Simplify and fix the code that handles session teardown on control
plane shutdown (either failure or graceful).
- Don't move the `NULL` check inside `free` functions that expect
data: it creates harder to understand flows.
- Add some new debug messages to aid visualizing session deletions.
- Add sanity check error message (if it ever happens).
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
Dealing with PRIu64 is unfortunately a bit hacky in the frr-format
plugin, as in, it works correctly with snprintfrr, but breaks on plain
snprintf. There's no good solution unfortunately :/.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
These are easy to get subtly wrong, and doing so can cause
nondeterministic failures when racing in parallel builds.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
Having a fixed set of parameters for each northbound callback isn't a
good idea since it makes it difficult to add new parameters whenever
that becomes necessary, as several hundreds or thousands of existing
callbacks need to be updated accordingly.
To remediate this issue, this commit changes the signature of all
northbound callbacks to have a single parameter: a pointer to a
'nb_cb_x_args' structure (where x is different for each type
of callback). These structures encapsulate all real parameters
(both input and output) the callbacks need to have access to. And
adding a new parameter to a given callback is as simple as adding
a new field to the corresponding 'nb_cb_x_args' structure, without
needing to update any instance of that callback in any daemon.
This commit includes a .cocci semantic patch that can be used to
update old code to the new format automatically.
Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
Replace all `random()` calls with a function called `frr_weak_random()`
and make it clear that it is only supposed to be used for weak random
applications.
Use the annotation described by the Coverity Scan documentation to
ignore `random()` call warnings.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
And again for the name. Why on earth would we centralize this, just so
people can forget to update it?
Signed-off-by: David Lamparter <equinox@diac24.net>
Same as before, instead of shoving this into a big central list we can
just put the parent node in cmd_node.
Signed-off-by: David Lamparter <equinox@diac24.net>
There is really no reason to not put this in the cmd_node.
And while we're add it, rename from pointless ".func" to ".config_write".
[v2: fix forgotten ldpd config_write]
Signed-off-by: David Lamparter <equinox@diac24.net>
The only nodes that have this as 0 don't have a "->func" anyway, so the
entire thing is really just pointless.
Signed-off-by: David Lamparter <equinox@diac24.net>
Don't attempt to handle out-of-memory situations: XMALLOC/XCALLOC will
`assert` if there is no memory left.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
Move most of the log messages to debug guards so they only get activated
if the user configured the proper debug level.
Current debug levels:
- Peer events.
- Zebra events.
- Network layer debugs.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
According to the RFC 5880 the transmission time should be mandated by
the slowest system.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
Lets avoid garbage data on packets by zeroing the packet before setting
the fields/flags.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
This is a full rewrite of the "back end" logging code. It now uses a
lock-free list to iterate over logging targets, and the targets
themselves are as lock-free as possible. (syslog() may have a hidden
internal mutex in the C library; the file/fd targets use a single
write() call which should ensure atomicity kernel-side.)
Note that some functionality is lost in this patch:
- Solaris printstack() backtraces are ditched (unlikely to come back)
- the `log-filter` machinery is gone (re-added in followup commit)
- `terminal monitor` is temporarily stubbed out. The old code had a
race condition with VTYs going away. It'll likely come back rewritten
and with vtysh support.
- The `zebra_ext_log` hook is gone. Instead, it's now much easier to
add a "proper" logging target.
v2: TLS buffer to get some actual performance
Signed-off-by: David Lamparter <equinox@diac24.net>
Some logging systems are, er, "allergic" to tabs in log messages.
(RFC5424: "The syslog application SHOULD avoid octet values below 32")
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
It's been a year search and destroy.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
The vrrpd one conflicts with the standalone vrrpd package; also we're
installing daemons to /usr/lib/frr on some systems so they're not on
PATH.
