Merge pull request #12601 from opensourcerouting/feature/bgp_neighbor_path-attribute_discard

bgpd: Add `neighbor path-attribute discard` command

Merge pull request #12597 from opensourcerouting/fix/bgp_sender_as_path_prevention

bgpd: Do not send routes back received from a peer

Merge pull request #12424 from opensourcerouting/static-route-bfd

staticd: BFD static route monitoring

Merge pull request #12621 from donaldsharp/route_scale_failure_micronet

tests: route_scale tests are failing on micronet

Merge pull request #12642 from anlancs/fix/bgpd-debug-name

bgpd: fix wrong vrf name for debug

bgpd: fix wrong vrf name for debug

For vrf name in debug, use `bgp->name_pretty` instead of `bgp->name`.

Before:
```
2023/01/15 05:04:19 BGP: [P4GAZ-JHRM3] evpn vrf VRF default nh init
2023/01/15 05:04:19 BGP: [ZZKY3-FX5JH] bgp_get: Registering BGP instance (null) to zebra <-
2023/01/15 05:04:19 BGP: [TNK7N-FJF7K] Registering VRF 0
```

After:
```
2023/01/15 21:38:16 BGP: [P4GAZ-JHRM3] evpn vrf VRF default nh init
2023/01/15 21:38:16 BGP: [ZZKY3-FX5JH] bgp_get: Registering BGP instance VRF default to zebra <-
2023/01/15 21:38:16 BGP: [TNK7N-FJF7K] Registering VRF 0
```

Signed-off-by: anlan_cs <vic.lan@pica8.com>

Merge pull request #12611 from vfreex/reload-missing-daemons

tools: Add missing daemons

tests: route_scale tests are failing on micronet

I'm seeing test failures after in micronet runs in CI
after 7 seconds * 30 attempts at seeing if it succeeds.
Let's see if another 60 seconds of attempts allows
this to work properly.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>

tools: Add missing daemons

Got `ERROR: Daemon babeld is not a valid option for 'show running-config'` when using `frr-reload.py --reload --daemon babeld`.

Adds `babeld` and `nhrpd` as valid daemons.

Signed-off-by: Yuxiang Zhu <vfreex@gmail.com>

Merge pull request #12639 from donaldsharp/breakup_bgp_vrf_dynamic_route_leak_topo4

tests: Breakup bgp_vrf_dynamic_route_leak_topo4

bgpd: Do not send routes back received from a peer

Before this patch, we needed to explicitly define a neighbor to be SOLO
(= separate update-group). Let's ease this functionality for an operator to
avoid confusions.

Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>

tests: Check if routes are not sent back from the sender

Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>

bgpd: Add `neighbor path-attribute discard` command

The idea is to drop unwanted attributes from the BGP UPDATE messages and
continue by just ignoring them. This improves the security, flexiblity, etc.

This is the command that Cisco has also.

Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>

doc: Add `neighbor path-attribute discard` command

Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>

tests: Check if we can discard unwanted attributes from the paths

Using `neighbor path-attribute discard ...` command.

Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>

Merge pull request #12635 from LabNConsulting/ziemba/topotests-lutil-wait-strict

topotests/lib/lutil.py: luSetWaitType("strict"|"nostrict")

tests: Breakup bgp_vrf_dynamic_route_leak_topo4

Single run of this test suite on my machine was 8 minutes.
Breaking this up into 3 test suites halves the run time.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>

topotests/lib/lutil.py: luSetWaitType("strict"|"nostrict")

    This change alters the behavior of existing test code. The
    default mode (before any call to luSetWaitType()) is now
    "strict".

    The historical behavior of luCommand(op="wait) is to ignore
    failures to match the specified regexp in the specified time.
    In those cases, no result was logged and no error was signaled.

    This change introduces a new "strict" mode for luCommand(op="wait):
    in "strict" wait mode, each invocation of luCommand(op="wait)
    generates an explicit, logged failure result when it fails to match
    the specified regexp in the specified time. These failures signal
    an error for the test.

    Calling luSetWaitType("nostrict") restores the historical behavior.

    Calling luSetWaitType("strict") (re)enables the new strict behavior.

