From 8adf83fcbc0778616701e487ddd841e66aa85234 Mon Sep 17 00:00:00 2001 From: "jose.lausuch" Date: Tue, 9 Feb 2016 18:29:01 +0100 Subject: Doc updates - Some clarifications, rephrasing - Remove more statements related to the DB - Add more info for vPing - Add more info about clean defaults Change-Id: I12e97443852dc38a23bda589945ca0333a37e410 Signed-off-by: jose.lausuch (cherry picked from commit e637e75435587014cb99edbc57e79970095d0464) --- docs/userguide/description.rst | 51 +++++------ docs/userguide/index.rst | 203 +++++++++++++++++++++++------------------ docs/userguide/runfunctest.rst | 123 +++++++++++++++---------- 3 files changed, 211 insertions(+), 166 deletions(-) diff --git a/docs/userguide/description.rst b/docs/userguide/description.rst index 288ad95af..b47d5ef62 100644 --- a/docs/userguide/description.rst +++ b/docs/userguide/description.rst @@ -1,12 +1,9 @@ -Description of the test cases +Overview of the test suites ============================= -Functest is an OPNFV project dedicated to functional testing. -In the continuous integration, it is launched after an OPNFV fresh installation. -The Functest target is to verify the basic functions of the infrastructure. - -Functest includes different test suites which several test cases within. -Test cases are developed in Functest and in feature projects. +Functest is the OPNFV project primarily targeting function testing. +In the Continuous Integration pipeline, it is launched after an OPNFV fresh +installation to validate and verify the basic functions of the infrastructure. The current list of test suites can be distributed in 3 main domains:: @@ -21,8 +18,8 @@ The current list of test suites can be distributed in 3 main domains:: | +----------------+-------------------------------------------+ |(Virtualised | Tempest | OpenStack reference test suite `[2]`_ | | Infrastructure +----------------+-------------------------------------------+ - | Manager) | Rally scenario | OpenStack testing tool testing OpenStack | - | | | modules `[3]`_ | + | Manager) | Rally bench | OpenStack testing tool benchmarking | + | | | OpenStack modules `[3]`_ | +----------------+----------------+-------------------------------------------+ | | OpenDaylight | Opendaylight Test suite | | +----------------+-------------------------------------------+ @@ -30,8 +27,8 @@ The current list of test suites can be distributed in 3 main domains:: | +----------------+-------------------------------------------+ | | OpenContrail | | +----------------+----------------+-------------------------------------------+ - | Features | vIMS | Show the capability to deploy a real NFV | - | | | test cases. | + | Features | vIMS | Example of a real VNF deployment to show | + | | | the NFV capabilities of the platform. | | | | The IP Multimedia Subsytem is a typical | | | | Telco test case, referenced by ETSI. | | | | It provides a fully functional VoIP System| @@ -46,31 +43,33 @@ The current list of test suites can be distributed in 3 main domains:: | | SDNVPN | | +----------------+----------------+-------------------------------------------+ - -Most of the test suites are developed upstream. -For example, `Tempest `_ is the -OpenStack integration test suite. -Functest is in charge of the integration of different functional test suites. +Functest includes different test suites with several test cases within. Some +of the tests are developed by Functest team members whereas others are integrated +from upstream communities or other OPNFV projects. For example, +`Tempest `_ is the +OpenStack integration test suite and Functest is in charge of the selection, +integration and automation of the tests that fit in OPNFV. The Tempest suite has been customized but no new test cases have been created. -Some OPNFV feature projects (e.g. SDNVPN) have created Tempest tests cases and -pushed to upstream. +Some OPNFV feature projects (e.g. SDNVPN) have written some Tempest tests cases +and pushed upstream to be used by Functest. -The tests run from CI are pushed into a database. -The goal is to populate the database with results and to show them on a Test -Dashboard. +The results produced by the tests run from CI are pushed and collected in a NoSQL +database. The goal is to populate the database with results from different sources +and scenarios and to show them