lib: Convert vrf.c to use new error-code subsystem

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

lib: Convert sigevent.c to use new error-code-subsystem

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

lib: Convert vty.c to use new error-code subsystem

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

lib: Convert routemap.c to use new error-code subsystem

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

lib: Convert privs.c to use new error-code subsystem

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

lib: Convert network.c to use new error subsystem codes

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

lib: Convert thread.c to use new error-code subsystem

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

lib: Convert netns_linux.c to new error-code work

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

lib: Convert libfrr.c to use new error code system

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

lib: Convert stream.c to use new error code subsystem

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

lib: Convert sockopt.c to flog_err

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

lib: Convert sockunion.c to use flog_err

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

lib: Update zclient.c warn code

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

lib: Convert to errors some issues in buffer.c

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

lib: Convert to flog_warn in agentx.c

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

bgpd: Update bgp_updgrp_packet.c to use flog_warn

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

bgpd: Convert bgp_open.c to use flog_warn

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

bgpd: convert bgp_nht.c to use flog_warn

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

bgpd: Modify bgp_label to use flog_warn

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

bgpd: Convert bgp_evpn.c to use flow_warn

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

bgpd: Convert bgp_network.c to use flog_warn

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

bgpd: Convert zlog_warn to flog_warn for bgp_packet.c

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

bgpd: Convert bgp_dump.c to use flog_warn

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

bgpd: Convert zlog_warn to flog_warn for bgp_aspath.c and bgp_attr.c

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

lib: add flog_warn

Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>

Merge pull request #2984 from donaldsharp/ak503_eigrp

eigrp: opnional metric in 'no redistribute'

Merge pull request #2974 from donaldsharp/v4_and_v6

V4 and v6

eigrp: opnional metric in 'no redistribute'

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

doc: Update sharp documentation

Update the sharp documentation to note v6 nexthops are able
to be installed now.

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

sharpd: Modify route install to allow v6 nexthops

Allow the user to create a v4 route with a v6 nexthop.

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

zebra: Fix _route_entry_dump to handle nexthop family as appropriate

The _route_entry_dump function was not handling the nexthop as passed
in from an upper level protocol appropriate and as such not displaying
the v4/v6 nexthop right in the case where we have both going.

Additionally dump the nexthop vrf as well.

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

Merge pull request #2959 from opensourcerouting/bfd-cleanup

bfdd: code clean up

Merge pull request #2873 from vivek-cumulus/evpn-extended-mobility

EVPN extended mobility support

bfdd: fix crash when receiving invalid echo packet

The log function would improperly format a string using an integer
causing a crash. This situation was found when switching echo mode with
a active connection.

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

bfdd: use THREAD_OFF() instead of thread_cancel()

Use a standardized function to cancel threads instead of hand rolling
our own.

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

Merge pull request #2970 from dslicenc/more-bool-use-json

bgpd, ospfd, pimd, zebra: Convert more use_json locations to bool

Merge pull request #2971 from qlyoung/misc-doc-updates

doc: fix building.rst index

doc: fix building.rst index

Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>

bgpd, ospfd, pimd, zebra: Convert more use_json locations to bool

Signed-off-by: Don Slice <dslice@cumulusnetworks.com>

Merge pull request #2909 from netravnen/feature/git-pl-template

Add github pull-request-template.md

Merge pull request #2934 from LabNConsulting/working/master/rfapi_topotest_support

add some testing support for use in rfapi topotest

Merge pull request #2939 from netravnen/feature/git-issue-teplate

Add ISSUE_TEMPLATE

Merge pull request #2966 from donaldsharp/v6_cleanup

lib: Remove dead code

lib: Remove dead code

The ZEBRA_IPV4_ROUTE_IPV6_NEXTHOP_ADD zapi message has no creators and
no handlers.  Let's just remove.

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

bfdd: use UDP sockets instead of raw

Avoid all complexities of handling ethernet/IP/UDP headers and just use
UDP sockets: let the OS kernel handle this task.

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

bfdd: enumerate all diagnositic codes

Enumerate all codes, update the string representation and fix the
misdiagnosed echo failure.

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

bfdd: make bfd_recv_* more generic

We'll reuse these functions later to receive echo packets, so we need
to despecialize them first. Another improvement here is that we are
going to remove the global buffer variable and opt to use the stack.

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

bfdd: remove VxLAN code

Remove all VxLAN code. There are probably better ways to write VxLAN
support without having to use raw sockets and depend on extra user
configuration input.

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

bfdd: remove global variables and improve logs

Remove some more global variables and make the log messages look like
the rest of the code.

