Donald Sharp [Tue, 22 Aug 2017 18:40:24 +0000 (14:40 -0400)]
eigrpd: Start split-horizon
EIGRP was not handling split-horizon. This code starts
down the path of properly considering it. There still
exists situations where we are not properly handling it
though.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Renato Westphal [Mon, 21 Aug 2017 22:43:43 +0000 (19:43 -0300)]
ospf6d: fix regression detected by topotest
With the old API, ospf6d always needed to send a nexthop address and a
nexthop interface when advertising a route to zebra. In the case where
the nexthop address didn't exist (e.g. connected route), zebra would
take care of ignore it in the zread_ipv6_add() function.
Now, if we have a nexthop interface but not a nexthop address, we not
only can but we should send a nexthop of type NEXTHOP_TYPE_IFINDEX. zebra
won't fix bad nexthops anymore because the clients have a proper API to
send correct messages.
Renato Westphal [Mon, 21 Aug 2017 01:10:50 +0000 (22:10 -0300)]
*: use zapi_route to send/receive redistributed routes as well
Some differences compared to the old API:
* Now the redistributed routes are sent using address-family
independent messages (ZEBRA_REDISTRIBUTE_ROUTE_ADD and
ZEBRA_REDISTRIBUTE_ROUTE_DEL). This allows us to unify the ipv4/ipv6
zclient callbacks in the client daemons and thus remove a lot of
duplicate code;
* Now zebra sends all nexthops of the redistributed routes to the client
daemons, not only the first one. This shouldn't have any noticeable
performance implications and will allow us to remove an ugly exception
we had for ldpd (which needs to know all nexthops of the redistributed
routes). The other client daemons can simply ignore the nexthops if
they want or consult just the first one (e.g. ospfd/ospf6d/ripd/ripngd).
As noticed in 657cde1, the zapi_ipv[4|6]_route functions are broken in
many ways and that's the reason that many client daemons (e.g. ospfd,
isisd) need to send handcrafted messages to zebra.
The zapi_route() function introduced by Donald solves the problem
by providing a consistent way to send ipv4/ipv6 routes to zebra with
nexthops of any type, in all possible combinations including IPv4 routes
with IPv6 nexthops (for BGP unnumbered routes).
This patch goes a bit further and creates two new address-family
independent ZAPI message types that the client daemons can
use to advertise route information to zebra: ZEBRA_ROUTE_ADD and
ZEBRA_ROUTE_DELETE. The big advantage of having address-family independent
messages is that it allows us to remove a lot of duplicate code in zebra
and in the client daemons.
This patch also introduces the zapi_route_decode() function. It will be
used by zebra to decode route messages sent by the client daemons using
zclient_route_send(), which calls zapi_route_encode().
Later on we'll use this same pair of encode/decode functions to
send/receive redistributed routes from zebra to the client daemons,
taking the idea of removing code duplication to the next level.
Renato Westphal [Sun, 20 Aug 2017 00:25:12 +0000 (21:25 -0300)]
lib: updates to zapi_route
This patch introduces the following changes to the zapi_route structure
and associated code:
* Use a fixed-size array to store the nexthops instead of a pointer. This
makes the zapi_route() function much easier to use when we have multiple
nexthops to send. It's also much more efficient to put everything on
the stack rather than allocating an array in the heap every time we
need to send a route to zebra;
* Use the new 'zapi_nexthop' structure. This will allow the client daemons
to send labeled routes without having to allocate memory for the labels
(the 'nexthop' structure was designed to be memory efficient and doesn't
have room for MPLS labels, only a pointer). Also, 'zapi_nexthop' is more
compact and more clean from an API perspective;
* Embed the route prefix inside the zapi_route structure. Since the
route's prefix is sent along with its nexthops and attributes, it makes
sense to pack everything inside the same structure.
Renato Westphal [Sat, 19 Aug 2017 17:26:00 +0000 (14:26 -0300)]
zserv: simplify handling of route delete requests
Route attributes like tag, distance and metric are irrelevant when we
want to delete a route from a client daemon. The same can be said about
the nexthops of the route. Only the IP prefix and client protocol are
enough to identify the route we want to remove, considering that zebra
maintains at most one route from each client daemon for each prefix. Once
rib_delete() is called, it deletes the selected route with all of its
nexthops.
Renato Westphal [Tue, 22 Aug 2017 20:01:54 +0000 (17:01 -0300)]
zebra: increase maximum label stack depth
* Bump MPLS_MAX_LABELS from 2 to 16;
* Adjust the static_nh_label structure and the mpls_label2str() function;
* On OpenBSD, print an error message when trying to push more than one
label at once (kernel limitation). While here, add support for MPLSv6
FTNs in OpenBSD.
