Motivation
The most obvious solution for taking full advantage of CSIT is to add the test cases under the same repository with the functionalities that they are testing (instead of having them in separate centralized CSIT repository as we currently have). This would have the following benefits:
- CSIT could be triggered by any commit to the project repo
- CSIT tests the code (or specifically, docker images that have been built) from the committed branch
- CSIT could have a vote on the commit based on the result of the test run
- If the implementation changes require changes in CSIT tests (to cover new functionality or to pass in the first place), that could be handled within the same commit
- ideally, local verification would become less complex (no need to work between CSIT repo and project repo)
- No need of Integration to merge any changes related to your project
Issues
Given the fact that CSITs are currently a very colourful collection of various suites with different scopes and strategies, the transition of CSITs to project repositories is not necessarily trivial:
- CSIT suites that test components from multiple project repositories at the same time
- such CSIT tests may have to be separated using additional simulators, or
- project repository structures themselves may have to be reconsidered
- In any case, the division between the images under test and images that are just necessary dependencies should be clearly made and documented
- the images under test should be coming from the commit branch
- in the case of necessary dependencies it must be decided whether they should be provided as simulators or real components
- if provided as real components, they should be referred to with a fixed, unchanging version number and should be stable and mature enough to develop on
- ONAP's (unintentional?) practice of allowing the same versioned image to be changed might prove problematic
- if provided as real components, they should be referred to with a fixed, unchanging version number and should be stable and mature enough to develop on
- project-specific simulators are trivial case from dependency handling point of view, but if common simulators are in use, dependency considerations for them are the same as with real components
- Jenkins templates may have to be redesigned to support unified approach for triggering review branch-specific docker build, CSIT execution and voting chain as part of review verification
- What is the most elegant and effortless solution for ensuring that the docker images built from commit branch are taken in use in the CSIT (both in Jenkins and in local development environment)?
- CSITs would become blockers for merging code
- local pre-commit verification should be supported better by common CSIT tools
are all projects and suites mature enough to deal with that?
Technical details to be decided
- Should we keep separate docker build and CSIT jobs and just chain them into review verification, or should we try to incorporate docker building and CSIT execution into existing review jobs?
- Reusing existing jobs and chaining them would require some docker image tag tuning to make sure CSIT tests pick up the exact image that was produced by preceding docker build job
- Either way, JJB templates will probably have to be touched
- Should we still have common CSIT scripts (run-csit.sh etc) in CSIT repo and related procedures (setup, tests, teardown and related result collection) as the basis of project-specific test execution?
- Execution of CSIT tests and incorporating locally built test images should be made as easy as possible following common guidelines
- Setting up testing environment (specific project-specific dependencies should be handled by the setup scripts)
- Specific environment variables expected by the test suite (like GERRIT_BRANCH)
Project status and readiness at the end of Guilin
| Projects with CSIT | CSIT jobs | CSIT job triggers | Tested docker images and repositories | Docker build jobs that create the tested images | Dependencies | Comments |
|---|---|---|---|---|---|---|
| aaf (disabled) | aaf-frankfurt-csit-certservice Frankfurt branch only? The tests triggered from here do not even exist in master (nor are there any jobs for still remaining aafapi and sms-test-plan?) | aaf-certservice-frankfurt-merge-java - this job neither builds a docker image nor triggers another job that does it, so why trigger CSIT? | nexus3.onap.org:10001/onap/ org.onap.aaf.certservice.aaf-certservice-api:frankfurt-latest (hardcoded) The source repository is aaf/certservice | aaf-certservice-maven-docker-stage-frankfurt (note that aaf-certservice-release-verify and aaf-certservice-release-merge also build aaf-certservice-api image, but not with frankfurt-latest tag) | primekey/ejbca-ce:6.15.2.5 (hardcoded, real image used) | In maintenance mode, CSIT disabled: AAF-1208 - Getting issue details... STATUS |
