$ ./bgpq4 -Kl eltel-v4 AS20597
/routing filter add action=accept chain="eltel-v4" prefix=81.9.0.0/20
/routing filter add action=accept chain="eltel-v4" prefix=81.9.32.0/20
/routing filter add action=accept chain="eltel-v4" prefix=81.9.96.0/20
/routing filter add action=accept chain="eltel-v4" prefix=81.222.128.0/20
/routing filter add action=accept chain="eltel-v4" prefix=81.222.160.0/20
/routing filter add action=accept chain="eltel-v4" prefix=81.222.192.0/18
/routing filter add action=accept chain="eltel-v4" prefix=85.249.8.0/21
/routing filter add action=accept chain="eltel-v4" prefix=85.249.224.0/19
/routing filter add action=accept chain="eltel-v4" prefix=89.112.0.0/17
/routing filter add action=accept chain="eltel-v4" prefix=217.170.64.0/19
> those objects will be excluded from expansion.
Generating named Juniper prefix-filter for `AS20597`:
# EXAMPLES
$ bgpq4 -Jl eltel-v4 AS20597
policy-options {
replace:
prefix-list eltel-v4 {
81.9.0.0/20;
81.9.32.0/20;
81.9.96.0/20;
81.222.128.0/20;
81.222.192.0/18;
85.249.8.0/21;
85.249.224.0/19;
89.112.0.0/19;
89.112.4.0/22;
89.112.32.0/19;
89.112.64.0/19;
217.170.64.0/20;
217.170.80.0/20;
}
}
Generating named juniper prefix-filter for AS20597:
$ bgpq4 -Jl eltel AS20597
policy-options {
replace:
prefix-list eltel {
81.9.0.0/20;
81.9.32.0/20;
81.9.96.0/20;
81.222.128.0/20;
81.222.192.0/18;
85.249.8.0/21;
85.249.224.0/19;
89.112.0.0/19;
89.112.4.0/22;
89.112.32.0/19;
89.112.64.0/19;
217.170.64.0/20;
217.170.80.0/20;
}
}
For Cisco we can use aggregation (-A) flag to make this prefix-filter
more compact:
$ bgpq4 -Al eltel-v4 AS20597
no ip prefix-list eltel-v4
ip prefix-list eltel-v4 permit 81.9.0.0/20
ip prefix-list eltel-v4 permit 81.9.32.0/20
ip prefix-list eltel-v4 permit 81.9.96.0/20
ip prefix-list eltel-v4 permit 81.222.128.0/20
ip prefix-list eltel-v4 permit 81.222.192.0/18
ip prefix-list eltel-v4 permit 85.249.8.0/21
ip prefix-list eltel-v4 permit 85.249.224.0/19
ip prefix-list eltel-v4 permit 89.112.0.0/18 ge 19 le 19
ip prefix-list eltel-v4 permit 89.112.4.0/22
ip prefix-list eltel-v4 permit 89.112.64.0/19
ip prefix-list eltel-v4 permit 217.170.64.0/19 ge 20 le 20
$ bgpq4 -Al eltel AS20597
no ip prefix-list eltel
ip prefix-list eltel permit 81.9.0.0/20
ip prefix-list eltel permit 81.9.32.0/20
ip prefix-list eltel permit 81.9.96.0/20
ip prefix-list eltel permit 81.222.128.0/20
ip prefix-list eltel permit 81.222.192.0/18
ip prefix-list eltel permit 85.249.8.0/21
ip prefix-list eltel permit 85.249.224.0/19
ip prefix-list eltel permit 89.112.0.0/18 ge 19 le 19
ip prefix-list eltel permit 89.112.4.0/22
ip prefix-list eltel permit 89.112.64.0/19
ip prefix-list eltel permit 217.170.64.0/19 ge 20 le 20
and, as you see, prefixes `89.112.0.0/19` and `89.112.32.0/19` now aggregated
into single entry
Prefixes 89.112.0.0/19 and 89.112.32.0/19 now aggregated
into single entry 89.112.0.0/18 ge 19 le 19.
ip prefix-list eltel permit 89.112.0.0/18 ge 19 le 19.
Well, for Juniper we can generate even more interesting policy-options,
using -M <extra match conditions>, -R <len> and hierarchical names:
Well, for Juniper we can generate even more interesting policy-statement,
using `-M <extra match conditions>`, `-r <len>`, `-R <len>` and hierarchical
names:
$ bgpq4 -AJEl eltel/specifics -r 29 -R 32 -M "community blackhole" AS20597
$ bgpq4 -AJEl eltel/specifics -r 29 -R 32 -M "community blackhole" AS20597
policy-options {
policy-statement eltel {
term specifics {
@@ -255,51 +247,83 @@ names:
}
}
generated policy-option term now allows all specifics with prefix-length
between /29 and /32 for eltel networks if they match with special community
blackhole (defined elsewhere in configuration).
generated policy-option term now allows more-specific routes in range
/29 - /32 for eltel networks if they marked with community 'blackhole'
(defined elsewhere in configuration).
Of course, this version supports IPv6 (-6):
Of course, `bgpq4` supports IPv6 (-6):
$ bgpq4 -6l as-retn-v6 AS-RETN6
no ipv6 prefix-list as-retn-v6
ipv6 prefix-list as-retn-v6 permit 2001:7fb:fe00::/48
ipv6 prefix-list as-retn-v6 permit 2001:7fb:fe01::/48
[....]
$ bgpq4 -6l as-retn-6 AS-RETN6
no ipv6 prefix-list as-retn-6
ipv6 prefix-list as-retn-6 permit 2001:7fb:fe00::/48
ipv6 prefix-list as-retn-6 permit 2001:7fb:fe01::/48
[....]
and assumes your device supports 32-bit ASNs
$ bgpq4 -Jf 112 AS-SPACENET
policy-options {
replace:
as-path-group NN {
as-path a0 "^112(112)*$";
as-path a1 "^112(.)*(1898|5539|8495|8763|8878|12136|12931|15909)$";
as-path a2 "^112(.)*(21358|23600|24151|25152|31529|34127|34906)$";
as-path a3 "^112(.)*(35052|41720|43628|44450|196611)$";
}
}
$ bgpq4 -Jf 112 AS-SPACENET
policy-options {
replace:
as-path-group NN {
as-path a0 "^112(112)*$";
as-path a1 "^112(.)*(1898|5539|8495|8763|8878|12136|12931|15909)$";
as-path a2 "^112(.)*(21358|23456|23600|24151|25152|31529|34127|34906)$";
as-path a3 "^112(.)*(35052|41720|43628|44450|196611)$";
}
}
see `AS196611` in the end of the list ? That's a 32-bit ASN.
see \`AS196611\` in the end of the list ? That's a 32-bit ASN.
USER-DEFINED FORMAT
-------------------
# USER-DEFINED FORMAT
If you want to generate configuration not for routers, but for some
other programs/systems, you may use user-defined formatting, like in
example below:
$ bgpq4 -F "ipfw add pass all from %n/%l to any\\n" as3254
$ bgpq4 -F "ipfw add pass all from %n/%l to any\n" as3254
ipfw add pass all from 62.244.0.0/18 to any
