Implemented logic to check if the system supports required services.
Logical edges with 'requires' fields are mapped only if physical edges
provide corresponding services. If not, the test is skipped.
This ensures that tests only run when the required services are available
in the test system configuration.
Fixes#654
On slow systems, we don't want test steps involving tools like scp or
ssh-keyscan to fail because the server had not come up yet.
Therefore, make sure that the server is ready to accept connections
before moving on to subsequent test steps.
Use CAPTIALIZED names for static data and collect them at the top.
Make a note about for future readers about why a test called
ssh_key_authentication suspiciously does not enable the SSH service :)
Being able to check for a TCP based service is generally useful, not
only for NETCONF. Therefore, break it out to a separate function that
allows any port to be queried.
Break up the --test parameter into either:
--capture: Write system command output to a local directory for later
consumption by...
--replay: Which can replay a previous capture
Rename --wrap-commands => --cmd-prefix
This should be a non functional change. The idea is to use the topdoc
dir from 9pm when generating the test report.
Signed-off-by: Richard Alpe <richard@bit42.se>
Existing tests relied on manually captured system data from various
points in Infix's history, making them hard to extend as we add new
features.
Move to a new model where we:
To update a unit test:
1. Use Infamy tests to setup a DUT in a known state
2. Use yanger's wrapper mode to capture the system state needed to
produce operational data for a set of YANG models
3. Use cli-pretty to create the expected output for a set of show
commands
When running unit tests, we can then use the captured data to verify
that yanger and cli-pretty works as expected without the need for a
running Infix system - just like before. The difference is that it is
now much easier to capture a new snapshot when the system evolves.
Ensure that all flavors of GRE tunnels can be configured. In addition
to the increased test coverage, it will also be useful as a basis for
collecting representative system state for CLI testing.
Ensure that VLANs can be configured on VLAN filtering bridges. In
addition to the increased test coverage, it will also be useful as a
basis for collecting representative system state for CLI testing.
This is just a first step to add VXLAN tunnels, much more remains.
Implemented right now is as GRE, local, remote and a VNI.
Also refactor gre_basic and gre_bridged to generic tunnel tests,
the same test now tests VXLAN as well.
This is only availible for tests that require paramters. If the test have a
parameter --data the description and test steps can contain {data} which will
be replaced with tha actual data of the parameter data in the test specification.
The suite can also contain meta data for the test specification,
One meta data that is implemented is title, which is used when the
same test code is used twice with parameters.
Also if a test case should not be present in the test specification
for some case. You can add specification: False as meta data.
The yanger source grown significantly, code is disorganized, the
namespace is all over the place, etc.
Take a stab at remedying the situation by splitting out the individual
YANG models in separate Python modules, starting with ietf-system.
Using `sys.argv[0]` does not quite achieve the effect we want, namely
that any (test-case, $PYTHONHASHSEED) tuple should always default to
the same transport.
For example, the following two examples should always use the same
transport:
$ make PYTHONHASHSEED=1337 test-sh
infamy0:test$ ./case/ietf_system/hostname/test.py
$ make PYTHONHASHSEED=1337 test-sh
infamy0:test$ cd ./case/ietf_system/hostname
infamy0:hostname$ ./test.py
But, since their argv[0]'s are different, they don't.
Therefore use the _last two components_ of argv[0]'s full path
instead. This should:
- Spread the allocation over an entire suite run, since the
penultimate component is usually the test name.
- Keep the allocation stable when test code is modified (which is
common during debugging of failing tests)
- Avoid instabilities stemming from the local storage location (i.e.,
home directory names etc.)
The somewhat clumsy construct of resolving a logical port name to its
physical counterpart has been with us from Infamy's inception:
_, physical_port = env.ltop.xlate(logical_node, logical_port)
This is needlessly verbose. Since devices already have access to the
mappings which applies to it, make them easy to access:
target = env.attach("target")
# Get physical port names via the subscript operator
target["data0"]
Basic support GRE and GRETAP tunnels both for IPv4 and IPv6.
Limited support right now, only possible to configure local
ip address and remote ip address.
Tobias Waldekranz