    Individual calls to luCommand() may also specify op="wait-nostrict"
    to override any default and use the historical behavior.

    Individual calls to luCommand() may also specify op="wait-strict"
    to override any default and use the new behavior.

Signed-off-by: G. Paul Ziemba <paulz@labn.net>

staticd: add debug static bfd command

This command helps in troubleshooting static bfd feature.
Add traces upon bfd events.

Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>

topotests: test BFD static route integration

Test that BFD static monitoring works:
When BFD session is up the routes are installed in the RIB and
distributed with routing protocol (in this case BGP). When the session
is down it is removed from RIB and propagated.

Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>

doc: document new static BFD commands

Let the user know how to use the static route monitoring commands.

Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>

staticd: fix bug on route reinstallation

When configuring a route with multiple next hops on boot and the
interface is missing the route never shows up (even after interface is
properly learned).

Fix: force route clean up by uninstalling it and putting it back
entirely instead of just updating the missing next hop.

Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>

staticd: add command to show monitored routes

Add command to visualize all monitored routes and their state.

Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>

zebra: send BFD messages to staticd

Add logic to allow `zebra` to reroute BFD messages for `staticd`.

Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>

lib,staticd: BFD integration CLI commands

Implement the static route monitoring feature with BFD integration.

Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>

staticd: BFD integration northbound support

Implement all BFD integration northbound callbacks and integrate BFD
with `staticd` route installation procedure.

Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>

yang: static route BFD monitoring

Define a generic BFD monitoring group template and use it to add support
for static route monitoring.

Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>

lib: BFD integration clean up function

Implement clean up function to be called on shutdown to make daemon exit
clean for valgrind and other memory sanitizers.

Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>

lib: BFD automatic source selection

Implement new BFD library issue to allow protocols to configure BFD
sessions with automatic source selection.

The source selection will be based on the Next Hop Tracking feature:
`zebra` will do RIB lookups to determine the output interface and the
primary source address of that interface will be used as source.

Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>

Merge pull request #12637 from donaldsharp/revert_pr_11127

Revert "Merge pull request #11127 from louis-6wind/bgp-leak"

Revert "tests: Fix wrong interface name in SRv6 topotest"

This reverts commit 647c38eab2d6a1d05f6f683185c58ecf3015a4dc.

Merge pull request #12626 from opensourcerouting/fix/bgpd_neighbor_password_unnumbered

bgpd: Warn user about mixed order of commands executed

Revert "Merge pull request #11127 from louis-6wind/bgp-leak"

This reverts commit 16aa1809e7c8caad37e8edd4e5aaac4f344bc7d3, reversing
changes made to f616e716089b16d9a678846282a6ac5c55e31a56.

Merge pull request #12629 from opensourcerouting/fix/use_static_repo

github: Use static quay.io repository for docker images

Merge pull request #12627 from donaldsharp/pim_igmp_vrf_micronet_failure

tests: the pim_igmp_vrf topotest fails under high load

Merge pull request #12595 from chiragshah6/fdev1

bgpd: show bgp vrfs per vrf level json support

doc: add documentation for show bgp vrfs

Signed-off-by: Chirag Shah <chirag@nvidia.com>

bgpd: show bgp vrfs per vrf level json support

json support extended for show [ip] bgp vrfs <vrf-name> json

Before:
```
tor-2# show ip bgp vrfs default json
% JSON option not yet supported for specific VRF
tor-2#
tor-2# show bgp vrfs sym_1 json
% JSON option not yet supported for specific VRF
tor-2#
```

After:
```
tor-1# show ip bgp vrfs default json
{
  "default":{
    "type":"DFLT",
    "vrfId":0,
    "routerId":"27.0.0.7",
    "numConfiguredPeers":2,
    "numEstablishedPeers":2,
    "l3vni":0,
    "rmac":"00:00:00:00:00:00",
    "interface":"unknown"
  }
}
tor-1#
tor-1# show bgp vrfs sym_1 json
{
  "sym_1":{
    "type":"VRF",
    "vrfId":21,
    "routerId":"27.0.0.81",
    "numConfiguredPeers":0,
    "numEstablishedPeers":0,
    "l3vni":8888,
    "rmac":"44:38:39:ff:ff:25",
    "interface":"vlan490_l3"
  }
}
tor-1#

tor-1# show bgp vrfs test json
{
}
tor-1#
```

