The messages level of this commit were downgraded because some of them
happen on 'non-error' situations. This should help diminish the error
log verbosity on the CI-system run.
It was found a regression on an edge case when the second number in the
comparison was (at least) 2 numbers longer the comparison would fail
with a wrong return value. It succeeded for some cases because the
first comparison in the exception was correct, but not the second.
- Generic Debian only has a single space in front of “proto” in the linux shell routing outptu
Signed-off-by: Martin Winter <mwinter@opensourcerouting.org>
PR #1213 in FRR changed the protocol of installed LSPs. To avoid breaking
older outstanding Pull Requests, remove the protocol check.
Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
- Distance used to be wrong (0), new commit fixes this (90). To avoid breaking older outstanding Pull Requests, remove the distance check
Signed-off-by: Martin Winter <mwinter@opensourcerouting.org>
The values saved for the route table and eigrp topo
were a bit off because they have been corrected
in the eigrp daemon
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
This feature from pytest can not be used to detect runtime errors.
Also, remove test_router_running() since all tests should actually check
if all daemons are up and running.
Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
Nothing fancy here, just add 3 routers in a row
attempt to let eigrp come up and start a mininet
xterm to debug, since eigrp doesn't work yet.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
json check accidently checked for the spfLastDurationMsecs which
isn't always 0 for slower system in our tests. ARM7 sometimes has
a slightly higher value (1).
Signed-off-by: Martin Winter <mwinter@opensourcerouting.org>
Add check for number of routes to convergence. InQ=0, OutQ=0
together with correct number of routes received shows that
BGP has converged
Signed-off-by: Martin Winter <mwinter@opensourcerouting.org>
Using relative path to start the exabgp python scripts didn't work out
of the box in my enviroment, so be more specific since we already know
where the scripts are.
Add more tests to the ospf-topo1 to include IPv6 testing. Since both IP
versions are running together, there is no need to wait OSPF convergence
per IP version.
Standardized function that removes format spaces (or tab) and carriage
returns characters. This function is useful to allow output text
processing without breaking diff capabilities.
Output example:
*N IA 2001:db8:2::/64 :: r2-eth0 00:03:39
Becomes:
*N IA 2001:db8:2::/64 :: r2-eth0 00:03:39
If you remove 'IA' you won't have space formatting problem anymore.
Having a generic start/stop methods for TopoGear allows TopoGen to call
start/stop for all equipments. This allows us to reduce the teardown
code by removing the necessity of having to always remember to call
each equipment clean up function.
Auto configure daemon logging files to the appropriated place. This
removes the responsibility from the test developer to set this in the
daemon configuration.
TopoRouters now create a logger (which logs to /tmp/{router_name}.log)
on start to record all commands and events that it goes through. All log
messages contain timestamps that may be used in the future to:
(1) correlate commands call with events
(2) benchmark/time command speed
Allow topotest subsystems to create their own loggers. This will help
increase log organization and allow different settings to fit the
subsystems needs.
The default logger (root) is already being used by Mininet, so to allow
customizing logging output and configuring log files Topolog was
created. Topolog is no more than a thin layer abstraction to call
logging functions without using the 'root' logger.
Use a configuration file for casual settings like:
* Verbosity level (helps when debugging mininet issues)
* Custom daemon directory (in order to support running different daemon
binaries without touching tests)
* Daemon type switch: allow running quagga without touching any test
files
Also fix the add_router() documentation to include all options.
Create a specialized assert and json_cmp() result to improve the
comparison output. With this we also got a way to display all comparison
failures instead of just the first one.
This allows old tests to be run with '--topology-only' without
generating tons of error messages, instead it will just stop the test
without trying anything else.
Allow vtysh_cmd() to convert JSON output to Python data structures.
This feature will be used to get vtysh JSON outputs for tests
comparsions.
Usage example:
```py
router = get_topogen().gears['r1']
json_output = router.vtysh_cmd('show ip ospf json', isjson=True)
json_cmp(json_output, {'important_key': 'important_value'})
```
Implemented a JSON compare function that tells you when a specific
subset of items exist or not inside a JSON dataset.
More details can be found in the function docstring or in the test file
lib/test_json.py.
After some feedback from mwinter@, the names of equipments are now
shorter to make it easier to type them and to keep consistency with
mininet documentation. While here, update the template and make it use
optional name parameters for clarity.
Adding the __init__.py file makes python and its linter recognize that
the test makes part of the package, this makes us save a few lines of
code to make the linter and auto complete engines happy.
This commit changes how topology links are made in order to support
discovering who and what link is the node interface connected to. After
that, the implementation of the link state change functions were trivial
as calling a command in the node shell.
The method run() was moved from TopoRouter to TopoGear so all equipment
types can benefit from this code.
Topogen (Topology Generator) is a helper that wraps around Topotest to
simplify some of the boilerplate code. This abstraction will help the
development of new tests and new APIs without breaking the existing
ones. It also makes the relation of objects clearer, since we no longer
touch the Mininet API directly, which in turn also makes us less
vulnerable to external API changes.
Implemented two functions to help setting sysctl values:
* set_sysctl: set a sysctl and return an auditable return value
* assert_sysctl: uses the previous function to assert that the sysctl
was set
Now zebra is adding the ldpd implicit-null labels to the RIB as well. We
don't want to hide them in the "show ip route" commands because knowing
that a route is associated with an implicit-null label is an useful
piece of information, specially when troubleshooting L2/L3 VPNs.
