The nexthop group code is installing routes and nexthop groups
and immediately expecting zebra to have processed the results
as a result there is a situation when the CI system is under
intense load that the nexthop group might not have been processed.
Add a bit of code to allow the test to give FRR some time
to finish work before declaring it not working.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
There is a bit of an impedance mismatch in the sequence of events here.
Depending on the dplane behavior, the `ROUTE_ENTRY_SELECTED` bit will be
inconsistent for rib_process_result().
With an asynchronous dataplane:
0. rib_process() is called
1. rib_install_kernel() is called, dplane action is queued
2. rib_install_kernel() returns
3. rib_process() sets the SELECTED bit appropriately, returns
4. dplane is done, triggers rib_process_result()
5. SELECTED bit is seen in "after" state
(5a. NHT code looks at the SELECTED bit, works correctly.)
With a synchronous dataplane:
0. rib_process() is called
1. rib_install_kernel() is called, dplane action is executed
2. dplane (should) trigger rib_process_result()
3. SELECTED bit is seen in "before" state
(3a. NHT code looks at the SELECTED bit, fails.)
4. rib_install_kernel() returns
5. rib_process() sets the SELECTED bit appropriately, too late.
Essentially, poking the dataplane is a sequencing point where control is
handed over to the dplane. Control may or may not return immediately.
Doing /anything/ after triggering the dataplane is a recipe for odd race
conditions.
(FWIW, I'm not sure rib_process_result() is called correctly in the
synchronous case, but that's a separate problem.)
Unfortunately, this change might have some unforeseen side effects. I
haven't dug through the code to see if anything breaks. There
/shouldn't/ be anything looking at the SELECTED bit here, but who knows.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
When the CI system is heavily loaded, we might see the following failures:
```
test failed at "test_config_timing/test_static_timing": assert 20.083204 <= 19.487716
```
Currently we allow each step to run 2 times slower than the initial
measurement. Let's allow them to run 3 times slower.
Signed-off-by: Igor Ryzhov <iryzhov@nfware.com>
On the first step, the test creates 10000 static routes. It passes 10000
to `get_ip_networks` and it generates 10000 /22 routes.
On the fourth step, the test tries to remove 5000 previously created
routes. It passes 5000 to `get_ip_networks` and here starts the problem.
Instead of generating 5000 /22 routes, it generates 5000 /21 routes. And
the whole step is a no-op, we constantly see the following logs:
```
% Refusing to remove a non-existent route
```
To consistently generate same routes, `get_ip_networks` must always use
the same prefix length.
Signed-off-by: Igor Ryzhov <iryzhov@nfware.com>
Do not return pointer to the newly created thread from various thread_add
functions. This should prevent developers from storing a thread pointer
into some variable without letting the lib know that the pointer is
stored. When the lib doesn't know that the pointer is stored, it doesn't
prevent rescheduling and it can lead to hard to find bugs. If someone
wants to store the pointer, they should pass a double pointer as the last
argument.
Signed-off-by: Igor Ryzhov <iryzhov@nfware.com>
rib_update() was mallocing memory then attempting to schedule
and if the schedule failed( it was already going to be run )
FRR would then free the memory. Fix this memory usage pattern
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
The function thread_is_scheduled allows us to know if
the particular thread is scheduled for execution or not.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
Our topotests send SIGBUS 2 seconds after a SIGTERM is
initiated. This is bad because under a heavily loaded
topotest system we may have a case where the system has
not had a chance to properly shut down the daemon.
Extend the time greatly before topotests send SIGBUS.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
pathd was never calling frr_fini and as such the
proper shutdown was not happening for libfrr.
This includes logging.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
When doing a normal exit from ospf we should close
the log file as that we are leaving a bunch of
unterminated logging processes by not doing so.
Signed-off-by: Donald Sharp <sharpd@nvidia.com>
```
exit1-debian-9(config)# route-map test1 permit 10
exit1-debian-9(config-route-map)# match community ?
(1-99) Community-list number (standard)
(100-500) Community-list number (expanded)
COMMUNITY_LIST_NAME Community-list name
testas
exit1-debian-9(config-route-map)# match large-community ?
(1-99) Large Community-list number (standard)
(100-500) Large Community-list number (expanded)
LCOMMUNITY_LIST_NAME Large Community-list name
LCL-ORIGINATED-ALL
exit1-debian-9(config-route-map)#
```
Signed-off-by: Donatas Abraitis <donatas.abraitis@gmail.com>
This removes a giant `switch { }` block from lib/zclient.c and
harmonizes all zclient callback function types to be the same (some had
a subset of the args, some had a void return, now they all have
ZAPI_CALLBACK_ARGS and int return.)
Apart from getting rid of the giant switch, this is a minor security
benefit since the function pointers are now in a `const` array, so they
can't be overwritten by e.g. heap overflows for code execution anymore.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
Dynamic peer count is inconsistent in
"show bgp summary json" and "show bgp summary failed json" due to
dynamic peer counter 'dn_count' being reused without resetting
Signed-off-by: Abhishek Naik <bhini@amazon.com>
These aren't appropriate for use in FRR. Among other things, this
enables running checkpatch by calling it in a git working tree with
`tools/checkpatch.pl -g origin/master..`
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
For the purpose of allowing the space in `frr_each (`, copy the list of
iterators from .clang-format and wire it up appropriately.
Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