Ticket:#3314672

Issue:3314672

Testing: UT done

Signed-off-by: Chirag Shah <chirag@nvidia.com>
Signed-off-by: Sindhu Parvathi Gopinathan <sgopinathan@nvidia.com>

Merge pull request #12609 from mjstapp/fix_bfd_check_len

bfdd: validate incoming control message length

github: Drop temporary docker-stable github action

Until we have a consensus how we ship stable/X.Y docker images.

For now we build images based on release tags, not based on HEAD of the branch.

Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>

github: Use static quay.io repository for docker images

Not dynamic from username, because tags do not allow using "+" character.

Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>

Merge pull request #12623 from anlancs/fix/zerbra-debug-cosmetic-changes

zebra: cosmetic changes for debug

tests: Pim vrf's need 4.19 or better to run properly not 4.15

Tests are failing in micronet because linux kernel needs are 4.19
not 4.15

2023-01-11 17:15:06,657.657 INFO: topolog.r1: vtysh command => "show zebra"
2023-01-11 17:15:06,657.657 DEBUG: topolog.r1: LinuxNamespace(r1): cmd_status("['/bin/bash', '-c', 'vtysh  -c "show zebra" 2>/dev/null']", kwargs: {'encoding': 'utf-8', 'stdout': -1, 'stderr': -2, 'shell': False, 'stdin': None})
2023-01-11 17:15:06,729.729 INFO: topolog.r1: vtysh result:
OS                             Linux(4.15.0-193-generic)

Notice the missing pimreg11 device needed in vrf blue:

2023-01-11 17:15:06,731.731 DEBUG: topolog.r1: LinuxNamespace(r1): cmd_status("['/bin/bash', '-c', 'vtysh  -c "show int brief" 2>/dev/null']", kwargs: {'encoding': 'utf-8', 'stdout': -1, 'stderr': -2, 'shell': False, 'stdin': None})
2023-01-11 17:15:06,781.781 INFO: topolog.r1: vtysh result:
Interface       Status  VRF             Addresses
---------       ------  ---             ---------
blue            up      blue            192.168.0.1/32
r1-eth0         up      blue            192.168.100.1/24
r1-eth1         up      blue            192.168.101.1/24

Interface       Status  VRF             Addresses
---------       ------  ---             ---------
erspan0         down    default
gre0            down    default
gretap0         down    default
lo              up      default
pimreg          up      default

Interface       Status  VRF             Addresses
---------       ------  ---             ---------
r1-eth2         up      red             192.168.100.1/24
r1-eth3         up      red             192.168.101.1/24
red             up      red             192.168.0.1/32

While on a 5.4 machine we have this:

mininet310# show int brief
Interface       Status  VRF             Addresses
---------       ------  ---             ---------
blue            up      blue
dummy1          up      blue
dummy2          up      blue
pimreg11        up      blue

As such let's limit the test to a 4.19 kernel or above that our
documentations states we need for proper pim operation.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>

Merge pull request #12628 from donaldsharp/underlines_too_short_in_doc

doc: Ensure that the underline length is correct

doc: Ensure that the underline length is correct

Builds are complaining about this issue.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>

bgpd: Don't warning twice for neighbor_bfd_check_controlplane_failure

When the peer is not defined first.

Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>

bgpd: Warn an operator, that the peer MUST be created before configuring rest

For now, if the order was mixed, most of the commands are just silently
ignored. Let the operator notice that.

Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>

Merge pull request #12622 from taspelund/adj-rib-json

bgpd: add "detail" for advertised/received-routes

Merge pull request #12577 from louis-6wind/fix-bgp-evpn-all

bgpd: fix show bgp all with evpn

zebra: cosmetic changes for debug

Just remove redundant white spaces in debug information.

