Donald Sharp [Thu, 5 May 2022 15:21:20 +0000 (11:21 -0400)]
bgpd: Add a bit of debug to give us data about how an update group is formed
The creation of the update group is a tiny bit of a black box and is hard
to figure out by hand if it is correct. Add a bit of code so that FRR
operators/developers can determine if the update group categorization
makes us happy.
Donald Sharp [Thu, 12 May 2022 20:58:29 +0000 (16:58 -0400)]
bfdd: Prevent coverity from thinking values are uninited
Coverity is claiming that bfdd is able got have bglobal.bg_use_dplane
can be true, while dplane_addr can be uninitialized. Not really
possible since global variables are initialized to all 0's. In
any event. Force it to think it can't go there.
Donald Sharp [Thu, 12 May 2022 20:47:05 +0000 (16:47 -0400)]
bgpd: Assign large enough data structure for coverity to be happy
When implementing the bgp_packet_mpunreach_prefix a uint8_t array
of 3 bytes was created and then assigned to a label type, which
is 4 bytes and then various pointer work is done on it. Eventually
coverity is complaining that the 3 -vs- 4 bytes is not enough
to properly dereference it. Just make the uint8_t 4 bytes
and be done with it.
Donald Sharp [Thu, 12 May 2022 17:23:36 +0000 (13:23 -0400)]
babeld: Check that bodylen is within some bounds of usable
Coverity believed that the bodylen value was read directly
from the incoming packet and then used as a loop variable.
Unfortunately it missed the fact that in babel_packet_examin
the bodylen was actually checked to ensure that it was long
enough. So instead of checking it 2 times, generate it one
time and let coverity figure it out from that.
Donald Sharp [Thu, 12 May 2022 12:06:14 +0000 (08:06 -0400)]
bgpd: Change single value bitfield to a bool
The maxpaths same_clusterlen value was a uint16_t
with a single bit being used. No other values are
being stored. Let's remove the bitfield and simplify
to a bool.
Donald Sharp [Thu, 12 May 2022 13:39:27 +0000 (09:39 -0400)]
bgpd: Prevent crash when issuing various forms of `bgp no-rib`
The `bgp no-rib` command cycles through all the bgp rib tables
and removes them from zebra. Modify the code so that FRR notices
that it is attempting to cycle through the safi's that are two level
tables. In addition these safi's cannot just blindly remove the routes
from the rib as that there are none explicitly.
This code just prevents the crash in bgpd. It does not properly cycle
through and remove the zebra changes made that are explicit to these afi's.
This should be handled as appropriate by the developers on these safi's when
it becomes important to them.
Fixes: #11178 Signed-off-by: Donald Sharp <sharpd@nvidia.com>
anlan_cs [Wed, 11 May 2022 01:48:50 +0000 (21:48 -0400)]
bgpd: remove unncessary check for evpn
In current code, `build_evpn_type2_prefix()` doesn't distinguish ARP
according to the `ip` parameter. The `ip` parameter from caller is
always non-NULL.
Be consistent and not confused, just remove the unnecessary check.
bgpd: remove one unnecessary parameter for evpn-mh
The "add" parameter of `bgp_evpn_mh_route_update()` makes no sense.
Just remove it to clarify this function, and remove the relevant check
with "add" as well.
anlan_cs [Fri, 6 May 2022 09:52:12 +0000 (05:52 -0400)]
bgpd: remove unnecessary check for evpn
When `bgp_evpn_new()` is called, the `bgp` parameter MUST be non-NULL,
remove this unnecessary check and remove the NULL check for returned
`struct bgpevpn *`, which should be non-NULL.
1) Remove check of `bgp_dest_has_bgp_path_info_data(rddest)`.
It is unnecessary(`dest->info` should not be NULL) and misleading.
`if (rddest && bgp_dest_has_bgp_path_info_data(rddest))`
Use (locked) node with this check, but unlock with `if (rddest)`,
The mismatched condition is misleading, there seems to be a
mistake to extra unlock.
Just make it clear, immediately exit with `(!rddest)`.
2) Remove checking returned value for it, and use `void` as return type.
