Currently, if `bgp max-med on-startup` is configured, after BGP session
is established for the first time, a timer for the specified time is
started. When the timer is expired, an UPDATE message should be sent to
reflect changes in the routes' MED value. The problem is that the routes
are being suppressed because based on the attributes they look like they
have not changed. However, in the case of max-med, the value is copied
to the packet directly from `bgp->maxmed_value`, not from the
attributes. Thus, changes in this case cannot be detected by comparing
attributes.
With this fix, avoid route suppressing when the `max-med on-startup`
timer expires and initiates an UPDATE.
Signed-off-by: Alexander Chernavin <achernavin@netgate.com>
The bgp_gr_restart_retain_routes test is looking for specific output
that does not include the routes nexthop id:
def _bgp_check_kernel_retained_routes():
output = (
r2.cmd("ip route show 172.16.255.1/32 proto bgp dev r2-eth0")
.replace("\n", "")
.rstrip()
)
expected = "172.16.255.1 via 192.168.255.1 metric 20"
diff = topotest.get_textdiff(
output, expected, "Actual IP Routing Table", "Expected IP RoutingTable"
)
if diff:
return False
return True
While the output includes nexthop group id's now:
root@r2:# ip route show 172.16.255.1 proto bgp dev r2-eth0
172.16.255.1 nhid 8 via 192.168.255.1 metric 20
Let's just mark r2 as not to use nexthop groups for installation
and this test issue will go away.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
No mistake, just to unify style for the parameter of function address - remove
ampersand. In current code, only this one place of `hook_register()`s needs
to be made.
Signed-off-by: anlan_cs <vic.lan@pica8.com>
There is code that sets the pi based upon matching it against
the same peer. In this code the type and sub-type are also
compared to the passed in type and sub-type. Let's just use
type and sub-type as that if we have a pi we know type and sub-type
are already correct. This should also make the first iteration
work correctly when the pi has not been created yet when we call
the martian_update function.bgpd: Remove unnecessary check for pi and setting type and sub-type
There is code that sets the pi based upon matching it against
the same peer. In this code the type and sub-type are also
compared to the passed in type and sub-type. Let's just use
type and sub-type as that if we have a pi we know type and sub-type
are already correct. This should also make the first iteration
work correctly when the pi has not been created yet when we call
the martian_update function.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
The issue fixed in the previous commit now correctly triggers a failure:
("assertion (list_add(&head, &itm[j]) == &itm[j]) failed")
Turns out the "shitty" hash function was not shitty enough.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
The typesafe hash data structure enforces items to be unique, but their
hash values may still collide. To this extent, when two items have the
same hash value, the compare function is called to see if it returns 0
(aka "equal").
While the _find() function handles this correctly, the _add() function
mistakenly only checked the first item with a colliding hash value for
equality, and if it was inequal proceeded to add the new item. There
may however be additional items with the same hash value collision, one
of which could still compare as equal. In that case, _add() would
mistakenly add the new element, failing to notice the already added
item. Breakage ensues.
Fix by looking for an equal element among *all* existing items with the
same hash value, not just the first.
Signed-off-by: Francois Dumontet <francois.dumontet@6wind.com>
[DL: rewrote commit message, fixed whitespace/formatting]
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
- for a given IP nexthop, dump all NH entries, including
colored entries, or entries with an ifindex.
- when a given IP nexthop is requested, the path is displayed.
For better readibility, remove the carriage return between
'Last update' and 'Paths', because ctime() function already
performs carriage return.
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
When zebra requests tunnel data it is sending a RTM_GETTUNNEL per
interface that is a VXLAN tunnel. If the kernel that is being
used does not support the particular request type then zebra
will get a error message per tunnel request back. Unfortunately
netlink_parse_info *stops* reading on the first error message.
Therefor one kernels that are returning an error message
let's gather all of those errors. This will allow things
like route reads to actually work properly
Fixes: #12056
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
Section 6.2.7 of RFC 4861 states that a router SHOULD log
inconsistencies in RA information detected on a given link:
```
- Cur Hop Limit values (except for the unspecified value of zero
other inconsistencies SHOULD be logged to system network
management).
- Values of the M or O flags.
- Reachable Time values (except for the unspecified value of zero).
- Retrans Timer values (except for the unspecified value of zero).
- Values in the MTU options.
- Preferred and Valid Lifetimes for the same prefix. If
AdvPreferredLifetime and/or AdvValidLifetime decrement in real
time as specified in Section 6.2.1 then the comparison of the
lifetimes cannot compare the content of the fields in the Router
Advertisement, but must instead compare the time at which the
prefix will become deprecated and invalidated, respectively. Due
to link propagation delays and potentially poorly synchronized
clocks between the routers such comparison SHOULD allow some time
skew.