on a Dashboard. -There is no real notion of Test domain or Test coverage yet. -Basic components (VIM, controllers) are tested through their own suites. -Feature projects also provide their own test suites. +There is no real notion of Test domain or Test coverage. Basic components +(VIM, controllers) are tested through their own suites. Feature projects also +provide their own test suites with different ways of running their tests. vIMS test case was integrated to demonstrate the capability to deploy a relatively complex NFV scenario on top of the OPNFV infrastructure. Functest considers OPNFV as a black box. -OPNFV, since Brahmaputra, offers lots of possible combinations: +OPNFV, since the Brahmaputra release, offers lots of potential combinations: * 3 controllers (OpenDayligh, ONOS, OpenContrail) * 4 installers (Apex, Compass, Fuel, Joid) -However most of the tests shall be runnable on any configuration. +Most of the tests are runnable on any combination, but some others might have +restrictions imposed by the installers or the available deployed features. diff --git a/docs/userguide/index.rst b/docs/userguide/index.rst index 39d795261..89e3ef115 100644 --- a/docs/userguide/index.rst +++ b/docs/userguide/index.rst @@ -11,7 +11,7 @@ Introduction ============ The goal of this documents is to describe the Functest test cases as well as -provide a procedure about how to execute (or launch) them. +provide a procedure about how to execute them. A presentation has been created for the first OPNFV Summit `[4]`_. @@ -21,79 +21,100 @@ It is assumed that Functest container has been properly installed `[1]`_. The different scenarios are described in the section hereafter. -VIM +VIM (Virtualized Infrastructure Manager) --- vPing ^^^^^ -The goal of this test can be described as follows:: +Given the script 'ping.sh':: + + #!/bin/sh + while true; do + ping -c 1 $1 2>&1 >/dev/null + RES=$? + if [ "Z$RES" = "Z0" ] ; then + echo 'vPing OK' + break + else + echo 'vPing KO' + fi + sleep 1 + done + +The goal of this test is described as follows:: vPing test case - +-------------+ +-------------+ - | | | | - | | Boot VM1 | | - | +------------------>| | - | | | | - | | Get IP VM1 | | - | +------------------>| | - | Tester | | System | - | | Boot VM2 | Under | - | +------------------>| Test | - | | | | - | | Create (VM2) | | - | | floating IP | | - | +------------------>| | - | | | | - | | SCP vPing script | | - | | to VM2 | | - | +------------------>| | - | | | | - | | SSH to VM2 | | - | +------------------>| | - | | | | - | | Ping VM1 | | - | | private IP | | - | +------------------>| | - | | | | - | | If ping: | | - | | exit OK | | - | | else (timeout) | | - | | exit KO | | - | | | | - +-------------+ +-------------+ - -This example can be considered as an "Hello World" example. -It is the first basic example, it must work on any configuration. + +-------------+ +-------------+ + | | | | + | | Boot VM1 with IP1 | | + | +------------------->| | + | Tester | | System | + | | Boot VM2 | Under | + | +------------------->| Test | + | | | | + | | Create floating IP | | + | +------------------->| | + | | | | + | | Assign floating IP | | + | | to VM2 | | + | +------------------->| | + | | | | + | | Stablish SSH | | + | | connection to VM2 | | + | | through floating IP| | + | +------------------->| | + | | | | + | | SCP ping.sh to VM2 | | + | +------------------->| | + | | | | + | | VM2 executes | | + | | ping.sh to VM1 | | + | +------------------->| | + | | | | + | | If ping: | | + | | exit OK | | + | | else (timeout) | | + | | exit Failed | | + | | | | + +-------------+ +-------------+ + +This test can be considered as an "Hello World" example. +It is the first basic use case which shall work on any deployment. vPing_userdata ^^^^^^^^^^^^^^ -The goal of this test can be described as follow:: +The goal of this test can be described as follows:: vPing_userdata test case - +-------------+ +-------------+ - | | | | - | | Boot VM1 | | - | +------------------>| | - | | | | - | | Get IP VM1 | | - | +------------------>| | - | Tester | | System | - | | Boot VM2 | Under | - | +------------------>| Test | - | | VM2 pings VM1 | | - | | (cloud-init) | | - | | Check console log | | - | | If ping: | | - | | exit OK | | - | | else (timeout) | | - | | exit KO | | - | | | | - +-------------+ +-------------+ + +-------------+ +-------------+ + | | | | + | | Boot VM1 with IP1 | | + | +------------------->| | + | | | | + | | Boot VM2 with | | + | | ping.sh as userdata| | + | | with IP1 as $1. | | + | +------------------->| | + | Tester | | System | + | | VM2 exeutes ping.sh| Under | + | | (ping IP1) | Test | + | +------------------->| | + | | | | + | | Monitor nova | | + | | console-log VM 2 | | + | | If ping: | | + | | exit OK | | + | | else (timeout) | | + | | exit Failed | | + | | | | + +-------------+ +-------------+ This scenario is similar to the previous one but it uses cloud-init (nova -metadata service) instead of floating IPs and SSH. +metadata service) instead of floating IPs and SSH connection. When the second VM +boots it will execute the script automatically and the ping will be detected +capturing periodically the output in the console-log of the second VM. Tempest @@ -107,27 +128,27 @@ Tempest has batteries of tests for: * Scenarios * Other specific tests useful in validating an OpenStack deployment -We use Rally `[3]`_ to run Tempest suite. -Rally generates automatically tempest.conf configuration file. -Before running actual test cases Functest creates needed resources. -Needed parameters are updated in the configuration file. -When the Tempest suite is run, each test duration is measured. -The full console output is stored in the tempest.log file. +Functest uses Rally `[3]`_ to run the Tempest suite. +Rally generates automatically the Tempest configuration file (tempest.conf). +Before running the actual test cases, Functest creates the needed resources and +updates the appropriate parameters to the configuration file. +When the Tempest suite is executed, each test duration is measured and the full +console output is stored in the tempest.log file for further analysis. As an addition of Arno, Brahmaputra runs a customized set of Tempest test cases. The list is specificed through *--tests-file* when running Rally. This option has been introduced in Rally in version 0.1.2. The customized test list is available in the Functest repo `[4]`_. -This list contains more than 200 Tempest test cases. -The list can be divied into two main parts: +This list contains more than 200 Tempest test cases and can be divided +into two main sections: 1) Set of tempest smoke test cases 2) Set of test cases from DefCore list `[8]`_ -The goal of Tempest test suite is to check the basic functionalities of +The goal of the Tempest test suite is to check the basic functionalities of different OpenStack components on an OPNFV fresh installation using -corresponding REST API interfaces. +the corresponding REST API interfaces. Rally bench test suites @@ -137,8 +158,8 @@ Rally `[3]`_ is a benchmarking tool that answers the question:: “How does OpenStack work at scale?”. -The goal of this test suite is to test the different modules of OpenStack and -get significant figures that could help us to define telco Cloud KPI. +The goal of this test suite is to benchmark the different OpenStack modules and +get significant figures that could help to define Telco Cloud KPIs. The OPNFV scenarios are based on the collection of the existing Rally scenarios: @@ -152,7 +173,7 @@ The OPNFV scenarios are based on the collection of the existing Rally scenarios: * quotas * requests -Basic SLA (stop test on errors) have been implemented. +A basic SLA (stop test on errors) have been implemented. SDN Controllers @@ -169,11 +190,11 @@ OpenDaylight ^^^^^^^^^^^^ The OpenDaylight (ODL) test suite consists of a set of basic tests inherited -from ODL project. +from the ODL project using the Robot `[11]`_ framework. The suite verifies creation and deletion of networks, subnets and ports with OpenDaylight and Neutron. -The list of tests can be described as follow: +The list of tests can be described as follows: * Restconf.basic: Get the controller modules via Restconf * Neutron.Networks @@ -216,14 +237,13 @@ The list of tests can be described as follow: ONOS ^^^^ -TestON Framework is used to test ONOS function. +TestON Framework is used to test the ONOS SDN controller functions. The test cases deal with L2 and L3 functions. -ONOS is configured through OPNFV scenario. The ONOS test suite can be run on any ONOS compliant scenario. -The test cases may be described as follow: +The test cases