When enabling adj-rib-in for global RIB table, it automatically
stores adj-rib-entries for VRF RIB tables. This commit introduces
a new vty command to dump the adj-rib-in entries per vrf:
show ip bgp vrf neighbor received-routes
Signed-off-by: Philippe Guibert
---
From: Julien Courtat
>From draft-ietf-bess-evpn-inter-subnet-forwarding-01#section-5.1.1,
Label-1 and Label-2 is not exported systematically. Actually, Label-1
refers to MAC-VRF and is exported to LAYER_2 VRF. Label-2 will act the
same with IP-VRF and is exported to
Enhancement of EVPN bgp information exportation to VRF RIBs.
Currently, only IPv4 unicast VRF RIBs are filled in, according from
EVPN information coming from Prefix route advertisements.
Labels are processed differently according to the VRF layer type.
In addition to this, routermac will not be
This commit appends or withdraw entries in ADJ-RIB-IN entries of VRF
RIB tables.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_route.c | 45 +
1 file changed, 33 insertions(+), 12 deletions(-)
diff --git a/bgpd/bgp_route.c
add vty command : show ipv6 bgp vrf [route]
This command permits to dump IPv6 entries in VRF RIB filled by export RT
facility.
Type 1 route distinguishers are now supported by command show ip bgp vrf
Signed-off-by: Philippe Guibert
---
bgpd/bgp_route.c | 41
From: Julien Courtat
On VRF creation, an additionnal parameter layer type can be provided to
distinguish between layer 2 and layer 3 vrf. layer 2 is used by EVPN
when dealing with MAC/IP route advertisement. layer 3 is used by any
other route L3VPN or EVPN route
When sending an EVPN MAC/IP attribute, the nexthop attribute is also
appended.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_attr.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/bgpd/bgp_attr.c b/bgpd/bgp_attr.c
index
When configuring MPLS or EVPN entries, the NLRI is duplicated.
Which is not wished.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_attr.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/bgpd/bgp_attr.c b/bgpd/bgp_attr.c
index
All current vty show commands related to evpn are moved from
show bgp evpn node to show bgp l2vpn evpn node. Only two
commands are kept: show bgp evpn rd <> [overlay] has been duplicated.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_evpn.c | 105
The introduction of L2VPN prefix in EVPN makes that the inner IP prefix
must be retrieved to display correct information.
Signed-off-by: Julien Courtat
---
bgpd/bgp_mpath.c | 14 +-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git
'packed' keyword had no effect on packing the afi_ethernet structure.
The attribute keyword has been eppended in order to take into account
the packed feature.
Signed-off-by: Julien Courtat
---
lib/prefix.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
As per draft-ietf-idr-tunnel-encaps-02, section 3.2.1, BGP Encap
attribute supports vxlan tunnel type. A new tunnel attribute has been
appended to subtlv list, describing the vxlan network identifier to
be used for the routing information of the BGP update message.
Signed-off-by: Philippe Guibert
This commits avoids keeping a warning of compilation.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_open.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/bgpd/bgp_open.c b/bgpd/bgp_open.c
index 75c840ef4958..ce1fea16971e 100644
--- a/bgpd/bgp_open.c
+++
As per draft-ietf-bess-evpn-inter-subnet-forwarding-01, chapter 6.1,
a new extended community called router's MAC Extended Community is
provided. This community is appended to extended community list.
Note that a show API has been changed in order to be able to not
display (or not) this new type.
A new vty command available under evpn address family. This command
takes following format:
(af-evpn)# network A.B.C.D/M rd ASN:nn_or_IP-address:nn ethtag WORD
label WORD esi WORD gwip A.B.C.D routermac WORD
Among new parameters, ethtag stands for the ethernet tag indentifier.
For each received routermac extended community, the mac address is
copied into routermac extended community context. For each emission,
a check is done against routermac extended community, if L2VPN is
enabled. If enabled, the extended community is appended.
Signed-off-by: Philippe Gubiert
The commit introduces the changes to be done to carry route type 5 EVPN
information in bgp extra attribute information. The commit also handles
the update processing for route type 5 information.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_attr.c | 16 -
This patch introduces show show bgp evpn commands to dump
NLRI entries configured or received on BGP, related to EVPN
route type 5. New command introduced is the following:
show bgp evpn [all | rd ] [overlay]
This command displays gw ip field of the RT-5 message in the
nexthop field of the show
This patch appends nexthop attribute to EVPN message, in addition
to appending gateway IP in RT-5 NLRI itself. In reception, if
both informations are stored, then the GW IP information will
supersede the NLRI nexthop attribute.
Signed-off-by: Philippe Guibert
---
This patch introduces code to receive a NLRI message with route type
5, as defined in draft-ietf-bess-evpn-prefix-advertisement-02. It
It increases the number of parameters to extract from the NLRI and
to store into bgp extra information structure. Those parameters are
the ESI (ethernet segment
This patch introduces the ability to make route type 5 message
when EVPN is enabled. Picked up paramters are collected from the
bgp extra attribute structure and are the ESI, the ethernet tag
information. In addition to this, nexthop attribute is collected too.