ipfw add pass all from 91.219.29.0/24 to any
ipfw add pass all from 91.219.30.0/24 to any
ipfw add pass all from 193.193.192.0/19 to any
Recognized format characters: '%n' - network, '%l' - mask length,
'%a' - aggregate low mask length, '%A' - aggregate high mask length,
'%N' - object name, '%m' - object mask and '%i' - inversed mask.
Recognized escape characters: '\n' - new line, '\t' - tabulation.
Recognized format sequences are:
**%n**
> network
**%l**
> mask length
**%a**
> aggregate low mask length
**%A**
> aggregate high mask length
**%N**
> object name
**%m**
> object mask
**%i**
> inversed mask
**&#92;n**
> new line
**&#92;t**
> tabulation
Please note that no new lines inserted automatically after each sentence,
you have to add them into format string manually, elsewhere output will
be in one line (sometimes it makes sense):
@@ -307,102 +331,116 @@ be in one line (sometimes it makes sense):
$ bgpq4 -6F "%n/%l; " as-eltel
2001:1b00::/32; 2620:4f:8000::/48; 2a04:bac0::/29; 2a05:3a80::/48;
DIAGNOSTICS
-----------
# NOTES ON SOURCES
When everything is OK, `bgpq4` generates result to standard output and
exits with status == 0. In case of errors they are printed to stderr and
program exits with non-zero status.
By default
*bgpq4*
trusts to data from all databases mirrored into NTT's IRR service.
Unfortunately, not all these databases are equal in how much can we
trust their data.
RIR maintained databases (AFRINIC, ARIN, APNIC, LACNIC and RIPE)
shall be trusted more than the others because they are indeed have the
knowledge about which address space allocated to this or that ASn,
other databases lack this knowledge and can (and, actually, do) contain
some stale data: noone but RIRs care to remove outdated route-objects
when address space revoked from one ASn and allocated to another.
In order to keep their filters both compact and actual,
*bgpq4 users*
are encouraged to use '-S' flag to limit database sources to only
ones they trust.
NOTES ON ULTRA-LARGE PREFIX-LISTS
---------------------------------
General recommendations:
To improve `bgpq4` performance when expanding extra-large AS-SETs you
Use minimal set of RIR databases (only those in which you and your
customers have registered route-objects).
Avoid using ARIN-NONAUTH and RIPE-NONAUTH as trusted source: these records
were created in database but for address space allocated to different RIR,
so the NONAUTH databases have no chance to confirm validity of this route
object.
$ bgpq4 -S RIPE,RADB as-space
no ip prefix-list NN
ip prefix-list NN permit 195.190.32.0/19
$ bgpq4 -S RADB,RIPE as-space
no ip prefix-list NN
ip prefix-list NN permit 45.4.4.0/22
ip prefix-list NN permit 45.4.132.0/22
ip prefix-list NN permit 45.6.128.0/22
ip prefix-list NN permit 45.65.184.0/22
[...]
# PERFORMANCE
To improve \`bgpq4\` performance when expanding extra-large AS-SETs you
shall tune OS settings to enlarge TCP send buffer.
FreeBSD can be tuned in the following way:
sysctl -w net.inet.tcp.sendbuf_max=2097152
sysctl -w net.inet.tcp.sendbuf_max=2097152
Linux can be tuned in the following way:
sysctl -w net.ipv4.tcp_window_scaling=1
sysctl -w net.core.rmem_max=2097152
sysctl -w net.core.wmem_max=2097152
sysctl -w net.ipv4.tcp_rmem="4096 87380 2097152"
sysctl -w net.ipv4.tcp_wmem="4096 65536 2097152"
sysctl -w net.ipv4.tcp_window_scaling=1
BUILDING
--------
sysctl -w net.core.rmem_max=2097152
sysctl -w net.core.wmem_max=2097152
sysctl -w net.ipv4.tcp_rmem="4096 87380 2097152"
sysctl -w net.ipv4.tcp_wmem="4096 65536 2097152"
# BUILDING
This project uses autotools. If you are building from the repository,
run the following command to prepare the build system:
./bootstrap
./bootstrap
In order to compile the software, run:
./configure
make
make install
./configure
make
make install
If you wish to remove the generated build system files from your
working tree, run:
make maintainer-clean
make maintainer-clean
In order to create a distribution archive, run:
make dist
make dist
PACKAGE INSTALLATION
--------------------
# DIAGNOSTICS
In FreeBSD binary package can be installed using
When everything is OK,
**bgpq4**
generates access-list to standard output and exits with status == 0.
In case of errors they are printed to stderr and program exits with
non-zero status.
```shell
pkg install bgpq4
```
Or from ports with `portmaster`
```shell
portmaster net-mgmt/bgpq4
```
On Arch Linux, BGPQ4 is [available in AUR](https://aur.archlinux.org/packages/bgpq4/):
```shell
yay -S bgpq4
```
On OpenBSD:
```shell
pkg_add bgpq4
```
MAILING LIST
------------
Users and interested parties can subscribe to the BGPQ4 mailing list [bgpq4@tcp0.com](https://tcp0.com/cgi-bin/mailman/listinfo/bgpq4).
SEE ALSO
--------
NLNOG's [BGP Filter Guide](http://bgpfilterguide.nlnog.net/)
AUTHORS
-------
# AUTHORS
Alexandre Snarskii, Christian David, Claudio Jeker, Job Snijders,
Massimiliano Stucchi, Michail Litvak, Peter Schoenmaker, Roelf Wichertjes,
and contributions from many others.
Project
-------
# SEE ALSO
BGPQ4 is maintained by Job Snijders `<job@sobornost.net>`.
**https://github.com/bgp/bgpq4**
BGPQ4 on Github.
[https://github.com/bgp/bgpq4](https://github.com/bgp/bgpq4)
**http://bgpfilterguide.nlnog.net/**
NLNOG's BGP Filter Guide.
**https://tcp0.com/cgi-bin/mailman/listinfo/bgpq4**
Users and interested parties can subscribe to the BGPQ4 mailing list bgpq4@tcp0.com
# PROJECT MAINTAINER
Job Snijders &lt;job@sobornost.net&gt;