may be described as follows: - * onosfunctest: The mainly executable file contains the initialization of + * onosfunctest: The main executable file contains the initialization of the docker environment and functions called by FUNCvirNetNB and FUNCvirNetNBL3 @@ -273,7 +293,7 @@ vIMS ^^^^ The goal of this test suite consists of: - * deploying a VNF orchestrator (cloudify) + * deploy a VNF orchestrator (Cloudify) * deploy a Clearwater vIMS (IP Multimedia Subsystem) VNF from this orchestrator based on a TOSCA blueprint defined in `[5]`_ * run suite of signaling tests on top of this VNF @@ -290,7 +310,7 @@ Two types of information are stored in the Test Database: * the test results The deployment of a complete functional VNF allows the test of most of the -essential functions needed for a NFV system. +essential functions needed for a NFV platform. Promise ^^^^^^^ @@ -340,6 +360,9 @@ include:: flavor_ram: 512 flavor_disk: 0 +However, these parameters must not be changed, as they are the values expected +by the Promise test suite. + .. include:: ./runfunctest.rst Test results @@ -351,7 +374,7 @@ VIM vPing ^^^^^ -vPing results are displayed in the console:: +vPing test case output is displayed in the console:: FUNCTEST.info: Running vPing test... 2016-01-23 03:18:20,153 - vPing- INFO - Creating neutron network vping-net... @@ -428,8 +451,8 @@ The Tempest results are displayed in the console:: 2016-01-28 08:19:32,133 - run_tempest - INFO - Pushing results to DB: 'http://testresults.opnfv.org/testapi/results'. 2016-01-28 08:19:32,278 - run_tempest - INFO - Deleting tenant and user for Tempest suite) -In order to check all the available test cases related debug information, please -inspect tempest.log file stored into related Rally deployment folder. +In order to check all the available test cases related debug information, inspect +tempest.log file stored under */home/opnfv/functest/results/tempest/*. Rally @@ -493,9 +516,9 @@ The results of ODL tests can be seen in the console:: 1 critical test, 1 passed, 0 failed 1 test total, 1 passed, 0 failed ============================================================================== - Output: /home/jenkins-ci/workspace/functest-opnfv-jump-2/output.xml - Log: /home/jenkins-ci/workspace/functest-opnfv-jump-2/log.html - Report: /home/jenkins-ci/workspace/functest-opnfv-jump-2/report.html + Output: /home/opnfv/repos/functest/output.xml + Log: /home/opnfv/repos/functest/log.html + Report: /home/opnfv/repos/functest/report.html .............................................................................. @@ -507,9 +530,9 @@ The results of ODL tests can be seen in the console:: 18 critical tests, 18 passed, 0 failed 18 tests total, 18 passed, 0 failed ============================================================================== - Output: /home/jenkins-ci/workspace/functest-opnfv-jump-2/output.xml - Log: /home/jenkins-ci/workspace/functest-opnfv-jump-2/log.html - Report: /home/jenkins-ci/workspace/functest-opnfv-jump-2/report.html + Output: /home/opnfv/repos/functest/output.xml + Log: /home/opnfv/repos/functest/log.html + Report: /home/opnfv/repos/functest/report.html 3 result files are generated: * output.xml @@ -659,7 +682,6 @@ Please note that vIMS traces can bee summarized in several steps: * INFO - Cloudify-manager server is UP ! => orchestrator deployed * INFO - The deployment of clearwater-opnfv is ended => VNF deployed * Multiple Identities (UDP) - (6505550771, 6505550675) Passed => tests run - * DEBUG - Pushing results to DB.... => tests saved Promise @@ -833,6 +855,7 @@ References .. _`[8]`: https://wiki.openstack.org/wiki/Governance/DefCoreCommittee .. _`[9]`: https://git.opnfv.org/cgit/functest/tree/testcases/VIM/OpenStack/CI/libraries/os_defaults.yaml .. _`[10]`: https://git.opnfv.org/cgit/functest/tree/testcases/VIM/OpenStack/CI/rally_cert/task.yaml +.. _`[11]`: http://robotframework.org/ OPNFV main site: opnfvmain_. diff --git a/docs/userguide/runfunctest.rst b/docs/userguide/runfunctest.rst index b186e539c..94e11a63a 100644 --- a/docs/userguide/runfunctest.rst +++ b/docs/userguide/runfunctest.rst @@ -5,17 +5,22 @@ Manual testing -------------- Once the Functest docker container is running and Functest environment ready -(through /home/opnfv/repos/functest/docker/prepare_env.sh script), the system is +(through */home/opnfv/repos/functest/docker/prepare_env.sh* script), the system is ready to run the tests. -The script *run_tests.sh* is located in $repos_dir/functest/docker and it has +The script *run_tests.sh* launches the test in an automated way. +Although it is possible to execute the different tests manually, it is +recommended to use the previous shell script which makes the call +to the actual scripts with the appropriate parameters. + +It is located in *$repos_dir/functest/docker* and it has several options:: ./run_tests.sh -h Script to trigger the tests automatically. usage: - bash run_tests.sh [--offline] [-h|--help] [-t ] + bash run_tests.sh [-h|--help] [-r|--report] [-n|--no-clean] [-t|--test ] where: -h|--help show this help text @@ -29,65 +34,85 @@ several options:: run_tests.sh --test vping,odl run_tests.sh -t tempest,rally --no-clean -The *-r* option is used by the Continuous Integration in order to push the test -results into a test collection database, see in next section for details. -In manual mode, you must not use it, your try will be anyway probably rejected -as your POD must be declared in the database to collect the data. +The *-r* option is used by the OPNFV Continuous Integration automation mechanisms +in order to push the test results into the NoSQL results collection database. +This database is read only for a regular user given that it needs special rights +and special conditions to push data. -The *-n* option is used for preserving all the existing OpenStack resources after -execution test cases. +The *-t* option can be used to specify the list of a desired test to be launched, +by default Functest will launch all the test suites in the following order: +vPing, Tempest, vIMS, Rally. -The *-t* option can be used to specify the list of test you want to launch, by -default Functest will try to launch all its test suites in the following order -vPing, odl, Tempest, vIMS, Rally. -You may launch only one single test by using *-t *. +A single or set of test may be launched at once using *-t * specifying +the test name or names separated by commas in the following list: +*[vping,vping_userdata,odl,rally,tempest,vims,onos,promise]*. -Within Tempest test suite you can define which test cases you want to execute in -your environment by editing test_list.txt file before executing *run_tests.sh* -script. +The *-n* option is used for preserving all the possible OpenStack resources created +by the tests after their execution. Please note that Functest includes cleaning mechanism in order to remove -everything except what was present after a fresh install. -If you create your own VMs, tenants, networks etc. and then launch Functest, -they all will be deleted after executing the tests. Use the *--no-clean* option with -run_test.sh in order to preserve all the existing resources. -However, be aware that Tempest and Rally create of lot of resources (users, +all the VIM resources except what was present before running any test. The script +*$repos_dir/functest/testcases/VIM/OpenStack/CI/libraries/generate_defaults.py* +is called once by *prepare_env.sh* when setting up the Functest environment +to snapshot all the OpenStack resources (images, networks, volumes, security groups, +tenants, users) so that an eventual cleanup does not remove any of this defaults. + +The script +*$repos_dir/functest/testcases/VIM/OpenStack/CI/libraries/clean_openstack.py* +is normally called after a test execution if the *-n* is not specified. It +is in charge of cleaning the OpenStack resources that are not specified +in the defaults file generated previously which is stored in +*/home/opnfv/functest/conf/os_defaults.yaml* in the docker +container. + +It is important to mention that if there are new OpenStack resources created +manually after preparing the Functest environment, they will be removed if this +flag is not specified in the *run_tests.sh* command. +The reason to include this cleanup meachanism in Functest is because some +test suites such as Tempest or Rally create a lot of resources (users, tenants, networks, volumes etc.) that are not always properly cleaned, so this -cleaning function has been set to keep the system as clean as possible after a -full Functest run. +cleaning function has been set to keep the system as clean as it was before a +full Functest