Signed-off-by: Philippe Guibert
This commit checks for the maximum path per VRF before running the
multipath algorithm. This makes it possible to limit the number of
paths for a specific VRF, and not for an other one.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_mpath.c | 34
There has been a change in API safi_valid_indices.
This modification is updating the call API.
Signed-off-by: Philippe Guibert
---
tests/bgp_capability_test.c | 13 +++--
tests/bgp_mp_attr_test.c| 9 +
2 files changed, 12 insertions(+), 10
As explained in a previous mail addressed to quagga-dev, Tue, Oct 25,
2016 at 2:46 PM, the proposed implementation covers the following:
- EVPN route type 2 aka MAC/IP advertisement
- EVPN route type 5 aka Prefix IP advertisement.
It is possible to statically set abovementioned messages.
It also
From: Christian Franke
This enhancement permits to export received or locally set route
entries for MPLS VPN, to the appropriate VRF. By introducing the
RFC4360 and BGP extended communities, some export route target are
received on BGP update, or some import route
From: Julien Courtat
Enhancement of new vty command to configure the maximum number of
multipath entries that are possible within a VRF RIB table.
The following command is available:
- vrf rd maximum-path <1-64>
- no vrf rd maximum-path <1-64>
Vrfs with a
This sanity check has been discovered when running bgp mpath_tests
that are available in tests folder.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_mpath.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/bgpd/bgp_mpath.c b/bgpd/bgp_mpath.c
index
From: David Lamparter
Support for multiple label support handling in NLRI messages.
Ability to receive and sent BGP updates with NLRI containing multiple
labels. The commit follows label encoding rules, contained in RFC3107.
Signed-off-by: David Lamparter
Introduction of EVPN address-family node that is reachable under
BGP node, like follow: address-family evpn. Under that node, only
some commands are available for configuring a neighbor or setting
some basic settings like attribute unchanged behaviour. Currently,
no vty command is available for
If the user preconfigures a vrf with import and export rules, before
creating a new network on vpnv4 address, then BGP update emitted will
contain the exported rts mentioned by the vrf export rule.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_route.c | 11 +++
From: Julien Courtat
Update extract.pl ignore list to prevent from installing
address_family_evpn_cmd twice and have "Ambiguous command" return by
vtysh when trying to configure address-family evpn.
Signed-off-by: Julien Courtat
---
bgp NLRI has a specific format for label, terminating with bottom
of stack bit. This value is updated in all BGP attribute tests. A
specific test has been added to check for a list of labels contained
in a single NLRI.
Signed-off-by: Philippe Guibert
---
To handle BGP NLRI EVPN messages, bgp is modified to handle AFI_L2VPN
and SAFI_EVPN values.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_attr.c | 25 ++-
bgpd/bgp_open.c | 29 ++-
bgpd/bgp_packet.c | 28
On adding/removing/updating NLRI entries into VRF RIB tables, a BGP
event is fired. Note that only selected entries, and entries that are
part of multipath entries, and where one of the multipath entry is
selected.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_route.c
sometimes, when electing best route algorithm, a static route
may be injected, thus causing all old multipath entries to be
removed from multipath list. As a loop is removing each mpath
entry from mpath list, it is necessary to update the pointer.
Signed-off-by: Philippe Guibert
This commit introduces the BGP VRF configuration, and BGP VRF RIB
table.
It includes the ability for a BGP to configure its own route
distinguisher ( aka VRF). New vty commands introduced:
(config-router)# vrf rd
This structure permits configuring import and export route targets,
which is
From: David Lamparter
API enhancement to be able to write and read 3 byte series in
stream structures.
Signed-off-by: David Lamparter
Signed-off-by: Philippe Guibert
---
lib/stream.c | 77
If vpnv4 capability has to be sent, then the correct safi parameter
is sent instead of internal value.
In case BGP open graceful restart capability is applied to EVPN, then
the flags to be set are modified to be official afi and safi parameters,
and not internal value.
Signed-off-by: Philippe
NLRI entries received from MP-BGP peer exchange are exported in VRF RIB
tables. This commit introduces the BGP VRF processing handler that can
be used for further processing into VRF RIB tables: best selection
algorithm, multipath.
Signed-off-by: Philippe Guibert
---
Because afi numbers can reach high values, a conversion is necessary
to return associate afi internal identifier. The safi_valid_indices
API has been changed for that.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_open.c | 22 +++---
bgpd/bgp_open.h
Enhance a function that permits knowing if two entries belong to the
same multipath group.
Signed-off-by: Philippe Guibert
---
bgpd/bgp_mpath.c | 15 +++
bgpd/bgp_mpath.h | 1 +
2 files changed, 16 insertions(+)
diff --git a/bgpd/bgp_mpath.c
From: David Lamparter
Signed-off-by: David Lamparter
Signed-off-by: Philippe Guibert
---
tests/Makefile.am | 2 +-
tests/bgp_mpath_test.c | 15 +++
2 files changed, 16 insertions(+), 1
42 matches
Mail list logo