View File

@@ -1 +1 @@
1.0
1.4

113
bgpq4.8
View File

@@ -178,7 +178,8 @@ ip prefix-list eltel permit 89.112.64.0/19
ip prefix-list eltel permit 217.170.64.0/19 ge 20 le 20
.Ed
.fi
- you see, prefixes 89.112.0.0/19 and 89.112.32.0/19 now aggregated
.Pp
Prefixes 89.112.0.0/19 and 89.112.32.0/19 now aggregated
into single entry 89.112.0.0/18 ge 19 le 19.
.Pp
Well, for Juniper we can generate even more interesting policy-options,
@@ -251,10 +252,29 @@ ipfw add pass all from 193.193.192.0/19 to any
.Ed
.fi
.Pp
Recognized format characters: %n - network, %l - mask length,
%a - aggregate low mask length, %A - aggregate high mask length,
%N - object name, %m - object mask and %i - inversed mask.
Recognized escape characters: \\n - new line, \\t - tabulation.
Recognized format sequences are:
.Pp
.Bl -tag -width Ds -offset indent -compact
.It Cm %n
network
.It Cm %l
mask length
.It Cm %a
aggregate low mask length
.It Cm \&%A
aggregate high mask length
.It Cm \&%N
object name
.It Cm %m
object mask
.It Cm %i
inversed mask
.It Cm \en
new line
.It Cm \et
tabulation
.El
.Pp
Please note that no new lines inserted automatically after each sentence,
you have to add them into format string manually, elsewhere output will
be in one line (sometimes it makes sense):
@@ -264,14 +284,97 @@ $ bgpq4 -6F "%n/%l; " as-eltel
2001:1b00::/32; 2620:4f:8000::/48; 2a04:bac0::/29; 2a05:3a80::/48;
.Ed
.fi
.Sh NOTES ON SOURCES
By default
.Em bgpq4
trusts to data from all databases mirrored into NTT's IRR service.
Unfortunately, not all these databases are equal in how much can we
trust their data.
RIR maintained databases (AFRINIC, ARIN, APNIC, LACNIC and RIPE)
shall be trusted more than the others because they are indeed have the
knowledge about which address space allocated to this or that ASn,
other databases lack this knowledge and can (and, actually, do) contain
some stale data: noone but RIRs care to remove outdated route-objects
when address space revoked from one ASn and allocated to another.
In order to keep their filters both compact and actual,
.Em bgpq4 users
are encouraged to use '-S' flag to limit database sources to only
ones they trust.
.Pp
General recommendations:
.Pp
Use minimal set of RIR databases (only those in which you and your
customers have registered route-objects).
.Pp
Avoid using ARIN-NONAUTH and RIPE-NONAUTH as trusted source: these records
were created in database but for address space allocated to different RIR,
so the NONAUTH databases have no chance to confirm validity of this route
object.
.Bd -literal
$ bgpq4 -S RIPE,RADB as-space
no ip prefix-list NN
ip prefix-list NN permit 195.190.32.0/19
$ bgpq4 -S RADB,RIPE as-space
no ip prefix-list NN
ip prefix-list NN permit 45.4.4.0/22
ip prefix-list NN permit 45.4.132.0/22
ip prefix-list NN permit 45.6.128.0/22
ip prefix-list NN permit 45.65.184.0/22
[...]
.Sh PERFORMANCE
To improve `bgpq4` performance when expanding extra-large AS-SETs you
shall tune OS settings to enlarge TCP send buffer.
.Pp
FreeBSD can be tuned in the following way:
.Pp
.Dl sysctl -w net.inet.tcp.sendbuf_max=2097152
.Pp
Linux can be tuned in the following way:
.Pp
.Dl sysctl -w net.ipv4.tcp_window_scaling=1
.Dl sysctl -w net.core.rmem_max=2097152
.Dl sysctl -w net.core.wmem_max=2097152
.Dl sysctl -w net.ipv4.tcp_rmem="4096 87380 2097152"
.Dl sysctl -w net.ipv4.tcp_wmem="4096 65536 2097152"
.Sh BUILDING
This project uses autotools. If you are building from the repository,
run the following command to prepare the build system:
.Pp
.Dl ./bootstrap
.Pp
In order to compile the software, run:
.Pp
.Dl ./configure
.Dl make
.Dl make install
.Pp
If you wish to remove the generated build system files from your
working tree, run:
.Pp
.Dl make maintainer-clean
.Pp
In order to create a distribution archive, run:
.Pp
.Dl make dist
.Sh DIAGNOSTICS
When everything is OK,
.Nm
generates access-list to standard output and exits with status == 0.
In case of errors they are printed to stderr and program exits with
non-zero status.
.Sh AUTHORS
Alexandre Snarskii, Christian David, Claudio Jeker, Job Snijders,
Massimiliano Stucchi, Michail Litvak, Peter Schoenmaker, Roelf Wichertjes,
and contributions from many others.
.Sh SEE ALSO
.Sy https://github.com/bgp/bgpq4
BGPQ4 on Github.
.Pp
.Sy http://bgpfilterguide.nlnog.net/
NLNOG's BGP Filter Guide.
.Pp
.Sy https://tcp0.com/cgi-bin/mailman/listinfo/bgpq4
Users and interested parties can subscribe to the BGPQ4 mailing list bgpq4@tcp0.com
.Sh PROJECT MAINTAINER
.An Job Snijders Aq job@sobornost.net

