Problem:
frr(config)# show ip pim rp-info
RP address group/prefix-list OIF I am RP Source Group-Type
109.0.0.3 224.0.0.0/4 ens192 no Static ASM
frr(config)# int ens192
frr(config-if)# no ip pim
frr(config) show ip pim rp-info
RP address group/prefix-list OIF I am RP Source Group-Type
109.0.0.3 224.0.0.0/4 ens192 no Static ASM
rp-info OIF is still having ens192 which is PIM disabled.
Fixing this as part of this PR.
Issue: #12044
Signed-off-by: Sarita Patra <saritap@vmware.com>
FRR implements a non-standard, but compatible approach for
sending update LSAs (it always send to 224.0.0.5) on P2MP
interfaces. This change makes it so acks are also sent to
224.0.0.5.
Since the acks are multicast, this allows an optimization
where we don't send back out the incoming P2MP interface
immediately allow time to rx multicast ack from neighbors
on the same net that rx'ed the original (multicast) update.
Signed-off-by: Lou Berger <lberger@labn.net>
Column headers in BGP routes table are not aligned with data when
RPKI status is available. This was fixed to insert a space at the
beginning of the header and at the beginning of lines that do not
have RPKI status.
This fix requires that several testing templates be adjusted to
match the new output.
Signed-off-by: Wayne Morrison <wmorrison@netgate.com>
Problem:
Multiple memory leaks in ospf6.
260 ==6637== 32 bytes in 1 blocks are definitely lost in loss record 5 of 24
261 ==6637== at 0x4C31FAC: calloc (vg_replace_malloc.c:762)
262 ==6637== by 0x4E8A1BF: qcalloc (memory.c:111)
263 ==6637== by 0x11EE27: ospf6_summary_add_aggr_route_and_blackhole (ospf6_asbr.c:2779)
264 ==6637== by 0x11EEBA: ospf6_originate_new_aggr_lsa (ospf6_asbr.c:2811)
265 ==6637== by 0x4E7C6A7: hash_clean (hash.c:325)
266 ==6637== by 0x11FA93: ospf6_handle_external_aggr_update (ospf6_asbr.c:3164)
267 ==6637== by 0x11FA93: ospf6_asbr_summary_process (ospf6_asbr.c:3386)
268 ==6637== by 0x4EB739B: thread_call (thread.c:1692)
269 ==6637== by 0x4E85B17: frr_run (libfrr.c:1068)
270 ==6637== by 0x119535: main (ospf6_main.c:228)
356 ==6637== 240 bytes in 12 blocks are indirectly lost in loss record 13 of 24
357 ==6637== at 0x4C2FE96: malloc (vg_replace_malloc.c:309)
358 ==6637== by 0x4E8A0DA: qmalloc (memory.c:106)
359 ==6637== by 0x13545C: ospf6_lsa_alloc (ospf6_lsa.c:724)
360 ==6637== by 0x1354E3: ospf6_lsa_create_headeronly (ospf6_lsa.c:756)
361 ==6637== by 0x1355F2: ospf6_lsa_copy (ospf6_lsa.c:790)
362 ==6637== by 0x13B58B: ospf6_dbdesc_recv_slave (ospf6_message.c:976)
363 ==6637== by 0x13B58B: ospf6_dbdesc_recv (ospf6_message.c:1038)
364 ==6637== by 0x13B58B: ospf6_read_helper (ospf6_message.c:1838)
365 ==6637== by 0x13B58B: ospf6_receive (ospf6_message.c:1875)
366 ==6637== by 0x4EB739B: thread_call (thread.c:1692)
367 ==6637== by 0x4E85B17: frr_run (libfrr.c:1068)
368 ==6637== by 0x119535: main (ospf6_main.c:228)
RCA:
1. when the ospf6 area is being deleted, the neighbor related information
was not being cleaned up.
2. when aggr route gets deleted from rt_aggr_tbl the corrsponding summary
route attched to the aggr route was not being deleted.
Fix:
Added the ospf6_neighbor_delete in ospf6_area_delete to free the
neighbor related information and added ospf6_route_delete while
freeing external aggr route to free the summary route.
