Donald Sharp [Tue, 10 Dec 2019 19:06:33 +0000 (14:06 -0500)]
zebra: Do not build mlag protobuf support if version 3 is not avail
Older versions of protobuf-c do not support version 3 of the
protocol. Add a check into the system to see if we have
version 3 available and if so, compile it in.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Donald Sharp [Tue, 10 Dec 2019 01:48:21 +0000 (20:48 -0500)]
zebra: Allow zebra_mlag to compile with no j factor
If you compile FRR with no j factor zebra_mlag.c fails to
build because the vtysh extraction methodology runs first
before the protobuf compiler runs and that compilation does
not have the proper dependancy chain built for the inclusions
that zebra_mlag.c had. Moving the DEF* code into a zebra_mlag_vty.c
which can be included in the vtysh extraction code and has
no mlag.proto dependancies makes the compilation work better.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
David Lamparter [Wed, 27 Nov 2019 21:52:50 +0000 (22:52 +0100)]
lib/linklist: flip the bitrot compost
The whole lib/linklist.c code shouldn't really be used for new code (the
lib/typesafe.h bits are better.) So, a new need for these unused
functions shouldn't be coming up.
Signed-off-by: David Lamparter <equinox@diac24.net>
David Lamparter [Wed, 13 Nov 2019 23:21:10 +0000 (00:21 +0100)]
lib: completely get rid of the MTYPE alias hack
Sometimes the easiest solution is hardest to find... the whole point of
all this "static const", aliasing, & co. was to make "MTYPE_FOO" usable
without adding the extra & as in "&MTYPE_FOO". Making it a size-1 array
does that perfectly through the magic of ISO C array decay...
Signed-off-by: David Lamparter <equinox@diac24.net>
Donald Sharp [Fri, 13 Dec 2019 00:30:21 +0000 (19:30 -0500)]
isisd: Free memory when confused
When you call into lsp_update with confusion, the lsp is purged
and we do not do anything with the created tlv's from parsing
the incoming data. To prevent the tlv's from being leaked
note confusion and delete the unneeded data.
Fixes: #5496 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
David Lamparter [Wed, 11 Dec 2019 12:33:36 +0000 (13:33 +0100)]
lib,nhrpd,bgpd/bmp: pass resolver failure details
To keep the calling code agnostic of the DNS resolver libary used, pass
a strerror-style string instead of a status code that would need extra
handling.
Signed-off-by: David Lamparter <equinox@diac24.net>
David Lamparter [Wed, 11 Dec 2019 11:27:05 +0000 (12:27 +0100)]
lib/resolver: support/bypass IP literals
libc-ares doesn't do IP literals, so we have to do that before running
off to do DNS. Since this isn't BMP specific, move to lib/ so NHRP can
benefit too.
Signed-off-by: David Lamparter <equinox@diac24.net>
Under some circumstances (apparently depends on several optimization
flags), gcc-9 throws an used-uninitialized warning for this variable in
the skiplist code. Just initialize to NULL.
Signed-off-by: David Lamparter <equinox@diac24.net>
Mark Stapp [Mon, 9 Dec 2019 21:02:57 +0000 (16:02 -0500)]
lib,zebra: add api to enforce nexthop sort order when copying
Add an api that creates a copy of a list of nexthops and
enforces the canonical sort ordering; consolidate some nhg
code to avoid copy-and-paste. The zebra dplane uses
that api when a plugin sets up a list of nexthops, ensuring
that the plugin's list is ordered when it's processed in
zebra.
Mark Stapp [Mon, 2 Dec 2019 17:03:57 +0000 (12:03 -0500)]
zebra: align dplane notify processing with nhg work
The processing of dataplane route notifications was a little
off-target after the nexthop-group re-work. This should allow
notifications to work better.
Donald Sharp [Fri, 6 Dec 2019 13:58:47 +0000 (08:58 -0500)]
lib, zebra: Allow for installation of a weighted nexthop
Linux has the idea of allowing a weight to be sent
down as part of a nexthop group to allow the kernel
to weight particular nexthop paths a bit more or less
than others.
Allow for installation into the kernel using the weight attribute
associated with the nexthop.
This code is foundational in that it just sets up the ability
to do this, we do not use it yet. Further commits will
allow for the pass through of this data from upper level protocols.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
when deleting an isis interface config with 'no [ip|ipv6] router isis'
we are destroying the isis yang container for that interface, but the
actual circuit struct is kept, and so are the flgs determining whether
that circuit is configured for ipv4 and/or ipv6. This caused issues
when removing and re-adding configuration, as the area counters for
ip circuits were not correctly updated and the topology was never
populated.
