FRR should create a bnc per peer. Not have
one's that write over others. Currently when
FRR has multiple Interface based peering, BGP wa
creating a single BNC. This is insufficient in that
we were accidently overwriting the one LL with other
data. This causes issues when there are multiple and
there is weird starting issues with those interfaces
that you are peering over.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
Just convert all uses of thread_cancel to THREAD_OFF. Additionally
use THREAD_ARG instead of t->arg to get the arguement. Individual
files should never be accessing thread private data like this.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
Let's just use THREAD_OFF consistently in the code base
instead of each daemon having a special macro that needs to
be looked at and remembered what it does.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
convert:
frr_with_mutex(..)
to:
frr_with_mutex (..)
To make all our code agree with what clang-format is going to produce
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
currently "show bgp summary" and "sho bgp summary wide" commands
provide a description string until a whitespace is occuring this
respectively with size limits of 20 and 60 chars
now theses two commands are providing strings with all
characters until the last witespace before size limit
Signed-off-by: Francois Dumontet <francois.dumontet@6wind.com>
A new command is available under SAFI_MPLS_VPN:
With this command, the BGP vpnvx prefixes received are
not kept, if there are no VRF interested in importing
those vpn entries.
A soft refresh is performed if there is a change of
configuration: retain cmd, vrf import settings, or
route-map change.
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Do not forget to cleanup after we are done:
==395247== 8,268 (32 direct, 8,236 indirect) bytes in 1 blocks are definitely lost in loss record 199 of 205
==395247== at 0x483DD99: calloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==395247== by 0x492EB8E: qcalloc (in /usr/local/lib/libfrr.so.0.0.0)
==395247== by 0x490BB12: hash_get (in /usr/local/lib/libfrr.so.0.0.0)
==395247== by 0x1FBF63: community_intern (in /usr/lib/frr/bgpd)
==395247== by 0x1FC0C5: community_parse (in /usr/lib/frr/bgpd)
==395247== by 0x1F0B66: bgp_attr_community (in /usr/lib/frr/bgpd)
==395247== by 0x1F4185: bgp_attr_parse (in /usr/lib/frr/bgpd)
==395247== by 0x26BC29: bgp_update_receive (in /usr/lib/frr/bgpd)
==395247== by 0x26E887: bgp_process_packet (in /usr/lib/frr/bgpd)
==395247== by 0x4985380: thread_call (in /usr/local/lib/libfrr.so.0.0.0)
==395247== by 0x491D521: frr_run (in /usr/local/lib/libfrr.so.0.0.0)
==395247== by 0x1EBEE8: main (in /usr/lib/frr/bgpd)
Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>
==361630== 24,780 (96 direct, 24,684 indirect) bytes in 3 blocks are definitely lost in loss record 94 of 97
==361630== at 0x483DD99: calloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==361630== by 0x492EB8E: qcalloc (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x490BB12: hash_get (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x1FD3CC: bgp_ca_alias_insert (in /usr/lib/frr/bgpd)
==361630== by 0x2CF8E5: bgp_community_alias_magic (in /usr/lib/frr/bgpd)
==361630== by 0x2C980B: bgp_community_alias (in /usr/lib/frr/bgpd)
==361630== by 0x48E3556: cmd_execute_command_real (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x48E384B: cmd_execute_command_strict (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x48E3D41: command_config_read_one_line (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x48E3EBA: config_from_file (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x499065C: vty_read_file (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x4990FF4: vty_read_config (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x491CB95: frr_config_read_in (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x4985380: thread_call (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x491D521: frr_run (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x1EBEE8: main (in /usr/lib/frr/bgpd)
==361630==
==361630== 24,780 (96 direct, 24,684 indirect) bytes in 3 blocks are definitely lost in loss record 95 of 97
==361630== at 0x483DD99: calloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==361630== by 0x492EB8E: qcalloc (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x490BB12: hash_get (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x1FD39C: bgp_ca_community_insert (in /usr/lib/frr/bgpd)
==361630== by 0x2CF8F4: bgp_community_alias_magic (in /usr/lib/frr/bgpd)
==361630== by 0x2C980B: bgp_community_alias (in /usr/lib/frr/bgpd)
==361630== by 0x48E3556: cmd_execute_command_real (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x48E384B: cmd_execute_command_strict (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x48E3D41: command_config_read_one_line (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x48E3EBA: config_from_file (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x499065C: vty_read_file (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x4990FF4: vty_read_config (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x491CB95: frr_config_read_in (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x4985380: thread_call (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x491D521: frr_run (in /usr/local/lib/libfrr.so.0.0.0)
==361630== by 0x1EBEE8: main (in /usr/lib/frr/bgpd)
Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>
According to RFC9234:
An UPDATE message with a malformed OTC Attribute SHALL be handled
using the approach of "treat-as-withdraw".
Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>
The bgp_nexthop must be the source bgp structure. It cannot be the
destination bgp one.
Use bgp_orig source bgp struct by default.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
The "struct bgp" variable names in the mplsvpn bgp code do not
explicitly say whether they refer to a source or destination BGP
instance. Some variable declarations are commented out with "from" and
"to" but this does not avoid confusion within the functions. The names
of "struct bgp" variables are reused in different functions but their
names sometimes refer to a source instance and sometimes to a
destination instance.
Rename the "struct bgp" variable names to from_bgp and to_bgp.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
==175785== 0 bytes in 1 blocks are definitely lost in loss record 1 of 88
==175785== at 0x483DD99: calloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==175785== by 0x492EB8E: qcalloc (in /usr/local/lib/libfrr.so.0.0.0)
==175785== by 0x269823: bgp_notify_decapsulate_hard_reset (in /usr/lib/frr/bgpd)
==175785== by 0x26C85D: bgp_notify_receive (in /usr/lib/frr/bgpd)
==175785== by 0x26E94E: bgp_process_packet (in /usr/lib/frr/bgpd)
==175785== by 0x4985349: thread_call (in /usr/local/lib/libfrr.so.0.0.0)
==175785== by 0x491D521: frr_run (in /usr/local/lib/libfrr.so.0.0.0)
==175785== by 0x1EBEE8: main (in /usr/lib/frr/bgpd)
==175785==
Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>
The GCC extended printf format checking plugin runs into some GCC
shortcomings regarding casts on printf function parameters. While this
can be fixed with a small GCC patch, patching GCC is "nontrivial" to say
the least. Luckily, it happens that this is /almost/ not an issue for
the FRR source base.
Since we fix SA "misunderstandings" too, let's just fix places where the
format checking plugin runs into this limitation to keep things working
extra smoothly.
(It's not a huge effort either, these two spots in bgpd are the only
places that trigger the plugin limitation, and it's been "clean" before
that for more than a year if my memory is right.)
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
`attr.rmac` is not set in debug as expected for its wrong place in code.
Just move the debug process (`bgp_debug_zebra(NULL)`) after possible `rmac`
value is set.
Signed-off-by: anlan_cs <vic.lan@pica8.com>
Description:
- When there are multiple policies configured with
route-map then the first matching policy is not
getting applied on default route originated with
default-originate.
- In BGP we first run through the BGP RIB and then
pass it to the route-map to find if its permit or
deny. Due to this behaviour the first route in
BGP RIB that passes the route-map will be applied.
Fix:
- Passing extra parameter to routemap_apply so that
we can get the preference of the matching policy,
keep comparing it with the old preference and finally
consider the policy with less preference.
Co-authored-by: Abhinay Ramesh <rabhinay@vmware.com>
Signed-off-by: Iqra Siddiqui <imujeebsiddi@vmware.com>
Description:
- On removing just the route-map from the default-originate config,
update message is not sent to the peer,
and the properties set by route-map persists on peer's end,
until we do a clear bgp.
Fix:
- The flag which is set when default route is originated,
should be unset once "neighbor X.X.X.X default-orginate",
to remove route-map from "neighbor X.X.X.X default-orginate route-map Y",
so as to trigger the flow for sending an update.
Co-authored-by: Abhinay Ramesh <rabhinay@vmware.com>
Signed-off-by: Iqra Siddiqui <imujeebsiddi@vmware.com>