Before:
```
2023/01/11 05:04:48 ZEBRA: [W8V7C-6W4DS] init neigh ctx NEIGH_INSTALL: ifp vlan100, mac  9a:68:e9:73:74:88, ip 88.88.88.88
2023/01/11 05:04:48 ZEBRA: [NH6N7-54CD1] Tx RTM_NEWNEIGH family ipv4 IF vlan100(8) Neigh 88.88.88.88 MAC  9a:68:e9:73:74:88 flags 0x10 state 0x40 ext_flags 0x0
```

After:
```
2023/01/11 05:17:26 ZEBRA: [W8V7C-6W4DS] init neigh ctx NEIGH_INSTALL: ifp vlan100, mac 9a:68:e9:73:74:88, ip 88.88.88.88
2023/01/11 05:17:26 ZEBRA: [NH6N7-54CD1] Tx RTM_NEWNEIGH family ipv4 IF vlan100(8) Neigh 88.88.88.88 MAC 9a:68:e9:73:74:88 flags 0x10 state 0x40 ext_flags 0x0
```

Signed-off-by: anlan_cs <vic.lan@pica8.com>

Merge pull request #12578 from Pdoijode/evpn-mac-vni-det

zebra: Evpn mac vni detail show command

Merge pull request #12521 from patrasar/issue_11347_fix

pimd, pim6d: Handling SGRPT prune received

Merge pull request #12590 from anlancs/fix/bgpd-remove-ec-log

bgpd: remove one EC log

Merge pull request #12602 from opensourcerouting/fix/drop_deprecated_BGP_ATTR_AS_PATHLIMIT

bgpd: Drop deprecated BGP_ATTR_AS_PATHLIMIT path attribute

Merge pull request #12620 from donaldsharp/continue_reading

zebra: Continue fpm_read when we decide a netlink message is not needed

doc: add "detail" advertised/received routes

Signed-off-by: Trey Aspelund <taspelund@nvidia.com>

bgpd: add "detail" for advertised/received-routes

Introduce a "detail" keyword for per-neighbor/per-afi-safi
advertised-routes and received-routes show commands.
Includes json support.

Signed-off-by: Trey Aspelund <taspelund@nvidia.com>

Merge pull request #12612 from opensourcerouting/fix/use_quay.io_for_github_actiobns

github: Use quay.io as docker registry for github action generated images

zebra: Continue fpm_read when we decide a netlink message is not needed

When FRR receives a netlink message that it decides to stop parsing
it returns a 0 ( instead of a -1 ).  Just make the dplane continue
reading other data instead of aborting the read.

Signed-off-by: Donald Sharp <sharpd@nvidia.com>

bfdd: validate incoming control message length

Apply upper bound on incoming control messages, drop invalid
messages.

Signed-off-by: Mark Stapp <mjs@labn.net>

Merge pull request #12567 from YutaroHayakawa/YutaroHayakawa/isl-dev

docker,alpine: Remove isl from dependencies

Merge pull request #12614 from chiragshah6/zdev

zebra: Add missing json attributes for show evpn

zebra: Add missing json attributes for show evpn

Missing json attributes added for show evpn json

Before:
```
tor-1# show evpn json
{
  "advertiseGatewayMacip":"No",
  "numVnis":26,
  "numL2Vnis":21,
  "numL3Vnis":5,
  "isDuplicateAddrDetection":true,
  "maxMoves":5,
  "detectionTime":180,
  "detectionFreezeTime":0,
  "macHoldtime":1080,
  "neighHoldtime":1080,
  "startupDelay":180,
  "startupDelayTimer":"--:--:--",
  "uplinkConfigCount":0,
  "uplinkActiveCount":0
}
tor-1#
```

After:
```
tor-1# show evpn json
{
  "advertiseGatewayMacip":"No",
  "advertiseSviMacip":"No",
  "advertiseSviMac":"No",
  "numVnis":26,
  "numL2Vnis":21,
  "numL3Vnis":5,
  "isDuplicateAddrDetection":true,
  "maxMoves":5,
  "detectionTime":180,
  "detectionFreezeTime":0,
  "macHoldtime":1080,
  "neighHoldtime":1080,
  "startupDelay":180,
  "startupDelayTimer":"--:--:--",
  "uplinkConfigCount":0,
  "uplinkActiveCount":0
}
tor-1#
```

