vivek [Thu, 9 Nov 2017 17:13:28 +0000 (09:13 -0800)]
zebra: Reinstall remote VTEP next hop when it becomes stale
When a remote VTEP next hop entry (for symmetric routing) becomes
stale, reinstall it. This makes the behavior the same as what is
done for remote host next hops (for asymmetric routing and ARP
suppression).
Mitesh Kanjariya [Wed, 18 Oct 2017 08:46:40 +0000 (01:46 -0700)]
bgpd: uninstall all nh/rmac when l3vni is deleted
Upon a l3vni delete (no vni under a vrf) is executed,
we should uninstall all the RMACs and NHs associated with the l3vni.
This is because by the time we get a route delete in zebra
l3vni is already deleted and we dont have refernce to RMACs and NHs
Rafael Zalamena [Mon, 11 Dec 2017 20:25:06 +0000 (18:25 -0200)]
isisd: fix l2 neighbor formations
Add a timestamp information for level 2 circuits, otherwise if the
circuit is marked as already processed on level 1 we will not process
level 2 areas.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
Donald Sharp [Thu, 7 Dec 2017 15:47:30 +0000 (10:47 -0500)]
zebra: Try to move non zapi cli commands from zserv.c
zserv.c has become a bit of a dumping ground for zebra cli.
I'd like to focus the zserv.c code into it's core functionality
which is handling the zapi interface.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Rafael Zalamena [Tue, 12 Dec 2017 13:47:04 +0000 (11:47 -0200)]
isisd: save a clock_gettime() call
Use the thread cached clock to use as start time. It will save a call to
clock_gettime() and also provide a more 'accurate' time measurement from
the start of the procedure.
Donald Sharp [Tue, 5 Dec 2017 15:09:36 +0000 (10:09 -0500)]
bgpd: Cleanup SA error in ignoring return from function
Ignoring the return from argv_find_and_parse_afi
makes the SA system assume that you could pass a AFI_MAX
value to bgp_show_route. Which in turn would cause
an array out of bounds read.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Donald Sharp [Tue, 5 Dec 2017 14:44:11 +0000 (09:44 -0500)]
bgpd: Fix prefix2str using BUFSIZ to PREFIX_STRLEN
PREFIX_STRLEN is the correct length for buffers needed to output
a prefix2str. Additionally cleanup some setting of the last
value to a `\0` this is handled by prefix2str.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Chirag Shah [Fri, 8 Dec 2017 17:33:53 +0000 (09:33 -0800)]
ospfd: prevent passive interface cmd crash
Current OSPF VRF configuration are allow pre-provisining even if
VRF is not configured. In such case ospf->vrf_id would VRF_UNKNOWN,
when passive interface configuration done under such ospf instance,
it would lookup all vrf_device and try to create ifp with unknown
vrf_id.
for passive interface config command lookup ifp for vrf_id is within range.
Donald Sharp [Thu, 7 Dec 2017 23:59:54 +0000 (18:59 -0500)]
ldpd: Switch over to new debug style
When compiling ldpd on a mac, there exists a #define MSG_SEND
which conflicts with a define in ldp_debug.h.
During discussion about this we decided that it would be
better to remove the macro massaging that was going on and
to just call our own #define for it.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Chirag Shah [Fri, 1 Dec 2017 01:45:12 +0000 (17:45 -0800)]
ospf6d: Fix multi nexthop route remove
Fix sorting of route storage to DB.
Fix two list comparison which allows route with
multiple nexthop to updates.
Ticket:CM-19025
Testing Done:
Configured a topology where ospf6 learn
ecmp route via one neighbor and upon removal
of intra-prefix route from origin, DUT removes
ECMP intra-prefix route from RIB.
Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
Don Slice [Wed, 6 Dec 2017 17:00:48 +0000 (09:00 -0800)]
zebra: do not display ipv6 ra commands created by bgpd
If the frr.conf file contains bgp unnumbered peering but the associated
interfaces do not have the commands "no ipv6 nd suppress-ra" and
"ipv6 nd ra-interval 10" configured, when frr-reload.py is issued the
interface commands are removed from the running config, causing peers to
got down and stay down after a link flap. This situation can occur if
the frr.conf file is created manually or via automation (like ansible)
but a subsequent "wr mem" has not been performed.
This fix changes the behavior so that the interface ipv6 nd ra commands
created by bgp are not displayed. Therefore, when the above condition
occurs, there is no difference between the running and stored configs
and peers work fine.
Ticket: CM-18702 Signed-off-by: Don Slice <dslice@cumulusnetworks.com> Reviewed-by: CCR-7004
Testing-done: Manual testing successful. L3-smoke has no new failures
Chirag Shah [Mon, 4 Dec 2017 22:08:23 +0000 (14:08 -0800)]
ospfd: fix crash no router ospf/show running
no router ospf removes default ospf instance,
if there are other non-default vrf instance present
with interface level configuration. Lookup ospf instance
for ifp->vrf_id, if ospf instnace present use that
to access 'instance id'.
Ticket: CM-19078
Testing Done:
run no router ospf and show running config along with other
non-default vrf aware ospf configurations.
Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
Chirag Shah [Mon, 27 Nov 2017 21:59:19 +0000 (13:59 -0800)]
ospfd: Display all vrf aware interface config
OSPF interface specific configuration can be done independent
of router ospf [vrf x] global config.
In cases where ospf interface non default vrf configuration
is done prior to 'router ospf vrf x', show running-config
would not display such configuration.
To display configuration now walk all vrfs and interface list
and only display where OSPF configure params are set.
Ticket:CM-18952
Testing Done:
Tried ospf interface specific configuration with VRF,
where router ospf vrf x is not present.
Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
projects / mirror_frr.git / log