View File

@@ -58,7 +58,15 @@ tentry_cmp(struct sx_tentry *a, struct sx_tentry *b)
return strcasecmp(a->text, b->text);
}
RB_GENERATE_STATIC(tentree, sx_tentry, entries, tentry_cmp);
RB_GENERATE_STATIC(tentree, sx_tentry, entry, tentry_cmp);
int
asn_cmp(struct asn_entry *a, struct asn_entry *b)
{
return (a->asn < b->asn ? -1 : a->asn > b->asn);
}
RB_GENERATE(asn_tree, asn_entry, entry, asn_cmp);
int
bgpq_expander_init(struct bgpq_expander *b, int af)
@@ -80,22 +88,12 @@ bgpq_expander_init(struct bgpq_expander *b, int af)
b->sources = "";
b->name = "NN";
b->aswidth = 8;
b->asn32s[0] = malloc(8192);
if (!b->asn32s[0]) {
sx_report(SX_FATAL,"Unable to allocate 8192 bytes: %s\n",
strerror(errno));
exit(1);
}
memset(b->asn32s[0], 0, 8192);
b->identify = 1;
b->server = "rr.ntt.net";
b->port = "43";
// b->wq = STAILQ_HEAD_INITIALZIER(b->wq);
// b->rq = STAILQ_HEAD_INITIALZIER(b->rq);
// b->rsets = STAILQ_HEAD_INITIALZIER(b->rsets);
// b->macroses = STAILQ_HEAD_INITIALZIER(b->macroses);
RB_INIT(&b->asnlist);
STAILQ_INIT(&b->wq);
STAILQ_INIT(&b->rq);
STAILQ_INIT(&b->rsets);
@@ -116,14 +114,14 @@ fixups:
int
bgpq_expander_add_asset(struct bgpq_expander *b, char *as)
{
struct sx_slentry *le;
struct slentry *le;
if (!b || !as)
return 0;
le = sx_slentry_new(as);
STAILQ_INSERT_TAIL(&b->macroses, le, entries);
STAILQ_INSERT_TAIL(&b->macroses, le, entry);
return 1;
}
@@ -131,7 +129,7 @@ bgpq_expander_add_asset(struct bgpq_expander *b, char *as)
int
bgpq_expander_add_rset(struct bgpq_expander *b, char *rs)
{
struct sx_slentry *le;
struct slentry *le;
if (!b || !rs)
return 0;
@@ -141,7 +139,7 @@ bgpq_expander_add_rset(struct bgpq_expander *b, char *rs)
if (!le)
return 0;
STAILQ_INSERT_TAIL(&b->rsets, le, entries);
STAILQ_INSERT_TAIL(&b->rsets, le, entry);
return 1;
}
@@ -183,46 +181,29 @@ bgpq_expander_add_stop(struct bgpq_expander *b, char *rs)
int
bgpq_expander_add_as(struct bgpq_expander *b, char *as)
{
char *eoa;
uint32_t asn1 = 0, asn2 = 0;
uint32_t asno = 0;
char *eoa;
uint32_t asno = 0;
struct asn_entry *asne;
if (!b || !as)
return 0;
asno = strtoul(as + 2, &eoa, 10);
if (eoa && *eoa != 0) {
sx_report(SX_ERROR,"Invalid symbol in AS number: '%c' in %s\n",
*eoa, as);
return 0;
}
if (asno > 65535) {
asn1 = asno / 65536;
asn2 = asno % 65536;
} else
asn1 = asno;
if (!expand_special_asn &&
((asno >= 4200000000ul) || (asno >= 64496 && asno <= 65551)))
return 0;
if (!b->asn32s[asn1]) {
b->asn32s[asn1] = malloc(8192);
if (!b->asn32s[asn1]) {
sx_report(SX_FATAL, "Unable to allocate 8192 "
"bytes: %s. Unable to add asn32 %s to "
" future expansion\n", strerror(errno), as);
return 0;
}
memset(b->asn32s[asn1], 0, 8192);
}
if ((asne = malloc(sizeof(struct asn_entry))) == NULL)
err(1, NULL);
if (asno > 65535)
b->asn32s[asn1][asn2 / 8] |= (0x80 >> (asn2 % 8));
else
b->asn32s[0][asn1 / 8] |= (0x80 >> (asn1 % 8));
asne->asn = asno;
RB_INSERT(asn_tree, &b->asnlist, asne);
return 1;
}
@@ -261,24 +242,24 @@ bgpq_expander_add_prefix_range(struct bgpq_expander *b, char *prefix)
static int
bgpq_expanded_macro(char *as, struct bgpq_expander *ex,
struct bgpq_request *req)
struct request *req)
{
bgpq_expander_add_as(ex, as);
return 1;
}
struct bgpq_request *bgpq_pipeline(struct bgpq_expander *b,
int (*callback)(char *, struct bgpq_expander *b, struct bgpq_request *req),
struct request *bgpq_pipeline(struct bgpq_expander *b,
int (*callback)(char *, struct bgpq_expander *b, struct request *req),
void *udata, char *fmt, ...);
int bgpq_expand_irrd(struct bgpq_expander *b,
int (*callback)(char*, struct bgpq_expander *b, struct bgpq_request *req),
int (*callback)(char*, struct bgpq_expander *b, struct request *req),
void *udata, char *fmt, ...);
static int
bgpq_expanded_macro_limit(char *as, struct bgpq_expander *b,
struct bgpq_request *req)
struct request *req)
{
if (!strncasecmp(as, "AS-", 3) || strchr(as, '-') || strchr(as, ':')) {
struct sx_tentry tkey = { .text = as };
@@ -300,8 +281,8 @@ bgpq_expanded_macro_limit(char *as, struct bgpq_expander *b,
req->depth + 1 < b->maxdepth)) {
bgpq_expander_add_already(b, as);
if (pipelining) {
struct bgpq_request *req1 = bgpq_pipeline(b,
bgpq_expanded_macro_limit, NULL, "!i%s",
struct request *req1 = bgpq_pipeline(b,
bgpq_expanded_macro_limit, NULL, "!i%s\n",
as);
req1->depth = req->depth + 1;
} else {
@@ -343,7 +324,7 @@ bgpq_expanded_macro_limit(char *as, struct bgpq_expander *b,
static int
bgpq_expanded_prefix(char *as, struct bgpq_expander *ex,
struct bgpq_request *req __attribute__((unused)))
struct request *req __attribute__((unused)))
{
char *d = strchr(as, '^');
@@ -357,7 +338,7 @@ bgpq_expanded_prefix(char *as, struct bgpq_expander *ex,
static int
bgpq_expanded_v6prefix(char *prefix, struct bgpq_expander *ex,
struct bgpq_request *req)
struct request *req)
{
char *d = strchr(prefix, '^');
@@ -371,16 +352,16 @@ bgpq_expanded_v6prefix(char *prefix, struct bgpq_expander *ex,
int bgpq_pipeline_dequeue(int fd, struct bgpq_expander *b);
static struct bgpq_request *
bgpq_request_alloc(char *request, int (*callback)(char *, struct bgpq_expander *,
struct bgpq_request *), void *udata)