Ticket:#3323248

Issue:3323248

Testing: UT done

Signed-off-by: Sindhu Parvathi Gopinathan's <sgopinathan@nvidia.com>

github: Use quay.io as docker registry for github action generated images

Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>

Merge pull request #12606 from opensourcerouting/fixes-20230106

*: various build fixes

Merge pull request #12573 from Pdoijode/bgp-nexthop-json-changes

Bgp nexthop json changes

Merge pull request #12586 from LabNConsulting/chopps/json-cli-note

doc: add guidance for CLI cmds that output JSON

Merge pull request #12605 from Jafaral/lsa2

ospfd: report the router IP with opaque capability mismatch

build: add doc files missing from dist tarball

The override.css/js files for sphinx docs were not being included into
the tarball created by `make dist`.

Signed-off-by: David Lamparter <equinox@opensourcerouting.org>

lib/clippy: fix symbol validity check

If symvalid is false, looking at symidx is bogus.

This fixes a build-time SEGV on mips64el.

Signed-off-by: David Lamparter <equinox@opensourcerouting.org>

lib: disable xref ELF note on mips64el

mips64el does not have a 64-bit PC-relative relocation, which is needed
to emit the ELF note for xrefs.  Disabling the ELF note means clippy
takes the fallback path using section headers, so everything does still
work (... unless you strip the section headers.)

Signed-off-by: David Lamparter <equinox@opensourcerouting.org>

zebra: do not load/store wider-than-ptr atomics

Most 32-bit architectures cannot do atomic loads and stores of data
wider than their pointer size, i.e. 32 bit.  Funnily enough they
generally *can* do a CAS2, i.e., 64-bit compare-and-swap, but while a
CAS can emulate atomic add/bitops, loads and stores aren't available.

Replace with a mutex;  since this is 99% used from the zserv thread, the
mutex should take the local-to-thread fast path anyway.  And while one
atomic might be faster than a mutex lock/unlock, we're doing several
here, and at some point a mutex wins on speed anyway.

This fixes build on armel, mipsel, m68k, powerpc, and sh4.

Signed-off-by: David Lamparter <equinox@opensourcerouting.org>

ospfd: report the router IP with Opaque capability mismatch

Signed-off-by: Jafar Al-Gharaibeh <jafar@atcorp.com>

docker,alpine: Remove isl from dependencies

Alpine upstream changed the name of the isl package to isl-dev. This
caused the build breakage. Since FRR doesn't use it, we chose to solve
this issue by removing it.

Signed-off-by: Yutaro Hayakawa <yutaro.hayakawa@isovalent.com>

Merge pull request #12568 from YutaroHayakawa/YutaroHayakawa/fpm-nexthop

fpm: Send NH message to FPM even if the local kernel doesn't support it

Merge pull request #12572 from ak503/lib

lib: fix display cputime-warning and walltime-warning

Merge pull request #12596 from opensourcerouting/fix/frrtrace_arguments

bgpd: Fix the number of arguments for frrtrace()

bgpd: Drop deprecated BGP_ATTR_AS_PATHLIMIT path attribute

Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>

doc: add guidance for CLI cmds that output JSON

Signed-off-by: Christian Hopps <chopps@labn.net>

bgpd: Fix the number of arguments for frrtrace()

For bgp_path_info_add_with_caller().

Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>

bgpd: AFI option to query nexthops based on AFI

Added ipv4 and ipv6 option to existing "show bgp nexthop"
command to be able to query nexthops that belong to a
particular address-family.

Also fixed the warnings of MR 12171

Signed-off-by: Pooja Jagadeesh Doijode <pdoijode@nvidia.com>

Merge pull request #12458 from liron-ze/aggre-route-withdraw

bgpd: Fix aggregated routes are withdrawn abnormally.