execution. -You may also add you own test by adding a section into the function run_test(). +Within the Tempest test suite it is possible to define which test cases to execute +by editing *test_list.txt* file before executing *run_tests.sh* script. This file +is located in *$repos_dir/functest/testcases/VIM/OpenStack/CI/custom_tests/test_list.txt* +Although *run_tests.sh* provides an easy way to run any test, it is possible to +do a direct call to the desired test script. For example:: + python $repos_dir/functest/testcases/vPing/vPing.py -d Automated testing ----------------- -As mentioned in `[1]`, the *prepare-env.sh* and *run_test.sh* can be executed within -the container from jenkins. -2 jobs have been created, one to run all the test and one that allows testing -test suite by test suite. -You thus just have to launch the acurate jenkins job on the target lab, all the -tests shall be automatically run. +As mentioned in `[1]`, the *prepare-env.sh* and *run_test.sh* can be called within +the container from Jenkins. There are 2 jobs that automate all the manual steps +explained in the previous section. One job runs all the tests and the other one allows testing +test suite by test suite specifying the test name. The user might use one or +the other job to execute the desired test suites. + +One of the most challenging task in the Brahmaputra release consists +in dealing with lots of scenarios and installers. Thus, when the tests are +automatically started from CI, a basic algorithm has been created in order to +detect whether a given test is runnable or not on the given scenario. +Some Functest test suites cannot be systematically run (e.g. ODL suite can not +be run on an ONOS scenario). + -When the tests are automatically started from CI, a basic algorithm has been -created in order to detect whether the test is runnable or not on the given -scenario. -In fact, one of the most challenging task in Brahmaputra consists in dealing -with lots of scenario and installers. -Functest test suites cannot be systematically run (e.g. run the ODL suite on an -ONOS scenario). -CI provides several information: +CI provides some useful information passed to the container as environment +variables: - * The installer (apex|compass|fuel|joid) - * The scenario [controller]-[feature]-[mode] with + * Installer (apex|compass|fuel|joid), stored in INSTALLER_TYPE + * Installer IP of the engine or VM running the actual deployment, stored in INSTALLER_IP + * The scenario [controller]-[feature]-[mode], stored in DEPLOY_SCENARIO with * controller = (odl|onos|ocl|nosdn) * feature = (ovs(dpdk)|kvm) * mode = (ha|noha) -Constraints per test case are defined in the Functest configuration file -/home/opnfv/functest/config/config_functest.yaml:: +The constraints per test case are defined in the Functest configuration file +*/home/opnfv/functest/config/config_functest.yaml*:: test-dependencies: functest: @@ -105,10 +130,11 @@ Constraints per test case are defined in the Functest configuration file .... At the end of the Functest environment creation (prepare_env.sh see `[1]`_), a -file (/home/opnfv/functest/conf/testcase-list.txt) is created with the list of +file */home/opnfv/functest/conf/testcase-list.txt* is created with the list of all the runnable tests. -We consider the static constraints as regex and compare them with the scenario. -For instance, odl can be run only on scenario including odl in its name. +Functest considers the static constraints as regular expressions and compare them +with the given scenario name. +For instance, ODL suite can be run only on an scenario including 'odl' in its name. The order of execution is also described in the Functest configuration file:: @@ -135,8 +161,5 @@ The tests are executed in the following order: * Feature projects (promise, vIMS) * Rally (benchmark scenario) -At the end of an automated execution, everything is cleaned. -Before running Functest, a snapshot of the OpenStack configuration (users, -tenants, networks, ....) is performed. After Functest, a clean mechanism is -launched to delete everything that would not have been properly deleted in order -to restitute the system as it was prior to the tests. +As explained before, at the end of an automated execution, the OpenStack resources +might be eventually removed. -- cgit 1.2.3-korg