Signed-off-by: David Lamparter <equinox@diac24.net>
Adding fields "detect-multiplier" and "remote-detect-multiplier"
for JSON to to reflect changes in "show bfd peers output"
Signed-off-by: Sayed Mohd Saquib sayed.saquib@broadcom.com
Adding new CLI clear bfd counters,
This CLI wil only reset Rx/Tx counters,
it will not reset session UP/DOWN and Zebra event count
Signed-off-by: Sayed Mohd Saquib <sayed.saquib@broadcom.com>
Use the interface VRF information instead of relying on the VRF specific
socket information.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
(cherry picked from commit c05c48621c)
Always bind the created sockets to their respective VRF devices. With
this it should be possible to run BFD on VRFs without needing to weaken
the security setting `net.ipv4.udp_l3mdev_accept=1`.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
Rearrange the bfdd northbound callbacks as following:
* bfd_nb.h: prototypes of all northbound callbacks.
* bfd_nb.c: definition of all northbound callbacks and their
associated YANG data paths.
* bfd_nb_config.c: implementation of YANG configuration nodes.
* bfd_nb_state.c: implementation of YANG state nodes.
This should help to keep to code more organized and easier to
maintain.
No behavior changes intended.
Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
Scenarios where this code change is required:
1. BFD is un-configured from BGP at remote end.
Neighbour BFD sends ADMIN_DOWN state, but BFD on local side will send
DOWN to BGP, resulting in BGP session DOWN.
Removing BFD session administratively shouldn't bring DOWN BGP session
at local or remote.
2. BFD is un-configured from BGP or shutdown locally.
BFD will send state DOWN to BGP resulting in BGP session DOWN.
(This is akin to saying do not use BFD for BGP)
Removing BFD session administratively shouldn't bring DOWN BGP session at
local or remote.
Signed-off-by: Sayed Mohd Saquib sayed.saquib@broadcom.com
When using link-local addresses we must provide scope-id to the
operating system so it knows where to send packets.
Spotted by Pavel Ivashchenko (@zays26).
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
Don't be selective about what to observe, always observe all possible
aspects of the session that may change on run-time (i.e. bind address,
interface and VRF existence).
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
If a session is no longer able to send/receive packets, it is very
likely it will be down in a few milliseconds so lets speed up the
process and correctly mark it as down.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
Otherwise the `thread_read` will keep waking us up to handle closing
sockets which are never unregistered.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
Current autocompletion works only for simple "vrf NAME" case.
This commit expands it also for the following cases:
- "nexthop-vrf NAME" in staticd
- usage of $varname in many daemons
All daemons are updated to use single varname "$vrf_name".
Signed-off-by: Igor Ryzhov <iryzhov@nfware.com>
Switch the zclient->interface_add functionality to have everyone
use the interface create callback in lib/if.c
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Start the conversion to allow zapi interface callbacks to be
controlled like vrf creation/destruction/change callbacks.
This will allow us to consolidate control into the interface.c
instead of having each daemon read the stream and react accordingly.
This will hopefully reduce a bunch of cut-n-paste stuff
Create 4 new callback functions that will be controlled by
lib/if.c
create -> A upper level protocol receives an interface creation event
The ifp is brand spanking newly created in the system.
up -> A upper level protocol receives a interface up event
This means the interface is up and ready to go.
down -> A upper level protocol receives a interface down
destroy -> A upper level protocol receives a destroy event
This means to delete the pointers associated with it.
At this point this is just boilerplate setup for future commits.
There is no new functionality.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Adding a lock to protect the global running configuration doesn't
help much since the FRR daemons are not prepared to process
configuration changes in a pthread that is not the main one (a
whole lot of new protections would be necessary to prevent race
conditions).
This means the lock added by commit 83981138 only adds more
complexity for no benefit. Remove it now to simplify the code.
All northbound clients, including the gRPC one, should either run
in the main pthread or use synchronization primitives to process
configuration transactions in the main pthread.