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

Merge pull request #2950 from qlyoung/misc-doc-updates

Misc doc updates

Merge pull request #2945 from dslicenc/bgp-ospf-json

bgpd/ospfd: make bgp and ospf json response a bit more consistent

.github: Update ISSUE_TEMPLATE.md

Addressed @rwestphal comments.

Signed-off-by: Christoffer <netravnen@gmail.com>

.github: Updated PULL_REQUEST_TEMPLATE.md

Addressed @qlyoung's suggestions.

Signed-off-by: Christoffer <netravnen@gmail.com>

.github: Updated ISSUE_TEMPALTE.md

Addressed @qlyoung's suggestions.

Signed-off-by: Christoffer <netravnen@gmail.com>

zebra: remove invalid ref "frr-configuration"

There is, in fact, no such additional information

Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>

doc: update subdir.am

Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>

doc: update misc build docs

Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>

doc: name correct sphinx package under FreeBSD 11

Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>

doc: rename building_frr_on* -> building_frr_for*

Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>

bgpd: convert from conditional operator to if-then for json changes

Signed-off-by: Don Slice <dslice@cumulusnetworks.com>

Merge pull request #2913 from opensourcerouting/non-recursive-prep

pre-final non-recursive make

Merge pull request #2955 from ton31337/feature/nexthop_blackhole

plist: Add safety check to avoid going over 4294967295 for prefix-list

Merge pull request #2940 from ton31337/fix/as-path_validation

bgpd: Validate as-path access-list before continuing regarding invali…

plist: Add safety check to avoid going over 4294967295 for prefix-list

bgpd: change conditional operators for bgp/ospf json changes

Signed-off-by: Don Slice <dslice@cumulusnetworks.com>

bgpd: Validate as-path access-list before continuing regarding invalid chars

Signed-off-by: Donatas Abraitis donatas.abraitis@gmail.com

.github: Updated ISSUE_TEMPALTE.md

Addressed all of @qlyoung's "presentational nits
and a formatting suggestion". As well as
@rwestphal's idea of OS and FRR name+version
numbers.

Signed-off-by: Christoffer <netravnen@gmail.com>

Merge pull request #2785 from donaldsharp/AGGanomics

Abstract `void *aggregate` pointer in `struct route_node`

Merge pull request #2951 from qlyoung/sort-route-maps

lib: sort route-maps for display

static: Put vty_frame around vrf output in staticd.

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

staticd: Fix mixup in vrf translations

When we store the nexthop for ref-counting, keep
track of the nexthop vrf_id as well.  This will allow
us to track the nexthop per vrf!

Additionally when we get the callback from zebra about
a nexthop update, iterate over all static routes to
see if the nexthop we are getting a callback is
one we are concerned about.

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

vtysh: Let staticd know about vrf sub-mode

staticd was not transitioning into vrf sub-mode, this fixes it.

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

Merge pull request #2923 from pguibert6WIND/fix_veth_namespace

zebra: when veth link is used across vrf, the link may not be good

lib: Move aggregate pointer into aggregate route/table

Move the aggregate pointer from the route_node into agg_node
so that people using struct route_node will see a savings
in data size.

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

bgpd, ripngd: Convert to using new agg_table/route

Switch bgp and ripngd to use the new aggregate table and
route data structures.  This was mainly a search and replace
operation.

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

lib: Add Aggregate Table and Aggregate_node

Add a abstraction for `struct route_node` and `struct route_table`
such that we can have an aggregate route_node and table.  This
is because only bgp/rfapi and ripng use the aggregate data pointer
in `struct route_node`.  For full route tables other routing
protocols and tables are paying a 8 byte overhead per node.
A full bgp table ends up being ~1.2 million routes in bgp
and zebra.  This is not an insiginificant amount of data.

So create the data structures for this replacement, but
do not replace the aggregate pointer yet.  This is because
later commits will convert rfapi and ripng over to this
new data, and finally we'll move the aggregate pointer.

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

lib: sort route-maps for display

Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>

ospfd: fix more warnings with the bgp ospf json commit

Signed-off-by: Don Slice <dslice@cumulusnetworks.com>

Merge pull request #2947 from mjstapp/fix_os_name

lib: fix use of frrpthread os_name

configure: Update version to 6.1-dev

We are now working on 6.1-dev for new work.

Signed-off-by: Martin Winter <mwinter@opensourcerouting.org>

bgpd/ospfd: resolve warnings for bgp/ospf json commit

Signed-off-by: Don Slice <dslice@cumulusnetwork.com>

lib: fix use of frrpthread os_name

Fix test for presence of configured os pthread name.