Merge pull request #12571 from Shikugawa/ospfd-json

ospfd: show ospf database info using formatted json

Merge pull request #12589 from mjstapp/fix_zeb_typos

zebra: fix a couple of typos

Merge pull request #12581 from cscarpitta/fix/fix-wrong-interface-name-srv6l3vpn-topotest

tests: Fix wrong interface name in `bgp_srv6l3vpn_route_leak` topotest

bgpd: remove one EC log

When creating one interface "vxlan66" ( ip link add vxlan66 type vxlan ... ),
which initially maintains down status, saw one unexpected EC log:

```
zebra[37906]: [ZAG0W-VSNSD] interface vxlan66 vrf default(0) index 35 becomes active.
zebra[37906]: [HPWGA-Y527W] IFLA_VXLAN_LINK missing from VXLAN IF message
...
zebra[37906]: [W6BZR-YZPAB] RTM_NEWLINK update for vxlan66(35) sl_type 0 master 0 flags 0x1002
zebra[37906]: [MR3ZF-ATDBY] Intf vxlan66(35) has gone DOWN
zebra[37906]: [T44X9-FFNVB] Intf vxlan66(35) L2-VNI 66 is DOWN
zebra[37906]: [KEGYY-K8XVV] Send EVPN_DEL 66 to bgp
zebra[37906]: [HPWGA-Y527W] IFLA_VXLAN_LINK missing from VXLAN IF message
bgpd[37911]: [TV0XP-3WR0A] Rx VNI del VRF default VNI 66 tenant-vrf default SVI ifindex 0
bgpd[37911]: [MDW89-YAXJG][EC 33554497] 0: VNI hash entry for VNI 66 not found at DEL
```

Since commit `6f908ded80eeba40a850e8d1f6246fb3ed31e648` support interfaces
from down to down, and bgpd doesn't know "VNI 66" at all. So, remove this
EC log.

Signed-off-by: anlan_cs <vic.lan@pica8.com>

zebra: Evpn mac vni detail show command

New show command "show evpn mac vni xx detail [json]"
to display details of all the mac entries for the
requested VNI.

Output of show evpn mac vni xx detail json:
        {
          "numMacs":2,
          "macs":{
            "ca:be:63:7c:81:05":{
              "type":"local",
              "intf":"veth100",
              "ifindex":8,
              "uptime":"00:06:55",
              "localSequence":0,
              "remoteSequence":0,
              "detectionCount":0,
              "isDuplicate":false,
              "syncNeighCount":0,
              "neighbors":{
                "active":[
                  "fe80::c8be:63ff:fe7c:8105"
                ],
                "inactive":[
                ]
              }
            }
          }
        }

Also added remoteEs field in the JSON output of
"show evpn mac vni xx json".

Output of show evpn mac vni xx json:
"00:02:00:00:00:0d":{
  "type":"remote",
  "remoteEs":"03:44:38:39:ff:ff:02:00:00:02",
  "localSequence":0,
  "remoteSequence":0,
  "detectionCount":0,
  "isDuplicate":false
}

Signed-off-by: Pooja Jagadeesh Doijode <pdoijode@nvidia.com>

zebra: fix a couple of typos

Fix a couple of typos in vty prompt and output text.

Signed-off-by: Mark Stapp <mjs@labn.net>

Merge pull request #12318 from gpnaveen/bgp_unique_rid

tests: adding bgp unique router id automation.

ospfd: show ospf database info using formatted json

Signed-off-by: Rei Shimizu <rshimizu@apache.org>

Merge pull request #12566 from Jafaral/lsa-op

ospfd: add advertising router IP to Opaque capability mismatch err msg

tests: Fix wrong interface name in SRv6 topotest

Previously, routes leaked from one VRF to another VRF were associated
with the original nexthop interface.

Commit 14aabc01565a918c223b7811572fae0316810422 replaced the nexthop
interface with the index of incoming VRF interface.

Due to this change, the `bgp_srv6l3vpn_route_leak` topotest always fails
because it still expects the nexthop interface.

This commit fixes the expected interface name in the
`bgp_srv6l3vpn_route_leak` topotest.

Signed-off-by: Carmine Scarpitta <carmine.scarpitta@uniroma2.it>

bgpd: fix show bgp all with evpn

Fix crash on "show bgp all" when BGP EVPN is set.