```
We were not logging inconsistencies if "the unspecified value of zero"
was used for Reachable Time but were logging them for Retrans Timer.
This updates the validation check to also skip the logging of Retrans
Timer inconsistencies if either local/rx value is 0.
Signed-off-by: Trey Aspelund <taspelund@nvidia.com>
When we process a received Router Advertisement we have some logic in
place to detect and log mismatches in a handful of flags/values.
However, these logs do not include what the actual values are, which
means it's up to the operator to grab a packet capture and compare that
against the local configuration...
So let's make life a little easier by including those in the log itself.
Before:
```
2022/09/30 20:37:16 ZEBRA: [KV2V1-7GM7G][EC 4043309149] enp1s0(2): Rx RA - our AdvCurHopLimit doesn't agree with fe80::5054:ff:feca:b085
2022/09/30 20:37:16 ZEBRA: [KS0BP-4GR8K][EC 4043309149] enp1s0(2): Rx RA - our AdvManagedFlag doesn't agree with fe80::5054:ff:feca:b085
2022/09/30 20:37:16 ZEBRA: [RE4EC-VYEJ2][EC 4043309149] enp1s0(2): Rx RA - our AdvOtherConfigFlag doesn't agree with fe80::5054:ff:feca:b085
2022/09/30 20:37:16 ZEBRA: [X6794-9MW18][EC 4043309149] enp1s0(2): Rx RA - our AdvReachableTime doesn't agree with fe80::5054:ff:feca:b085
2022/09/30 20:37:16 ZEBRA: [S1KXC-H8F4W][EC 4043309149] enp1s0(2): Rx RA - our AdvRetransTimer doesn't agree with fe80::5054:ff:feca:b085
```
After:
```
Sep 30 20:45:18 ub20-2 zebra[47487]: [GSW5Z-V7DZN][EC 4043309149] enp1s0(2): Rx RA - our AdvCurHopLimit (14) doesn't agree with fe80::5054:ff:fe9a:e2ca (64)
Sep 30 20:45:18 ub20-2 zebra[47487]: [RHHTS-F96DR][EC 4043309149] enp1s0(2): Rx RA - our AdvManagedFlag (0) doesn't agree with fe80::5054:ff:fe9a:e2ca (1)
Sep 30 20:45:18 ub20-2 zebra[47487]: [MNBY3-FTN6W][EC 4043309149] enp1s0(2): Rx RA - our AdvOtherConfigFlag (0) doesn't agree with fe80::5054:ff:fe9a:e2ca (1)
Sep 30 20:45:18 ub20-2 zebra[47487]: [GG62B-XXWR0][EC 4043309149] enp1s0(2): Rx RA - our AdvReachableTime (20) doesn't agree with fe80::5054:ff:fe9a:e2ca (777)
Sep 30 20:45:18 ub20-2 zebra[47487]: [YG220-D6B4H][EC 4043309149] enp1s0(2): Rx RA - our AdvRetransTimer (13) doesn't agree with fe80::5054:ff:fe9a:e2ca (0)
```
Signed-off-by: Trey Aspelund <taspelund@nvidia.com>
When fabricd is configured to use an interface and there will be
no peers out that interface, the log file is filling up with:
Oct 04 10:50:03 host2 fabricd[1444769]: [HHXDJ-1DA93] ISIS-Adj (1): Threeway state change Initializing to Up
Oct 04 10:50:03 host2 fabricd[1444769]: [R18GA-MS9R7] OpenFabric: Started initial synchronization with 1111.1111.1111 on enp1s0f1np1
Oct 04 10:50:06 host2 fabricd[1444769]: [HHXDJ-1DA93] ISIS-Adj (1): Threeway state change Up to Initializing
Oct 04 10:50:07 host2 fabricd[1444769]: [NT6J7-1RYRF] OpenFabric: Initial synchronization on enp1s0f1np1 timed out!
Oct 04 10:50:07 host2 fabricd[1444769]: [R18GA-MS9R7] OpenFabric: Started initial synchronization with 3333.3333.3333 on enp1s0f0np0
Oct 04 10:50:08 host2 fabricd[1444769]: [HHXDJ-1DA93] ISIS-Adj (1): Threeway state change Up to Initializing
Oct 04 10:50:11 host2 fabricd[1444769]: [NT6J7-1RYRF] OpenFabric: Initial synchronization on enp1s0f0np0 timed out!