static struct request *
request_alloc(char *request, int (*callback)(char *, struct bgpq_expander *,
struct request *), void *udata)
{
struct bgpq_request *bp = malloc(sizeof(struct bgpq_request));
struct request *bp = malloc(sizeof(struct request));
if (!bp)
return NULL;
memset(bp, 0, sizeof(struct bgpq_request));
memset(bp, 0, sizeof(struct request));
bp->request = strdup(request);
bp->offset = 0;
bp->size = strlen(bp->request);
@@ -391,7 +372,7 @@ bgpq_request_alloc(char *request, int (*callback)(char *, struct bgpq_expander *
}
static void
bgpq_request_free(struct bgpq_request *req)
request_free(struct request *req)
{
if (req->request)
free(req->request);
@@ -399,14 +380,14 @@ bgpq_request_free(struct bgpq_request *req)
free(req);
}
struct bgpq_request *
struct request *
bgpq_pipeline(struct bgpq_expander *b,
int (*callback)(char *, struct bgpq_expander *, struct bgpq_request *),
int (*callback)(char *, struct bgpq_expander *, struct request *),
void *udata, char *fmt, ...)
{
char request[128];
int ret;
struct bgpq_request *bp = NULL;
struct request *bp = NULL;
va_list ap;
va_start(ap, fmt);
@@ -415,11 +396,11 @@ bgpq_pipeline(struct bgpq_expander *b,
SX_DEBUG(debug_expander,"expander: sending %s", request);
bp = bgpq_request_alloc(request, callback, udata);
bp = request_alloc(request, callback, udata);
if (!bp) {
sx_report(SX_FATAL,"Unable to allocate %lu bytes: %s\n",
(unsigned long)sizeof(struct bgpq_request),
(unsigned long)sizeof(struct request),
strerror(errno));
exit(1);
}
@@ -449,23 +430,35 @@ bgpq_pipeline(struct bgpq_expander *b,
static void
bgpq_expander_invalidate_asn(struct bgpq_expander *b, const char *q)
{
char *eptr;
unsigned long asn = 0;
struct asn_entry find, *res;
if (!strncmp(q, "!gas", 4) || !strncmp(q, "!6as", 4)) {
char *eptr;
unsigned long asn = strtoul(q+4, &eptr, 10), asn0, asn1 = 0;
if (!asn || asn == ULONG_MAX || asn >= 4294967295
|| (eptr && *eptr != '\n')) {
sx_report(SX_ERROR, "some problem invalidating asn"
" %s\n", q);
errno = 0;
asn = strtoul(q + 4, &eptr, 10);
if (*eptr != '\n') {
sx_report(SX_ERROR, "not a number: %s\n", q);
return;
}
asn1 = asn % 65536;
asn0 = asn / 65536;
if (!b->asn32s[asn0] ||
!(b->asn32s[asn0][asn1/8] & (0x80 >> (asn1 % 8)))) {
sx_report(SX_NOTICE, "strange, invalidating inactive "
"asn %lu(%s)\n", asn, q);
} else {
b->asn32s[asn0][asn1 / 8] &= ~(0x80 >> (asn1 % 8));
if (errno == ERANGE && asn == ULONG_MAX) {
sx_report(SX_ERROR, "overflow: %s\n", q);
return;
}
if (asn == 0 || asn >= 4294967295) {
sx_report(SX_ERROR, "out of range: %s\n", q);
return;
}
if (eptr && *eptr != '\n') {
sx_report(SX_ERROR, "no number? %s\n", q);
return;
}
find.asn = asn;
if ((res = RB_FIND(asn_tree, &b->asnlist, &find)) == NULL) {
RB_REMOVE(asn_tree, &b->asnlist, res);
free(res);
}
}
}
@@ -474,7 +467,7 @@ static void
bgpq_write(struct bgpq_expander *b)
{
while(!STAILQ_EMPTY(&b->wq)) {
struct bgpq_request *req = STAILQ_FIRST(&b->wq);
struct request *req = STAILQ_FIRST(&b->wq);
int ret = write(b->fd, req->request + req->offset,
req->size-req->offset);
@@ -500,8 +493,8 @@ bgpq_write(struct bgpq_expander *b)
static int
bgpq_selread(struct bgpq_expander *b, char *buffer, int size)
{
fd_set rfd, wfd;
int ret;
fd_set rfd, wfd;
int ret;
repeat:
FD_ZERO(&rfd);
@@ -542,7 +535,7 @@ bgpq_read(struct bgpq_expander *b)
while(!STAILQ_EMPTY(&b->rq)) {
int ret = 0;
char *cres;
struct bgpq_request *req = STAILQ_FIRST(&b->rq);
struct request *req = STAILQ_FIRST(&b->rq);
SX_DEBUG(debug_expander > 2, "waiting for answer to %s,"
"init %i '%.*s'\n", req->request, off, off, response);
@@ -580,7 +573,7 @@ have:
"bytes: %s\n", togot + 2, strerror(errno));
}
memset(recvbuffer,0,togot+2);
memset(recvbuffer, 0, togot + 2);
if (!eon || *eon != '\n') {
sx_report(SX_ERROR,"A-code finished with wrong"
@@ -602,7 +595,7 @@ have:
/* response is not yet fully buffered */
memcpy(recvbuffer, eon + 1,
off - ((eon + 1) - response));
offset = off - ((eon+1) - response);
offset = off - ((eon + 1) - response);
memset(response, 0, sizeof(response));
off = 0;
}
@@ -674,7 +667,7 @@ have3:
if (c[0] == 0)
break;
req->callback(c, b, req);
c += spn+1;
c += spn + 1;
}
assert(c == recvbuffer + togot);
memset(recvbuffer, 0, togot + 2);
@@ -704,7 +697,7 @@ have3:
}
memmove(response, cres + 1, off - ((cres + 1) - response));
off -= (cres+1) - response;
off -= (cres + 1) - response;
memset(response + off, 0, sizeof(response) - off);
SX_DEBUG(debug_expander > 5,
"fixed response of %i, %.*s\n", off, off, response);
@@ -712,7 +705,7 @@ have3:
STAILQ_REMOVE_HEAD(&b->rq, next);
b->piped--;
bgpq_request_free(req);
request_free(req);
}
return 0;
@@ -720,30 +713,26 @@ have3:
int
bgpq_expand_irrd(struct bgpq_expander *b,
int (*callback)(char *, struct bgpq_expander *, struct bgpq_request *),
int (*callback)(char *, struct bgpq_expander *, struct request *),
void *udata, char *fmt, ...)
{
char request[128], response[128];
va_list ap;
ssize_t ret;
int off = 0;
struct bgpq_request *req;
struct request *req;
va_start(ap, fmt);
vsnprintf(request, sizeof(request), fmt, ap);
va_end(ap);