> #0  raise (sig=11) at ../sysdeps/unix/sysv/linux/raise.c:50
> #1  0x00007fdfe03cf53c in core_handler (signo=11, siginfo=0x7ffdebbffe30, context=0x7ffdebbffd00) at lib/sigevent.c:261
> #2  <signal handler called>
> #3  0x00000000004d4fec in bgp_attr_get_community (attr=0x41) at bgpd/bgp_attr.h:553
> #4  0x00000000004eee84 in bgp_show_table (vty=0x1a790d0, bgp=0x19d0a00, safi=SAFI_EVPN, table=0x19f6010, type=bgp_show_type_normal, output_arg=0x0, rd=0x0, is_last=1, output_cum=0x0,
>     total_cum=0x0, json_header_depth=0x7ffdebc00bf8, show_flags=4, rpki_target_state=RPKI_NOT_BEING_USED) at bgpd/bgp_route.c:11329
> #5  0x00000000004f7765 in bgp_show (vty=0x1a790d0, bgp=0x19d0a00, afi=AFI_L2VPN, safi=SAFI_EVPN, type=bgp_show_type_normal, output_arg=0x0, show_flags=4,
>     rpki_target_state=RPKI_NOT_BEING_USED) at bgpd/bgp_route.c:11814
> #6  0x00000000004fb53b in show_ip_bgp_magic (self=0x6752b0 <show_ip_bgp_cmd>, vty=0x1a790d0, argc=3, argv=0x19cb050, viewvrfname=0x0, all=0x1395390 "all", aa_nn=0x0, community_list=0,
>     community_list_str=0x0, community_list_name=0x0, as_path_filter_name=0x0, prefix_list=0x0, accesslist_name=0x0, rmap_name=0x0, version=0, version_str=0x0, alias_name=0x0,
>     orr_group_name=0x0, detail_routes=0x0, uj=0x0, detail_json=0x0, wide=0x0) at bgpd/bgp_route.c:13040
> #7  0x00000000004fa322 in show_ip_bgp (self=0x6752b0 <show_ip_bgp_cmd>, vty=0x1a790d0, argc=3, argv=0x19cb050) at ./bgpd/bgp_route_clippy.c:519
> #8  0x00007fdfe033ccc8 in cmd_execute_command_real (vline=0x19c9300, filter=FILTER_RELAXED, vty=0x1a790d0, cmd=0x0, up_level=0) at lib/command.c:996
> #9  0x00007fdfe033c739 in cmd_execute_command (vline=0x19c9300, vty=0x1a790d0, cmd=0x0, vtysh=0) at lib/command.c:1056
> #10 0x00007fdfe033cdf5 in cmd_execute (vty=0x1a790d0, cmd=0x19c9eb0 "show bgp all", matched=0x0, vtysh=0) at lib/command.c:1223
> #11 0x00007fdfe03f65c6 in vty_command (vty=0x1a790d0, buf=0x19c9eb0 "show bgp all") at lib/vty.c:486
> #12 0x00007fdfe03f603b in vty_execute (vty=0x1a790d0) at lib/vty.c:1249
> #13 0x00007fdfe03f533b in vtysh_read (thread=0x7ffdebc03838) at lib/vty.c:2148
> #14 0x00007fdfe03e815d in thread_call (thread=0x7ffdebc03838) at lib/thread.c:2006
> #15 0x00007fdfe0379b54 in frr_run (master=0x1246880) at lib/libfrr.c:1198
> #16 0x000000000042b2a8 in main (argc=7, argv=0x7ffdebc03af8) at bgpd/bgp_main.c:520

Link: https://github.com/FRRouting/frr/issues/12576
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>

bgpd: Detail option for nexthop and import-check to display paths

1. Updated "show bgp vrf <vrf-name> nexthop detail"
   and "show bgp vrf <vrf-name> import-check-table
   detail" show commands to display paths associated with
   nexthop. "detail" option was previously unused.
2. Added 'ipv4' and 'ipv6' JSON object under top level JSON.
3. Removed the "nexthops" JSON object which was under the top
   level JSON object
4. Renamed "ifname" to "interfaceName"
5. Renamed "gates" JSON obejct to "nexthops"
6. Changed "flags" JSON array to JSON object and changed the
   flags from string to boolean
7. "lastUpdate" will display only epoch time for "detail" option