Oct 04 10:50:11 host2 fabricd[1444769]: [HHXDJ-1DA93] ISIS-Adj (1): Threeway state change Initializing to Up
Oct 04 10:50:11 host2 fabricd[1444769]: [R18GA-MS9R7] OpenFabric: Started initial synchronization with 1111.1111.1111 on enp1s0f1np1
Oct 04 10:50:14 host2 fabricd[1444769]: [HHXDJ-1DA93] ISIS-Adj (1): Threeway state change Up to Initializing
Oct 04 10:50:15 host2 fabricd[1444769]: [NT6J7-1RYRF] OpenFabric: Initial synchronization on enp1s0f1np1 timed out!
Oct 04 10:50:16 host2 fabricd[1444769]: [R18GA-MS9R7] OpenFabric: Started initial synchronization with 1111.1111.1111 on enp1s0f1np1
Oct 04 10:50:18 host2 fabricd[1444769]: [HHXDJ-1DA93] ISIS-Adj (1): Threeway state change Initializing to Up
The `Threeway state change..` message is guarded by a debug, but the other 2 are not.
Let's guard those with debugs since the log will be filled up rather quickly
with any sort of aggressive timers.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
```
donatas-pc# show ip bgp 100.100.100.0/24 longer-prefixes
BGP table version is 13, local router ID is 10.10.10.10, vrf id 0
Default local pref 100, local AS 65000
Status codes: s suppressed, d damped, h history, * valid, > best, = multipath,
i internal, r RIB-failure, S Stale, R Removed
Nexthop codes: @NNN nexthop's vrf id, < announce-nh-self
Origin codes: i - IGP, e - EGP, ? - incomplete
RPKI validation codes: V valid, I invalid, N Not found
Network Next Hop Metric LocPrf Weight Path
100.100.100.0/24 0.0.0.0 0 32768 i
Displayed 1 routes and 15 total paths
donatas-pc# show ip bgp 100.100.100.0/24
BGP routing table entry for 100.100.100.0/24, version 0
Paths: (1 available, no best path)
Not advertised to any peer
Local
0.0.0.0 (inaccessible, import-check enabled) from 0.0.0.0 (10.10.10.10)
Origin IGP, metric 0, weight 32768, invalid, sourced, local
Last update: Tue Oct 4 11:31:44 2022
donatas-pc# show ip bgp 100.100.100.0/24 json
{
"prefix":"100.100.100.0\/24",
"version":0,
"paths":[
{
"aspath":{
"string":"Local",
"segments":[
],
"length":0
},
"origin":"IGP",
"metric":0,
"weight":32768,
"valid":false,
"version":0,
"sourced":true,
"local":true,
"lastUpdate":{
"epoch":1664872304,
"string":"Tue Oct 4 11:31:44 2022\n"
},
"nexthops":[
{
"ip":"0.0.0.0",
"hostname":"donatas-pc",
"afi":"ipv4",
"accessible":false,
"importCheckEnabled":true,
"used":true
}
],
"peer":{
"peerId":"0.0.0.0",
"routerId":"10.10.10.10"
}
}
]
}
donatas-pc#
```
Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>
The issuing of `show bgp nexthop A.B.C.D` fails even if that
nexthop exists:
eva# show bgp nexthop 192.168.119.120
specified nexthop does not have entry
Fixed:
eva# show bgp nexthop 192.168.119.120
192.168.119.120 valid [IGP metric 0], #paths 0, peer 192.168.119.120
if enp39s0
Last update: Fri Sep 30 14:55:13 2022
Paths:
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
In bgp_open_option_parse the code is checking that the
stream has at least 2 bytes to read ( the opt_type and
the opt_length). However if BGP_OPEN_EXT_OPT_PARAMS_CAPABLE(peer)
is configured then FRR is reading 3 bytes. Which is not good
since the packet could be badly formateed. Ensure that
FRR has the appropriate data length to read the data.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
In peek_for_as4_capability the code is checking that the
stream has at least 2 bytes to read ( the opt_type and the
opt_length ). However if BGP_OPEN_EXT_OPT_PARAMS_CAPABLE(peer)
is configured then FRR is reading 3 bytes. Which is not good
since the packet could be badly formated. Ensure that
FRR has the appropriate data length to read the data.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
Fixed as per rfc7761 section 4.3.1.
"""
Sending Hello Messages
The Address List option advertises all the secondary addresses
associated with the source interface of the router originating the
message. The option MUST be included in all Hello messages if there
are secondary addresses associated with the source interface and MAY
be omitted if no secondary addresses exist.
"""
Issue: #12015
Signed-off-by: Sarita Patra <saritap@vmware.com>
We might disable sending unconfig/shutdown notifications when
Graceful-Restart is enabled and negotiated.
Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>
When primary global v6 unicast address is configured on an
unnumbered interface, BGP does not re-advertise updates out
with the new global v6 address as the nexthop
Signed-off-by: Pdoijode <pdoijode@nvidia.com>