Signed-off-by: Emanuele Di Pascale <emanuele@voltanet.io>
Santosh P K [Mon, 25 Nov 2019 16:49:38 +0000 (08:49 -0800)]
bgpd: BGP assert when it tries to access peer which is closed.
Problem: BGP peer pointer is present in keepalive hash table
even when socket has been closed in some race condition.
When keepalive tries to access this peer it asserts.
RCA: Below sequence of events causing assert.
1. Config node peer has went down due to TCP reset
it's FD has been set to -1.
2. Doppelganger peer goes to established state and it has
been added to peer hash table for keepalive when it was
in openconfirm state.
3. Config node parameters including FD are exchanged with
doppelganger. Doppelganger will not have FD -1.
4. Doppelganger will be deleted as part of this it will
remove it from the keepalive peer hash table.
5. While removing from hash table it tries to acquire lock.
6. During this time keepalive thread has the lock and in
a loop trying to send keepalive for peers in hash table.
7. It tries to send keepalive for doppelganger peer with fd
set to -1 and asserts.
Quentin Young [Tue, 3 Dec 2019 20:48:27 +0000 (15:48 -0500)]
bgpd: more attribute parsing cleanup & paranoia
* Move VNC interning to the appropriate spot
* Use existing bgp_attr_flush_encap to free encap sets
* Assert that refcounts are correct before exiting to keep the demons
contained in their fiery prison
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
Stephen Worley [Thu, 5 Dec 2019 22:00:32 +0000 (17:00 -0500)]
pbrd: use spaces in show pbr map vty output
We were using a mix of spaces and tabsin show pbr map vty output.
Tabs can be inconsistent depending on the system settings.
Using spaces is a safer option for more consistent output.
Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
Stephen Worley [Thu, 5 Dec 2019 21:17:42 +0000 (16:17 -0500)]
pbrd: make vty nexthop/nexthop-group output consistent
The vty output for pbr maps with a nexthop-group was not
consistent with those configured with an individual nexthop.
Fix that so its easier for users to read.
Stephen Worley [Thu, 5 Dec 2019 20:52:08 +0000 (15:52 -0500)]
pbrd: make show pbr map detail actually work
The `detail` keyword was doing literally nothing. Changed the
default show to be a bit more user friendly and detail
to give the information you might would need for
debugging.
David Lamparter [Tue, 13 Aug 2019 16:24:11 +0000 (18:24 +0200)]
tools/frr-reload.py: remove stderr redirects
These make no sense. stderr=subprocess.STDOUT means that vtysh's stdout
and stderr are combined and returned by check_output. We don't expect
errors in that, and we certainly don't log them.
Leaving vtysh's stderr as stderr is perfectly fine, it'll be captured
for logging just like stderr output from frr-reload.py.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
David Lamparter [Thu, 1 Aug 2019 16:50:56 +0000 (18:50 +0200)]
bgpd: use new defaults system (v2)
This moves all the DFLT_BGP_* stuff over to the new defaults mechanism.
bgp_timers_nondefault() added to get better file-scoping.
v2: moved everything into bgp_vty.c so that the core BGP code is
independent of the CLI-specific defaults. This should make the future
northbound conversion easier.
Signed-off-by: David Lamparter <equinox@diac24.net>
David Lamparter [Fri, 26 Oct 2018 17:33:20 +0000 (19:33 +0200)]
build: apply DFLT_NAME/profiles in init script
Hopefully at some point we can get rid of the --enable-datacenter switch
and just have the init script do magic. Should already work for Cumulus
as it is.
NB: the profile name can't be baked into the package. The whole point
is to make the package profile-agnostic; in theory at some point the
exact same package files should work on both, say, a Cumulus switch and
a Linux software BGP DFZ router.
Signed-off-by: David Lamparter <equinox@diac24.net>
This commit modified the interface up handling code in
ZAPI such that the zclient handled the decoding for you.
Prior to this commit ospf assumed that it could use the
old ifp pointer to know state before reading the stream.
This lead to a situation where ospf would `smartly` track
and do the right thing in this situation. This commit
changed this assumption and in certain scenarios, say
a interface was changed after it was already up would
lead to situations where ospf would not properly handle
the new interface up.
Modify ospf to track data that is important to it in
it's interface->info pointer.
This code pattern was followed in both eigrp and pim.
In eigrp's case it was just behaving weirdly in any event
so fixing this pattern is not a big deal. In pim's
case it was not properly using this so it's a no-op
to fix.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
projects / mirror_frr.git / log