req = bgpq_request_alloc(request, callback, udata);
req = request_alloc(request, callback, udata);
SX_DEBUG(debug_expander, "expander: sending '%s'\n", request);
SX_DEBUG(debug_expander, "expander sending: %s", request);
if ((ret = write(b->fd, request, strlen(request)) == 0) || ret == -1)
err(1, "write");
// push the button by sending a newline
if ((ret = write(b->fd, "\n", 1) == 0) || ret == -1)
err(1, "write");
memset(response, 0, sizeof(response));
repeat:
@@ -767,7 +756,7 @@ repeat:
if (response[0] == 'A') {
char *eon, *c;
long togot = strtoul(response+1, &eon, 10);
long togot = strtoul(response + 1, &eon, 10);
char *recvbuffer = malloc(togot + 2);
int offset = 0;
@@ -777,13 +766,13 @@ repeat:
}
if (eon && *eon != '\n') {
sx_report(SX_ERROR,"A-code finised with wrong char "
sx_report(SX_ERROR,"A-code finished with wrong char "
"'%c' (%s)\n", *eon, response);
exit(1);
}
if (off - ((eon + 1)-response) > togot) {
memcpy(recvbuffer, eon+1, togot);
if (off - ((eon + 1) - response) > togot) {
memcpy(recvbuffer, eon + 1, togot);
offset = togot;
memmove(response, eon + 1 + togot,
off - ((eon + 1) - response) - togot);
@@ -863,7 +852,7 @@ have3:
sx_report(SX_ERROR,"Wrong reply: %s\n", response);
exit(0);
}
bgpq_request_free(req);
request_free(req);
return 0;
}
@@ -872,9 +861,10 @@ int
bgpq_expand(struct bgpq_expander *b)
{
int fd = -1, err, ret, aquery = 0;
struct sx_slentry *mc;
struct slentry *mc;
struct addrinfo hints, *res = NULL, *rp;
struct linger sl;
struct asn_entry *asne;
sl.l_onoff = 1;
sl.l_linger = 5;
@@ -1027,24 +1017,24 @@ bgpq_expand(struct bgpq_expander *b)
if (pipelining)
fcntl(fd, F_SETFL, O_NONBLOCK|(fcntl(fd, F_GETFL)));
STAILQ_FOREACH(mc, &b->macroses, entries) {
STAILQ_FOREACH(mc, &b->macroses, entry) {
if (!b->maxdepth && RB_EMPTY(&b->stoplist)) {
if (aquery)
bgpq_expand_irrd(b, bgpq_expanded_prefix, b,
"!a%s%s",
"!a%s%s\n",
b->family == AF_INET ? "4" : "6",
mc->text);
else
bgpq_expand_irrd(b, bgpq_expanded_macro, b,
"!i%s,1", mc->text);
"!i%s,1\n", mc->text);
} else {
bgpq_expander_add_already(b, mc->text);
if (pipelining)
bgpq_pipeline(b, bgpq_expanded_macro_limit,
NULL, "!i%s", mc->text);
NULL, "!i%s\n", mc->text);
else
bgpq_expand_irrd(b, bgpq_expanded_macro_limit,
NULL, "!i%s", mc->text);
NULL, "!i%s\n", mc->text);
}
}
@@ -1056,42 +1046,35 @@ bgpq_expand(struct bgpq_expander *b)
}
if (b->generation >= T_PREFIXLIST || b->validate_asns) {
uint32_t i, j, k;
STAILQ_FOREACH(mc, &b->rsets, entries) {
STAILQ_FOREACH(mc, &b->rsets, entry) {
if (b->family == AF_INET)
bgpq_expand_irrd(b, bgpq_expanded_prefix,
NULL, "!i%s,1", mc->text);
NULL, "!i%s,1\n", mc->text);
else
bgpq_expand_irrd(b, bgpq_expanded_v6prefix,
NULL, "!i%s,1", mc->text);
NULL, "!i%s,1\n", mc->text);
}
for (k=0; k < sizeof(b->asn32s) / sizeof(unsigned char *); k++) {
if (!b->asn32s[k])
continue;
for (i=0; i<8192; i++) {
for (j=0; j<8; j++) {
if (b->asn32s[k][i] & (0x80 >> j)) {
if (b->family == AF_INET6) {
if (!pipelining) {
bgpq_expand_irrd(b, bgpq_expanded_v6prefix,
NULL, "!6as%" PRIu32, ( k << 16) + i * 8 + j);
} else {
bgpq_pipeline(b, bgpq_expanded_v6prefix,
NULL, "!6as%" PRIu32, (k << 16) + i * 8 + j);
}
} else {
if (!pipelining) {
bgpq_expand_irrd(b, bgpq_expanded_prefix,
NULL, "!gas%" PRIu32, (k << 16) + i * 8 + j);
} else {
bgpq_pipeline(b, bgpq_expanded_prefix,
NULL, "!gas%" PRIu32, ( k<< 16) + i* 8 + j);
}
}
}
RB_FOREACH(asne, asn_tree, &b->asnlist) {
if (b->family == AF_INET6) {
if (!pipelining) {
bgpq_expand_irrd(b, bgpq_expanded_v6prefix,
NULL, "!6as%" PRIu32 "\n", asne->asn);
} else {
bgpq_pipeline(b, bgpq_expanded_v6prefix,
NULL, "!6as%" PRIu32 "\n", asne->asn);
}
} else {
if (!pipelining) {
bgpq_expand_irrd(b, bgpq_expanded_prefix,
NULL, "!gas%" PRIu32 "\n", asne->asn);
} else {
bgpq_pipeline(b, bgpq_expanded_prefix,
NULL, "!gas%" PRIu32 "\n", asne->asn);
}
}
}
if (pipelining) {
if (!STAILQ_EMPTY(&b->wq))
bgpq_write(b);
@@ -1151,17 +1134,18 @@ bgpq_prequest_freeall(struct bgpq_prequest *bpr) {
void
expander_freeall(struct bgpq_expander *expander) {
struct sx_tentry *var, *nxt;
struct asn_entry *asne, *asne_next;
while (!STAILQ_EMPTY(&expander->macroses)) {
struct sx_slentry *n1 = STAILQ_FIRST(&expander->macroses);
STAILQ_REMOVE_HEAD(&expander->macroses, entries);
struct slentry *n1 = STAILQ_FIRST(&expander->macroses);
STAILQ_REMOVE_HEAD(&expander->macroses, entry);
free(n1->text);
free(n1);
}
while (!STAILQ_EMPTY(&expander->rsets)) {
struct sx_slentry *n1 = STAILQ_FIRST(&expander->rsets);
STAILQ_REMOVE_HEAD(&expander->rsets, entries);
struct slentry *n1 = STAILQ_FIRST(&expander->rsets);
STAILQ_REMOVE_HEAD(&expander->rsets, entry);
free(n1->text);
free(n1);
}
@@ -1180,10 +1164,11 @@ expander_freeall(struct bgpq_expander *expander) {
free(var);
}
for (int i = 0; i < 65536; i++) {
if (expander->asn32s[i] != NULL) {
free(expander->asn32s[i]);
}
for (asne = RB_MIN(asn_tree, &expander->asnlist);
asne != NULL; asne = asne_next) {
asne_next = RB_NEXT(asn_tree, &expander->asnlist, asne);
RB_REMOVE(asn_tree, &expander->asnlist, asne);
free(asne);
}
sx_radix_tree_freeall(expander->tree);