This is not the full package. We still can't pop multiple labels at once,
or do things like swap a label and push other ones. We'll address that
in the future.
Renato Westphal [Tue, 22 Aug 2017 13:52:07 +0000 (10:52 -0300)]
zebra: fix display of static routes pointing to nonexistent interfaces
Bug introduced a couple of weeks ago by myself. Only happens when the
route has an IP nexthop + a nexthop interface.
Example:
debian(config)# ip route 10.0.1.0/24 172.16.1.10 fake1
debian(config)# do sh run
Building configuration...
[snip]
!
ip route 10.0.1.0/24 172.16.1.10 unknown
!
end
David Lamparter [Tue, 22 Aug 2017 13:27:08 +0000 (15:27 +0200)]
lib: fix const-check in route_node
route_node->lock is "const" if --enable-dev-build is used. This is done
to deter people from messing with internals of the route_table...
unfortunately, the inline'd route_[un]lock_node runs into this.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
When an external route is received, eigrpd will crash. For now,
quietly discard the TLV. Work must be done to handle tlv
processing and changes needed to FSM so it understands an
external should not be chooses if an internal exist.
Signed-off-by: Donnie Savage <diivious@hotmail.com>
Donald Sharp [Sun, 20 Aug 2017 01:21:00 +0000 (21:21 -0400)]
ospf6d: Remove ospf6d version string
This version string has not been updated in over 11 years.
I cannot see any viable reason that we should use or update
or anything with this value, remove.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Donald Sharp [Sun, 20 Aug 2017 00:56:50 +0000 (20:56 -0400)]
eigrpd: Setup eigrp to send FRR version
Send to our peer the major/minor version of FRR that we are
working with. I decided to use FRR instead of the os major/minor
because the os itself doesn't tell you what version of EIGRP
you are actually using.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Donald Sharp [Fri, 18 Aug 2017 13:45:16 +0000 (09:45 -0400)]
bgpd: Fix bgp clear help string
The bgp clear help string was misordered.
New output:
robot.cumulusnetworks.com# clear bgp
(1-4294967295) Clear peers with the AS number
* Clear all peers
A.B.C.D BGP neighbor address to clear
WORD BGP neighbor on interface to clear
X:X::X:X BGP IPv6 neighbor to clear
external Clear all external peers
ipv4 Address Family
ipv6 Address Family
peer-group Clear all members of peer-group
prefix Clear bestpath and re-advertise
view BGP view
vrf BGP VRF
Fixes: #1005 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Jorge Boncompte [Sat, 5 Aug 2017 10:59:05 +0000 (12:59 +0200)]
bgpd: bgp process queue optimization
There are several code paths that dump nodes to the queue for route
processing in a loop. This patch tries to reduce memory allocations/freeing
(work item, list node) and thread scheduling overhead by batching the nodes
in a simple queue list.
In the past when route processing wasn't event driven (bgp_scan()), this used
to have a noticeable impact in table loading, convergence time and memory heap
fragmentation due to reduced alloc's/free's.
Jorge Boncompte [Tue, 8 Aug 2017 18:32:30 +0000 (20:32 +0200)]
lib: cleanup the work queue implementation
Convert the work queue implementation to not use the generic linked list
to mantain the item list and use instead a simple queue from queue.h that
does not allocate memory for each node.
Jorge Boncompte [Mon, 7 Aug 2017 09:47:51 +0000 (11:47 +0200)]
lib: standardize use of queue.h
The simple queue implementation in OpenBSD and FreeBSD are called diferently,
standardize in the use of the FreeBSD version and map the missing names only
if we compile on OpenBSD.
Daniel Walton [Thu, 17 Aug 2017 15:04:25 +0000 (15:04 +0000)]
pimd: "No Path to RP address specified" should exit 0
Signed-off-by: Daniel Walton <dwalton@cumulusnetworks.com>
This could happen if routing isn't up yet but the command did take so we
should exit 0 here. Testing:
root@cel-redxp-10[frr-dwalton76]# vtysh -c 'conf t' -c' ip pim rp
11.11.11.11'
% No Path to RP address specified: 11.11.11.11
root@cel-redxp-10[frr-dwalton76]# echo $?
0
root@cel-redxp-10[frr-dwalton76]#
Donald Sharp [Thu, 17 Aug 2017 14:02:54 +0000 (10:02 -0400)]
eigrpd: Rework ack sent to neighbor
EIGRP schedules and builds packets to send in the future.
Before those packets are being sent, we are receiving
update information from our peer. At that time we
need to reach into the built packet and reset the
outgoing ack to a better value.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
projects / mirror_frr.git / log