Note: preserve the original output for cli version 1 (stable/2.0).
Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
Fixes the issue for topotest to fail ot end of skipped LDP test on
a system without MPLS support
Signed-off-by: Martin Winter <mwinter@opensourcerouting.org>
If daemon crashed at a later stage (not startup), then the test scripts didn't properly detect it and report unpredictable errors. This will properly log the daemon crashes
Signed-off-by: Martin Winter <mwinter@opensourcerouting.org>
When we have a kernel sub version > 10 the float conversion
of the kernel version causes 4.10 to be less than 4.5
Get the kernel version in groups on <major>.<minor> and do
comparison that way
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
* Added test to check for version
* Adopted all tests to verify against the correct version of output
Signed-off-by: Martin Winter <mwinter@opensourcerouting.org>
It's been a year since we added the new optional parameters
to instantiation. Let's switch over to the new name.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
The motivation for this patch is to address a concerning behavior of
tx-addpath-bestpath-per-AS. Prior to this patch, all paths' TX ID was
pre-determined as the path was received from a peer. However, this meant
that any time the path selected as best from an AS changed, bgpd had no
choice but to withdraw the previous best path, and advertise the new
best-path under a new TX ID. This could cause significant network
disruption, especially for the subset of prefixes coming from only one
AS that were also communicated over a bestpath-per-AS session.
The patch's general approach is best illustrated by
txaddpath_update_ids. After a bestpath run (required for best-per-AS to
know what will and will not be sent as addpaths) ID numbers will be
stripped from paths that no longer need to be sent, and held in a pool.
Then, paths that will be sent as addpaths and do not already have ID
numbers will allocate new ID numbers, pulling first from that pool.
Finally, anything left in the pool will be returned to the allocator.
In order for this to work, ID numbers had to be split by strategy. The
tx-addpath-All strategy would keep every ID number "in use" constantly,
preventing IDs from being transferred to different paths. Rather than
create two variables for ID, this patch create a more generic array that
will easily enable more addpath strategies to be implemented. The
previously described ID manipulations will happen per addpath strategy,
and will only be run for strategies that are enabled on at least one
peer.
Finally, the ID numbers are allocated from an allocator that tracks per
AFI/SAFI/Addpath Strategy which IDs are in use. Though it would be very
improbable, there was the possibility with the free-running counter
approach for rollover to cause two paths on the same prefix to get
assigned the same TX ID. As remote as the possibility is, we prefer to
not leave it to chance.
This ID re-use method is not perfect. In some cases you could still get
withdraw-then-add behaviors where not strictly necessary. In the case of
bestpath-per-AS this requires one AS to advertise a prefix for the first
time, then a second AS withdraws that prefix, all within the space of an
already pending MRAI timer. In those situations a withdraw-then-add is
more forgivable, and fixing it would probably require a much more
significant effort, as IDs would need to be moved to ADVs instead of
paths.
Signed-off-by Mitchell Skiba <mskiba@amazon.com>
This commit introduces lib/id_alloc, which has facilities for both an ID number
allocator, and less efficient ID holding pools. The pools are meant to be a
temporary holding area for ID numbers meant to be re-used, and are implemented
as a linked-list stack.
The allocator itself is much more efficient with memory. Based on sizeof
values on my 64 bit desktop, the allocator requires around 155 KiB per
million IDs tracked.
IDs are ultimately tracked in a bit-map split into many "pages." The
allocator tracks a list of pages that have free bits, and which sections
of each page have free IDs, so there isn't any scanning required to find
a free ID. (The library utility ffs, or "Find First Set," is generally a
single CPU instruction.) At the moment, totally empty pages will not be
freed, so the memory utilization of this allocator will remain at the
high water mark.
The initial intended use case is for BGP's TX Addpath IDs to be pulled
from an allocator that tracks which IDs are in use, rather than a free
running counter. The allocator reserves ID #0 as a sentinel value for
an invalid ID numbers, and BGP will want ID #1 reserved as well. To
support this, the allocator allows for IDs to be explicitly reserved,
though be aware this is only practical to use with low numbered IDs
because the allocator must allocate pages in order.
Signed-off-by Mitchell Skiba <mskiba@amazon.com>
We weren't cleaning up some files (a whole lot of python foobar) and had
some files in the dist tarball that don't quite belong there.
Signed-off-by: David Lamparter <equinox@diac24.net>
libunwind provides an alternate to backtrace() for printing out the call
stack of a particular location. It doesn't use the frame pointer, it
goes by the DWARF debug info. In most cases the traces have exactly the
same information, but there are some situations where libunwind traces
are better.
(On some platforms, the libc backtrace() also uses the DWARF debug info
[e.g.: ARM backtraces are impossible without it] but this is not the
case everywhere, especially not on BSD libexecinfo.)
Signed-off-by: David Lamparter <equinox@diac24.net>
Do a straight conversion of `struct bgp_info` to `struct bgp_path_info`.
This commit will setup the rename of variables as well.
This is being done because `struct bgp_info` is not descriptive
of what this data actually is. It is path information for routes
that we keep to build the actual routes nexthops plus some extra
information.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