102
extern.h
View File

@@ -29,19 +29,24 @@
#include "sx_prefix.h"
struct sx_slentry {
STAILQ_ENTRY(sx_slentry) entries;
char* text;
struct slentry {
STAILQ_ENTRY(slentry) entry;
char *text;
};
struct sx_slentry* sx_slentry_new(char* text);
struct slentry *sx_slentry_new(char *text);
struct sx_tentry {
RB_ENTRY(sx_tentry) entries;
char* text;
RB_ENTRY(sx_tentry) entry;
char *text;
};
struct sx_tentry* sx_tentry_new(char* text);
struct sx_tentry *sx_tentry_new(char *text);
struct asn_entry {
RB_ENTRY(asn_entry) entry;
uint32_t asn;
};
typedef enum {
V_CISCO = 0,
@@ -70,44 +75,47 @@ typedef enum {
struct bgpq_expander;
struct bgpq_request {
STAILQ_ENTRY(bgpq_request) next;
char *request;
int size, offset;
void *udata;
unsigned int depth;
int (*callback)(char *, struct bgpq_expander *,
struct bgpq_request *);
struct request {
STAILQ_ENTRY(request) next;
char *request;
int size, offset;
void *udata;
unsigned int depth;
int (*callback)(char *, struct bgpq_expander *,
struct request *);
};
struct bgpq_expander {
struct sx_radix_tree *tree;
STAILQ_HEAD(sx_slentries, sx_slentry) macroses, rsets;
RB_HEAD(tentree, sx_tentry) already, stoplist;
int family;
char *sources;
uint32_t asnumber;
int aswidth;
char *name;
bgpq_vendor_t vendor;
bgpq_gen_t generation;
int identify;
int sequence;
unsigned int maxdepth;
unsigned int cdepth;
int validate_asns;
unsigned char *asn32s[65536];
struct bgpq_prequest *firstpipe, *lastpipe;
int piped;
char *match;
char *server;
char *port;
char *format;
unsigned int maxlen;
STAILQ_HEAD(bgpq_requests, bgpq_request) wq, rq;
int fd;
struct sx_radix_tree *tree;
int family;
char *sources;
uint32_t asnumber;
int aswidth;
char *name;
bgpq_vendor_t vendor;
bgpq_gen_t generation;
int identify;
int sequence;
unsigned int maxdepth;
unsigned int cdepth;
int validate_asns;
struct bgpq_prequest *firstpipe, *lastpipe;
int piped;
char *match;
char *server;
char *port;
char *format;
unsigned int maxlen;
int fd;
RB_HEAD(asn_tree, asn_entry) asnlist;
STAILQ_HEAD(requests, request) wq, rq;
STAILQ_HEAD(slentries, slentry) macroses, rsets;
RB_HEAD(tentree, sx_tentry) already, stoplist;
};
int asn_cmp(struct asn_entry *, struct asn_entry *);
RB_PROTOTYPE(asn_tree, asn_entry, entry, asn_cmp);
int bgpq_expander_init(struct bgpq_expander *b, int af);
int bgpq_expander_add_asset(struct bgpq_expander *b, char *set);
int bgpq_expander_add_rset(struct bgpq_expander *b, char *set);
@@ -118,12 +126,12 @@ int bgpq_expander_add_stop(struct bgpq_expander *b, char *object);
int bgpq_expand(struct bgpq_expander *b);
int bgpq4_print_prefixlist(FILE *f, struct bgpq_expander *b);
int bgpq4_print_eacl(FILE *f, struct bgpq_expander *b);
int bgpq4_print_aspath(FILE *f, struct bgpq_expander *b);
int bgpq4_print_asset(FILE *f, struct bgpq_expander *b);
int bgpq4_print_oaspath(FILE *f, struct bgpq_expander *b);
int bgpq4_print_route_filter_list(FILE *f, struct bgpq_expander *b);
void bgpq4_print_prefixlist(FILE *f, struct bgpq_expander *b);
void bgpq4_print_eacl(FILE *f, struct bgpq_expander *b);
void bgpq4_print_aspath(FILE *f, struct bgpq_expander *b);
void bgpq4_print_asset(FILE *f, struct bgpq_expander *b);
void bgpq4_print_oaspath(FILE *f, struct bgpq_expander *b);
void bgpq4_print_route_filter_list(FILE *f, struct bgpq_expander *b);
void sx_radix_node_freeall(struct sx_radix_node *n);
void sx_radix_tree_freeall(struct sx_radix_tree *t);
@@ -134,5 +142,5 @@ void expander_freeall(struct bgpq_expander *expander);
int sx_maxsockbuf(int s, int dir);
#ifndef HAVE_STRLCPY
size_t strlcpy(char* dst, const char* src, size_t size);
size_t strlcpy(char *dst, const char *src, size_t size);
#endif

23
main.c
View File

@@ -124,8 +124,8 @@ version(void)
static void
exclusive(void)
{
fprintf(stderr,"-E, -f <asnum>, -G <asnum>, and -t are mutually "
"exclusive\n");
fprintf(stderr,"-E, -F, -K , -f <asnum>, -G <asnum>, and -t are mutually"
" exclusive\n");
exit(1);
}
@@ -344,7 +344,7 @@ main(int argc, char* argv[])
mc = md = expander.match;
while (*mc) {
if (*mc == '\\') {
if (*(mc+1) == '\n') {
if (*(mc + 1) == '\n') {
*md = '\n';
md++;
mc += 2;
@@ -449,14 +449,18 @@ main(int argc, char* argv[])
case V_CISCO:
case V_MIKROTIK:
expander.aswidth = 4;
break;
case V_CISCO_XR:
expander.aswidth = 6;
break;
case V_JUNIPER:
case V_NOKIA:
case V_NOKIA_MD:
expander.aswidth = 8;
break;
case V_BIRD:
expander.aswidth = 10;
break;
}
} else if (expander.generation == T_OASPATH) {
int vendor = expander.vendor;
@@ -464,12 +468,15 @@ main(int argc, char* argv[])
case V_ARISTA:
case V_CISCO:
expander.aswidth = 5;
break;
case V_CISCO_XR:
expander.aswidth = 7;
break;
case V_JUNIPER:
case V_NOKIA:
case V_NOKIA_MD:
expander.aswidth = 8;
break;
}
}
}
@@ -729,19 +736,19 @@ main(int argc, char* argv[])
sx_report(SX_FATAL,"Unreachable point... call snar\n");
exit(1);
case T_ASPATH:
bgpq4_print_aspath(stdout,&expander);
bgpq4_print_aspath(stdout, &expander);
break;
case T_OASPATH:
bgpq4_print_oaspath(stdout,&expander);
bgpq4_print_oaspath(stdout, &expander);
break;
case T_ASSET:
bgpq4_print_asset(stdout,&expander);
bgpq4_print_asset(stdout, &expander);
break;
case T_PREFIXLIST:
bgpq4_print_prefixlist(stdout,&expander);
bgpq4_print_prefixlist(stdout, &expander);
break;
case T_EACL:
bgpq4_print_eacl(stdout,&expander);
bgpq4_print_eacl(stdout, &expander);
break;
case T_ROUTE_FILTER_LIST:
bgpq4_print_route_filter_list(stdout, &expander);