This reverts commit 83981138fe.
upon vrf disable, an event informs bfd daemon that the vrf contexts
should be removed. in the case a vrf backend is netns based, all sockets
opened under that netns have to be closed. otherwise it is impossible
for the system to completely close the network namespace. that implies
that some interfaces may not be deleted, and may not be given back to
default vrf.
PR=65291
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Acked-by: Julien Floret <julien.floret@6wind.com>
bfd operational & config data may already applied and available, while
an external event requests for changing the vrf name. this change
updates the config and operational context of yang.
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
if the bfd session is already enabled, then dynamically change the vrf
name if the vrf where bfd is executed changed its name.
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Add source address to northbound when creating sessions with this
information. It is not possible to change source address after the
session was created, but we should be able to set it to make IPv6 work.
Spotted by Philippe Guibert.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
This helps northbound to create the `bfd` node on the configuration
output sooner than adding a peer.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
Build will fail with `Werror` enabled with:
```
CC bfdd/bfdd_cli.o
In file included from ./lib/frratomic.h:21:0,
from ./lib/memory.h:22,
from ./lib/vector.h:25,
from ./lib/command.h:25,
from bfdd/bfdd_cli.c:23:
./config.h:665:0: error: "_FILE_OFFSET_BITS" redefined [-Werror]
#define _FILE_OFFSET_BITS 64
^
In file included from /usr/include/iso/stdlib_iso.h:49:0,
from /usr/include/stdlib.h:37,
from ./lib/memory.h:20,
from ./lib/vector.h:25,
from ./lib/command.h:25,
from bfdd/bfdd_cli.c:23:
/opt/gcc-5.1.0/lib/gcc/i386-pc-solaris2.11/5.1.0/include-fixed/sys/feature_tests.h:231:0: note: this is the location of the previous definition
#define _FILE_OFFSET_BITS 32
^
In file included from ./lib/thread.h:24:0,
from ./lib/vty.h:27,
from ./lib/command.h:26,
from bfdd/bfdd_cli.c:23:
./lib/zebra.h:271:2: error: #warning "assuming 4-byte alignment for CMSG_SPACE" [-Werror=cpp]
#warning "assuming 4-byte alignment for CMSG_SPACE"
^
./lib/zebra.h:277:2: error: #warning "assuming 4-byte alignment for CMSG_LEN" [-Werror=cpp]
#warning "assuming 4-byte alignment for CMSG_LEN"
^
cc1: all warnings being treated as errors
Makefile:6876: recipe for target 'bfdd/bfdd_cli.o' failed
gmake[1]: *** [bfdd/bfdd_cli.o] Error 1
```
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
Lets allow specification to accept microseconds, but limit the timers
configuration in FRR to milliseconds (minimum is 10 ms and maximum is 60
seconds).
This matches the RFC 5880 and the IETF BFD YANG draft model.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
Add command 'no bfd' to remove all BFD sessions configuration and fix
other daemon integration.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
```
warnings: BFD: [EC 100663317] YANG model "frr-bfdd@*" not embedded, trying external file
```
Embed the YANG model into the binary to avoid reading an external file.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
When the remote discriminator hasn't been assigned yet, then we can't
return a value of 0. The value '0' is an invalid discriminator and the
yang handlers will complain about it.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
Lets start using the new BFD yang model and translate the BFD session
configuration commands to use the northbound.
One important change: all sessions will default to use
`VRF_DEFAULT_NAME` (usually "default") when no VRF is configured. All
places which search for BFD sessions must now take this into account.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
These functions are going to be used by the future northbound
implementation to handle BFD sessions.
Signed-off-by: Rafael Zalamena <rzalamena@opensourcerouting.org>
when working with a standard vrf backend, bfdd ignores that and tries to
create and configure bfd sockets for each vrf, which will fail for the
second vrf discovered, since the network namespace used is the same, and
it is not possible to use same socket settings twice. Handle this case,
and avoids to reinitialise sockets.
This patch however does not leverage bfd support for vrf-lite.
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
MTYPE definitions should be local to the file using them whereever
possible. Also remove some superfluous ;
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