It is unnecessary and wrong. Even the check failed, it should continue
to delete other types of routes.
Just remove the check and go through.
Donald Sharp [Sat, 7 May 2022 01:24:56 +0000 (21:24 -0400)]
isisd: Prevent direct copy of different size prefix'es
Memory is allocated for a `struct prefix_ipv6` but
it was directly copied into a `struct prefix` via
direct pointer copy, which leads to a read past
end of memory. Fix by using prefix_copy
anlan_cs [Tue, 3 May 2022 13:26:15 +0000 (09:26 -0400)]
bfdd: fix override between sessions
After two single-hop sessions (*no local address are configured*) on two
interfaces are UP, remove one address of one interface, both of them
(actually, quite independent sessions) come to be DOWN, not just one.
Consider two boxes: A with `a1` and `a2` adddress on two interfaces,
and B with `b1` and `b2`.
Two sessions are set up and ok: `s1` with <a1,b1> and `s2` with <a2,b2>.
After `a1` of A is removed, there is an unhappy coincidence:
1) On A: `s1` changes local address, and sends <a2,b1> packets with help
of route.
2) On B: wrongly regarded <a2,b1> packets with non-zero remote descriminator
as part of `s2`, and are dropped for mismatched remote remote descriminator.
3) On A: `s1` sends <a2,b1> packets with zero remote descriminator to
initialize this session.
4) On B: wrongly regarded <a2,b1> packets with zero remote descriminator as
part of `s2`. Then `s2` will vibrate.
So the good sessions are overridden.
In this case, the <a2,b1> packets with zero remote descriminator won't take
effect until the current good sessions become bad.
Since single-hop sessions are allowed to be set without bound inteface in
current code, this commit adds one check in `bfd_recv_cb()` to avoid wrong
override.
Donald Sharp [Tue, 3 May 2022 19:24:28 +0000 (15:24 -0400)]
bgpd: peer groups should not toggle interface based nht tracking
Peer groups when various forms of `bgp capability extended-nexthop`
is entered on them are toggling the nexthop tracking status of peers
in their peer group. This is ok when the peer is not interface based.
But it is not ok when the peer is interface based as that it will turn
off the ability of FRR to properly work with that peer type.
Donald Sharp [Tue, 3 May 2022 16:51:21 +0000 (12:51 -0400)]
bgpd: enhanced capability is always turned on for interface based peers
FRR is displaying that the peer enhanced capability command is not
turned on when the interface is part of a peer group. Saving the
config and then reloading actually turns it off.
Fix the code so that FRR does not display the enhanced capability
for interface based peers.
Fixes: #11108 Signed-off-by: Donald Sharp <sharpd@nvidia.com>
*: remove the checking returned value for hash_get()
Firstly, *keep no change* for `hash_get()` with NULL
`alloc_func`.
Only focus on cases with non-NULL `alloc_func` of
`hash_get()`.
Since `hash_get()` with non-NULL `alloc_func` parameter
shall not fail, just ignore the returned value of it.
The returned value must not be NULL.
So in this case, remove the unnecessary checking NULL
or not for the returned value and add `void` in front
of it.
Importantly, also *keep no change* for the two cases with
non-NULL `alloc_func` -
1) Use `assert(<returned_data> == <searching_data>)` to
ensure it is a created node, not a found node.
Refer to `isis_vertex_queue_insert()` of isisd, there
are many examples of this case in isid.
2) Use `<returned_data> != <searching_data>` to judge it
is a found node, then free <searching_data>.
Refer to `aspath_intern()` of bgpd, there are many
examples of this case in bgpd.
Here, <returned_data> is the returned value from `hash_get()`,
and <searching_data> is the data, which is to be put into
hash table.
ewlumpkin [Thu, 10 Feb 2022 22:07:38 +0000 (17:07 -0500)]
bfdd: remove "local_address" of bfd session
The "local_address" of bfd is only used in `show bfd peers brief`
for single hop sessions which are configured without "local address".
Since it is set by destination address of received packet, not
completely correct, so remove it.
projects / mirror / frr.git / log