Signed-off-by: Mark Stapp <mjs@voltanet.io>

Merge pull request #2859 from LabNConsulting/working/master/meminfo

lib: qmem show changes (header and max)

bgpd/ospfd: make bgp and ospf json response a bit more consistent

Problem reported that some bgp and ospf json commands did not return
any json output at all if the bgp/ospf instance did not exist.
Additionally, some bgp and ospf json commands did not return any json
output if the instance existed but no neighbors were defined.  This
fix makes these commands more consistent in returning empty braces for
json output and issue a message if not using json output.  Additionally,
made the flag "use_json" a bool to make it consistent since previously,
it had been defined as an int, char, u_char, and bool at various places.

Ticket: CM-21040
Signed-off-by: Don Slice <dslice@cumulusnetworks.com>

doc: drop colspan/rowspan elements

The sphinx 'text' writer apparently doesn't support these, and the
'dummy' writer is only available on recent versions, and all of this
makes the build a bit annoying...

Signed-off-by: David Lamparter <equinox@diac24.net>

doc: html default off

... RPM builds break if we enable this unconditionally; should really
change this in CI instead.

Signed-off-by: David Lamparter <equinox@diac24.net>

build: add Makefile variable easy access

This adds 2 helper targets for use in scripts to easily get at Makefile
variables without parsing the Makefile or config.status.

Signed-off-by: David Lamparter <equinox@diac24.net>

build: improve python search pattern

- try pythonN.N-config after pythonN-config
- use "python-config --ldflags" instead of --libs
- add Python 3.6 to explicitly searched versions
- if linking fails, try with "-lz" added

Signed-off-by: David Lamparter <equinox@diac24.net>

build: add warning when overwriting vars on "make"

doing things like `make CC="mmix-linux-musl-gcc"` breaks the hosttools/
cross-compilation setup pretty hard and just straight up should not be
done.  These vars belong on `configure`, not `make`.

Signed-off-by: David Lamparter <equinox@diac24.net>

build: cache a few more results in ./configure

Makes configure re-runs faster if the `-C` option is used.

Signed-off-by: David Lamparter <equinox@diac24.net>

build: -Werror does not belong in AM_CPPFLAGS

(probably neither do the ASAN/... FLAGS, but let's leave that alone.)

Also, it was intentional that it's $(WERROR) and not @WERROR@.  The
former can be overwritten at comandline as "make WERROR=" while the
latter can't.

Signed-off-by: David Lamparter <equinox@diac24.net>

build: non-recursive solaris

Signed-off-by: David Lamparter <equinox@diac24.net>

build: fix some oddities in solaris/

target_* is always wrong for FRR because we are not a compiler.  We only
use host_* and build_*.

Signed-off-by: David Lamparter <equinox@diac24.net>

build: trigger fewer unneccessary clippy rebuilds

Make sure we include the clippy binary file in our timestamp mangling so
`make` will correctly determine when clippy files don't need to be
rebuilt.

Signed-off-by: David Lamparter <equinox@diac24.net>

build: non-recursive debianpkg

Signed-off-by: David Lamparter <equinox@diac24.net>

build: non-recursive doc + parallel sphinx

Sphinx actually does work with a parallel build, if the doctree creation
is a separate step (which the other builds will then just read
unmodified.)  This can be done with the "dummy" target.

This also adds "-j6" to sphinx-build and adds a "--disable-doc-html"
switch on ./configure to turn on/off building HTML docs separately.

Also, HTML docs are now installed by "make install" to
/usr/share/doc/frr/html.

Signed-off-by: David Lamparter <equinox@diac24.net>

Merge pull request #2942 from mjstapp/fix_new_atomics

libs: fix missing atomics for some gcc versions

Merge pull request #2754 from chiragshah6/mdev1

*: pthread set name abstraction

*: pthread set name abstraction

Testing Done:

 TOR#cat /proc/2670/task/2672/comm
 bgpd_ka

TOR# ps H -C bgpd -o 'pid tid cmd comm'
  PID   TID CMD                         COMMAND
  2670  2670 /usr/lib/frr/bgpd -M snmp - bgpd
  2670  2671 /usr/lib/frr/bgpd -M snmp - bgpd
  2670  2672 /usr/lib/frr/bgpd -M snmp - bgpd_ka

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

libs: fix missing atomics for some gcc versions

stdatomic.h does not have aliases for all of the useful gcc
atomic primitives; add them in for that path through
frratomic.h.

Signed-off-by: Mark Stapp <mjs@voltanet.io>