JSON output:
r4# show bgp vrf default nexthop detail json
{
  "ipv4":{
    "10.0.7.1":{
      "valid":true,
      "complete":true,
      "igpMetric":0,
      "pathCount":3,
      "peer":"10.0.7.1",
      "nexthops":[
        {
          "interfaceName":"r4-r2-eth0"
        }
      ],
      "lastUpdate":1672265350,
      "paths":[
        {
          "afi":"IPv4",
          "safi":"unicast",
          "prefix":"11.0.20.2/32",
          "vrf":"default",
          "flags":{
            "igpChanged":false,
            "damped":false,
            "history":false,
            "bestpath":true,
            "valid":true,
            "attrChanged":false,
            "deterministicMedCheck":false,
            "deterministicMedSelected":false,
            "stale":false,
            "removed":false,
            "counted":true,
            "multipath":false,
            "multipathChanged":false,
            "ribAttributeChanged":false,
            "nexthopSelf":false,
            "linkBandwidthChanged":false,
            "acceptOwn":false
          }
        }
      ]
    }
  }
}
}

Signed-off-by: Pooja Jagadeesh Doijode <pdoijode@nvidia.com>

bgpd: add json option to show commands in bgp_nexthop

Commands with json option:
 - show bgp nexthop
 - show bgp import-check-table

Example output below, "nexthop" and "import-check-table" are only
different in the nexthop entries, the format is the same
```
leaf-A# show bgp nexthop 10.11.10.1 detail json
{
  "nexthops":{
    "10.11.10.1":{
      "valid":true,
      "complete":true,
      "igpMetric":0,
      "pathCount":1,
      "peer":"10.11.10.1",
      "gates":[
        {
          "ifname":"eth1"
        }
      ],
      "lastUpdate":{
        "epoch":1669161758,
        "string":"Wed Nov 23 00:02:38 2022\n"
      },
      "paths":[
        {
          "afi":"IPv4",
          "safi":"unicast",
          "prefix":"192.168.11.0/24",
          "vrf":"default",
          "flags":[
            "valid",
            "dmedSelected",
            "counted"
          ]
        }
      ]
    }
  }
}
leaf-A# show bgp nexthop json
{
  "nexthops":{
    "10.10.10.1":{
      "valid":true,
      "complete":true,
      "igpMetric":0,
      "pathCount":1,
      "peer":"10.10.10.1",
      "gates":[
        {
          "ifname":"eth0"
        }
      ],
      "lastUpdate":{
        "epoch":1669161758,
        "string":"Wed Nov 23 00:02:38 2022\n"
      }
    },
    "10.11.10.1":{
      "valid":true,
      "complete":true,
      "igpMetric":0,
      "pathCount":1,
      "peer":"10.11.10.1",
      "gates":[
        {
          "ifname":"eth1"
        }
      ],
      "lastUpdate":{
        "epoch":1669161758,
        "string":"Wed Nov 23 00:02:38 2022\n"
      }
    }
  }
}

```

Signed-off-by: Yaroslav Fedoriachenko <yar.fed99@gmail.com>

lib: fix display cputime-warning and walltime-warning

Before patch:
```
# vtysh -c "c t" -c "service cputime-warning 1" -c "service walltime-warning 2"
# vtysh  -c "show run" | grep warn
service cputime-warning 1000
service walltime-warning 2000
```

After patch:
```
# vtysh -c "c t" -c "service cputime-warning 1" -c "service walltime-warning 2"
# vtysh  -c "show run" | grep warn
service cputime-warning 1
service walltime-warning 2
```

Signed-off-by: Dmitrii Turlupov <dturlupov@factor-ts.ru>

Merge pull request #12562 from opensourcerouting/fix/add_frrtrace_points_for_peer_lock_unlock

bgpd: A bit more tracepoints for lttng

Merge pull request #12558 from donaldsharp/bgp_static_route_mem_leak

bgpd: static routes are leaked on shutdown

Merge pull request #12515 from opensourcerouting/fix/show_ip_bgp_detail_weird_output

bgpd: Show the real prefix for `show bgp detail`

Merge pull request #11127 from louis-6wind/bgp-leak

bgpd: multiple fixes for route leaking