Signed-off-by: Manoj Naragund <mnaragund@vmware.com>
vtysh show running-config is showing:
frr version 8.3.1_git
frr defaults traditional
hostname test
log file /etc/frr/frr.log informational
log timestamp precision 3
domainname
service integrated-vtysh-config
domainname should not be printed in this case at all. If the
host has no search/domainname configured, frr_reload.py
crashes on invalid config from `vtysh show running-config`
Basically the same change as commit a7141b8
Signed-off-by: Nico Berlee <nico.berlee@on2it.net>
Add a default limit to the InQ for messages off the bgp peer
socket. Make the limit configurable via cli.
Adding in this limit causes the messages to be retained in the tcp
socket and allow for tcp back pressure and congestion control to kick
in.
Before this change, we allow the InQ to grow indefinitely just taking
messages off the socket and adding them to the fifo queue, never letting
the kernel know we need to slow down. We were seeing under high loads of
messages and large perf-heavy routemaps (regex matching) this queue
would cause a memory spike and BGP would get OOM killed. Modifying this
leaves the messages in the socket and distributes that load where it
should be in the socket buffers on both send/recv while we handle the
mesages.
Also, changes were made to allow the ringbuffer to hold messages and
continue to be filled by the IO pthread while we wait for the Main
pthread to handle the work on the InQ.
Memory spike seen with large numbers of routes flapping and route-maps
with dozens of regex matching:
```
Memory statistics for bgpd:
System allocator statistics:
Total heap allocated: > 2GB
Holding block headers: 516 KiB
Used small blocks: 0 bytes
Used ordinary blocks: 160 MiB
Free small blocks: 3680 bytes
Free ordinary blocks: > 2GB
Ordinary blocks: 121244
Small blocks: 83
Holding blocks: 1
```
With most of it being held by the inQ (seen from the stream datastructure info here):
```
Type : Current# Size Total Max# MaxBytes
...
...
Stream : 115543 variable 26963208 15970740 3571708768
```
With this change that memory is capped and load is left in the sockets:
RECV Side:
```
State Recv-Q Send-Q Local Address:Port Peer Address:Port Process
ESTAB 265350 0 [fe80::4080:30ff:feb0:cee3]%veth1:36950 [fe80::4c14:9cff:fe1d:5bfd]:179 users:(("bgpd",pid=1393334,fd=26))
skmem:(r403688,rb425984,t0,tb425984,f1816,w0,o0,bl0,d61)
```
SEND Side:
```
State Recv-Q Send-Q Local Address:Port Peer Address:Port Process
ESTAB 0 1275012 [fe80::4c14:9cff:fe1d:5bfd]%veth1:179 [fe80::4080:30ff:feb0:cee3]:36950 users:(("bgpd",pid=1393443,fd=27))
skmem:(r0,rb131072,t0,tb1453568,f1916,w1300612,o0,bl0,d0)
```
Signed-off-by: Stephen Worley <sworley@nvidia.com>
Ensure that un-configuring allowas-in for a peer or group
clears the related flags and integer value. Tighten the use
of the integer counter so that it's only used when the config
flag is set. Add show output if allowas-in is enabled.
Signed-off-by: Mark Stapp <mstapp@nvidia.com>
Fix potential access to NULL pointer in isis_route_update even if it is
not related with the fast-reroute implementation.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
Add a switchover test that consists in:
- Setting up ISIS BFD between rt1 and rt2
- The no link-detect setting on rt1 eth-rt2 is still present so that
zebra does not take account linkdown events on this interface.
- Shutting down rt1 eth-rt2 from the switch side
- Wait for BFD to comes down
Check that the switchover between primary and backup happens before the
SPF re-computation.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
Add a switchover test that consists in clearing the rt2 neighbor on rt1.
Check that the switchover between primary and backup happens before the
SPF re-computation.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
Add a switchover test that consists in:
- Setting no link-detect on rt1 eth-rt2 so that zebra does not take
account linkdown events on this interface.
- Shutting down rt1 eth-rt2 from the switch side
- Wait for the hello timer expiration
Check that the switchover between primary and backup happens before the
SPF re-computation.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
Add a switchover test that consists in shutting down an interface.
Check that the switchover between primary and backup happens before the
SPF re-computation.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
If ISIS is running on an IPv6 only topology, the command "spf interval"
has no effect.
Only the IPv4 SPF tree timers are taken into account.