1306
printer.c

File diff suppressed because it is too large Load Diff

View File

@@ -39,19 +39,15 @@
#include "extern.h"
#include "sx_report.h"
#ifndef SX_MAXSOCKBUF_MAX
#define SX_MAXSOCKBUF_MAX (2 * 1024 * 1024)
#endif
int
sx_maxsockbuf(int s, int dir)
{
{
int optval = 0, voptval;
int hiconf = -1, loconf = -1;
unsigned int voptlen;
int phase = 0, iterations = 0;
if (s < 0) {
if (s < 0) {
sx_report(SX_FATAL,"Unable to maximize sockbuf on invalid "
"socket %i\n", s);
exit(1);
@@ -65,30 +61,27 @@ sx_maxsockbuf(int s, int dir)
return -1;
}
for (;;) {
for (;;) {
iterations++;
if (phase == 0)
optval<<=1;
else {
optval <<= 1;
else {
if (optval == (hiconf + loconf) / 2)
break;
optval = (hiconf + loconf) / 2;
}
if (optval > SX_MAXSOCKBUF_MAX && phase == 0)
break;
if (setsockopt(s, SOL_SOCKET, dir, (void*)&optval,
sizeof(optval)) == -1) {
if (phase == 0)
phase = 1;
phase = 1;
hiconf = optval;
hiconf = optval;
continue;
} else {
} else {
loconf = optval;
}
@@ -99,39 +92,26 @@ sx_maxsockbuf(int s, int dir)
sx_report(SX_ERROR,"getsockopt failed: %s\n",
strerror(errno));
return -1;
} else if (voptval < optval) {
if (phase == 0) {
} else if (voptval < optval) {
if (phase == 0) {
phase = 1;
optval >>= 1;
continue;
} else if (phase == 1) {
} else if (phase == 1) {
phase = 2;
optval -= 2048;
continue;
} else
break;
} else if (voptval >= SX_MAXSOCKBUF_MAX) {
/*
* ... and getsockopt not failed and voptval>=optval.
* Do not allow to increase sockbuf too much even in
* case OS permits it
*/
break;
}
}
voptlen = sizeof(voptval);
if (getsockopt(s, SOL_SOCKET, dir, (void*)&voptval,
&voptlen) == -1) {
sx_report(SX_ERROR,"getsockopt(final stage) failed: %s\n",
sx_report(SX_ERROR,"getsockopt(final stage) failed: %s\n",
strerror(errno));
return -1;
} else {
/*
printf("Finally got %i bytes of recvspace in %i interations\n",
voptval, iterations);
*/
}
return voptval;
} else
return voptval;
}

View File

@@ -133,7 +133,7 @@ sx_prefix_parse(struct sx_prefix *p, int af, char *text)
{
char *c = NULL;
int masklen, ret;
char mtext[INET6_ADDRSTRLEN+5];
char mtext[INET6_ADDRSTRLEN + 5];
strlcpy(mtext, text, sizeof(mtext));
@@ -142,7 +142,7 @@ sx_prefix_parse(struct sx_prefix *p, int af, char *text)
if (c) {
char* eod;
*c = 0;
masklen = strtol(c+1, &eod, 10);
masklen = strtol(c + 1, &eod, 10);
if (eod && eod[0] && !isspace(eod[0])) {
*c = '/';
sx_report(SX_ERROR, "Invalid masklen in prefix %s\n",
@@ -297,7 +297,7 @@ sx_prefix_range_parse(struct sx_radix_tree *tree, int af, unsigned int maxlen,
if (!sx_prefix_parse(p, 0, text)) {
sx_report(SX_ERROR, "Unable to parse prefix %s^%s\n", text,
d+1);
d + 1);
return 0;
}
@@ -323,7 +323,7 @@ sx_prefix_range_parse(struct sx_radix_tree *tree, int af, unsigned int maxlen,
max = maxlen;
} else if (isdigit(d[1])) {
char *dm = NULL;
min = strtoul(d+1, &dm, 10);
min = strtoul(d + 1, &dm, 10);
if (dm && *dm == '-' && isdigit(dm[1])) {
max = strtoul(dm + 1, NULL, 10);
} else if (dm && *dm) {
@@ -483,7 +483,7 @@ sx_prefix_snprintf_fmt(struct sx_prefix *p, FILE *f,
}
c += 2;
} else if (*c == '\\') {
switch(*(c+1)) {
switch(*(c + 1)) {
case 'n':
fprintf(f, "\n");
break;
@@ -925,7 +925,7 @@ sx_radix_node_aggregate(struct sx_radix_node *node)
if (!node->r->isAggregate && !node->l->isAggregate
&& !node->r->isGlue && !node->l->isGlue
&& node->r->prefix->masklen == node->l->prefix->masklen) {
if (node->r->prefix->masklen == node->prefix->masklen+1) {
if (node->r->prefix->masklen == node->prefix->masklen + 1) {
node->isAggregate = 1;
node->r->isGlue = 1;
node->l->isGlue = 1;
@@ -1019,7 +1019,7 @@ sx_radix_node_aggregate(struct sx_radix_node *node)
&& node->l->aggregateHi == node->r->son->aggregateHi
&& node->l->aggregateLow == node->r->son->aggregateLow) {
if (node->l->prefix->masklen == node->prefix->masklen + 1
&& node->r->prefix->masklen == node->prefix->masklen+1) {
&& node->r->prefix->masklen == node->prefix->masklen + 1) {
if (node->isGlue) {
node->l->isGlue = 1;
node->r->son->isGlue = 1;

View File

@@ -31,15 +31,15 @@
#include "extern.h"
struct sx_slentry *
struct slentry *
sx_slentry_new(char *t)
{
struct sx_slentry *e = malloc(sizeof(struct sx_slentry));
struct slentry *e = malloc(sizeof(struct slentry));
if (!e)
return NULL;
memset(e, 0, sizeof(struct sx_slentry));
memset(e, 0, sizeof(struct slentry));
e->text = strdup(t);