Base the next SPF scheduling on the last running SPF tree.
Fixes: be985ba059 ("isisd: make use of advanced concepts like arrays and loops")
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
Add a switchover test that consists in:
- Setting up ISIS BFD between rt5 and rt6
- Setting no link-detect on rt6 eth-rt5 so that zebra does not take
account linkdown events on this interface.
- Shutting down rt6 eth-rt5 from the switch side
Check that the switchover between primary and backup happens before the
SPF re-computation.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
The "RTLIST..." list should be maintained integrity. If wildcard check
failed, it should immediately return failure. Otherwise user configuration
will be partial.
```
anlan(config-router-af)# route-target export *:55 33:33
% Wildcard '*' only applicable for import
anlan(config-router-af)# route-target both *:55 33:33
% Wildcard '*' only applicable for import
```
With this commit, the RTs without wildcard will not be executed as before. And
the same for `no` form.
Signed-off-by: anlan_cs <vic.lan@pica8.com>
When disabling and re-enabling the "mpls-te on" command, the LS TED
database is deleted but not re-created.
Cleanup the LS TED database instead of deleting it when disabling
"mpls-te".
Fixes: 1fa6385040.. ("isisd: Correct Valgrind errors")
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
If MPLS TE is enabled, the router encodes the local and remote interface
IP address in the "Extended Reachability" TLV.
> east-vm(config)# do show isis database detail east-vm.00-00
> Extended Reachability: 0007.e901.3333.00 (Metric: 10)
> Local Interface IP Address(es): 10.126.0.2
> Remote Interface IP Address(es): 10.126.0.3
> Maximum Bandwidth: 1.76258e+08 (Bytes/sec)
The remote interface is added when the circuit adjacency comes up after
setting MPLS TE. However, if MPLS TE is enabled after, the remote
address is not added. It happens after disabling and re-enabling the
MPLS TE.
> east-vm(config)# router isis 1
> east-vm(config-router)# no mpls on
> east-vm(config-router)# mpls on
> east-vm(config)# do show isis database detail east-vm.00-00
> Extended Reachability: 0007.e901.3333.00 (Metric: 10)
> Local Interface IP Address(es): 10.126.0.2
> Maximum Bandwidth: 1.76258e+08 (Bytes/sec)
Update the remote IPv4 and IPv6 of all adjacencies after enabling MPLS
TE.
Fixes: 1b3f47d04c ("isisd: Update TLVs processing for TE, RI & SR")
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
Add a switchover test that consists in shutting down an interface.
Check that the switchover between primary and backup happens before the
SPF re-computation.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
isis_ifp_down() may in some circumstances be called twice on a down
interface event.
Avoid applying fast-reroute on an already down interface.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
When an IS-IS interface is coming down, fast-reroute may be triggered
twice: a first time after the detection of the interface down event and
a second time after the detection of the adjacency down (because of the
expiration of the ISIS Hello or BFD timers).
Avoid a BFD down event from running fast-reroute another time if the
interface was already detected down.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
Backup routes are sent to zebra by routing daemons such as isisd so that
the dataplane can pre-install them with a lower priority. When an
interface comes down, the associated primary routes are discarded by the
dataplane and the backup ones take over.
However, some dataplanes (e.g. Netlink ones) do not pre-install the
backup routes. Associated prefixes have no next-hop until SPF is
recomputed.
Apply fast-reroute as soon as an interface falls down by sending route
UPDATEs to zebra.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
Allow the nexthoplookup function to return the first nexthop found on
ifindex interface if the IP is unspecified.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
When a adjacency falls down, the primary routes are not deleted on the
dataplane until the SPF is recomputed. Even the backup routes are
pre-installed on the dataplane, there is no fast-route optimization.
Reasons for an adjacency to come down are:
- BFD down
- Hello timer timeout
- User adjacency clear
Apply the backup route switchover for fast-reroute as soon an IS-IS
adjacency falls down before the first SPF re-computation. Pre-computed
backup routes are applied sooner.
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
Fix the vni_str NULL check for wildcard route-targets
in evpn show run. This will never be NULL if we add 1
here. Though it should also never be NULL since ":" should
always exist. Better to be safe than sorry.
Signed-off-by: Stephen Worley <sworley@nvidia.com>