| appc (disabled) | appc-master-csit-healthcheck | - | nexus3.onap.org:10001/onap/appc-image:1.7.2-SNAPSHOT-latest nexus3.onap.org:10001/onap/appc-cdt-image:1.7.2-SNAPSHOT-latest (hardcoded) docker-compose.yml as well as appc-image and appc-cdt-image come from appc/deployment | No job is producing these images at the moment! appc-deployment-master-docker-java-daily produces 1.8.0-SNAPSHOT-latest and appc-deployment-frankfurt-docker-java-daily is producing 1.7.3-SNAPSHOT-latest 1.7.2-SNAPSHOT-latest images have not been created since May 27 | mariadb:10.4.3 nexus3.onap.org:10001/onap/ccsdk-dgbuilder-image:0.6.0 (no longer produced, last update to Nexus on Aug 30 2019) nexus3.onap.org:10001/onap/ccsdk-ansible-server-image:0.4.4 (no longer produced, last update to Nexus on Jun 06 2019) (hardcoded, real images used) | In maintenance mode, CSIT disabled: APPC-1910 - Getting issue details... STATUS |
| ccsdk (temporarily disabled) | See CCSDK-2850 - Getting issue details... STATUS CCSDK-2851 - Getting issue details... STATUS CCSDK-2856 - Getting issue details... STATUS | |||||
| ccsdk-oran |
The source repository is ccsdk/oran | ccsdk-oran/job/ccsdk-oran-maven-merge-master and ccsdk-oran-maven-merge-guilin produce the tested images respectively (note that these are different images that the STAGING-tagged ones from triggering jobs!) | nexus3.onap.org:10003/onap/sdnc-image (2.1.0-STAGING-latest for master, nexus3.onap.org:10003/onap/sdnc-image:2.0.1-STAGING-latest for Guilin) nexus3.o-ran-sc.org:10004/o-ran-sc/nonrtric-controlpanel:2.0.0 | These tests are currently cloned from ORAN repository and they are essentially plain shell scripts that are just wrapped into Robot In Honolulu the test cases should be brought under ONAP repository and rewritten with Robot and ONAP CSIT scripts as much as possible | ||
| clamp | Invalid trigger jobs removed in CLAMP-950 - Getting issue details... STATUS | nexus3.onap.org:10001/onap/clamp-backend:4.1-STAGING-latest nexus3.onap.org:10001/onap/clamp-backend:4.1-STAGING-latest (hardcoded) The docker-compose.yml and source of the images is clamp | clamp-maven-docker-stage-elalto | mariadb:10.5.4 python:2-slim | The versions are a mess; master docker builds are producing 5.1 versions, frankfurt branch 5.0 versions, and el-alto 4.1 versions. However, CSIT master and frankfurt branches are testing versions 4.1 and CSIT does not even have el-alto branch? | |
| dcaegen2 | dcaegen2-services-son-handler-master-csit-testsuites dcaegen2-services-pmsh-master-csit-testsuite dcaegen2-services-bbs-event-processor-master-csit-bbs-testsuites dcaegen2-pmmapper-master-csit-pmmapper dcaegen2-master-csit-testsuites dcaegen2-master-csit-prh-testsuites dcaegen2-collectors-restconf-master-csit-testsuites dcaegen2-collectors-hv-ves-master-csit-testsuites dcaegen2-collectors-datafile-master-csit-ManagementInterface-suite | dcaegen2-services-son-handler-master-merge-java dcaegen2-services-pmsh-docker-merge-master dcaegen2-services-master-bbs-event-processor-merge-java (pmmapper has no trigger job that actually exists) dcaegen2-collectors-ves-master-merge-java dcaegen2-services-prh-master-merge-java dcaegen2-collectors-restconf-master-merge-java dcaegen2-collectors-hv-ves-master-merge-java (datafile CSITs have no trigger jobs that actually exist) Ticket raised about the non-existent trigger jobs: DCAEGEN2-2487 - Getting issue details... STATUS | nexus3.onap.org:10003/onap/org.onap.dcaegen2.services.son-handler:latest nexus3.onap.org:10001/onap/org.onap.dcaegen2.services.pmsh:latest nexus3.onap.org:10001/onap/org.onap.dcaegen2.services.components.bbs-event-processor:latest nexus3.onap.org:10001/onap/org.onap.dcaegen2.services.pm-mapper:latest nexus3.onap.org:10003/onap/org.onap.dcaegen2.collectors.ves.vescollector:latest nexus3.onap.org:10001/onap/org.onap.dcaegen2.services.prh.prh-app-server:latest nexus3.onap.org:10001/onap/org.onap.dcaegen2.collectors.restconfcollector:latest nexus3.onap.org:10001/onap/org.onap.dcaegen2.collectors.hv-ves.hv-collector-main:latest nexus3.onap.org:10001/onap/org.onap.dcaegen2.collectors.datafile.datafile-app-server:latest (all hardcoded in various places in the tests) TBD: most or all of these are in their own dedicated repositories, TBD: are the repo-image-CSIT relations straightforward 1:1 in all cases (or any case)? | Some of the triggering jobs produce the tested images, others do not produce any image; requires a deeper look | A lot of simulators used, TBD | |
| dmaap | dmaap-buscontroller-master-csit-ssl dmaap-buscontroller-master-csit-with_dr dmaap-buscontroller-master-csit-with_mr dmaap-datarouter-master-csit-ssl-dr-suite | dmaap-datarouter-maven-docker-stage-master Invalid trigger jobs removed in DMAAP-1501 - Getting issue details... STATUS | nexus3.onap.org:10001/onap/dmaap/buscontroller nexus3.onap.org:10001/onap/dmaap/datarouter-prov nexus3.onap.org:10001/onap/dmaap/datarouter-node nexus3.onap.org:10001/onap/dmaap/datarouter-subscriber (all hardcoded, version not given so latest by default) At least three different source repositories: dmaap/messagerouter/messageservice | There are no jobs producing onap/dmaap/buscontroller - the latest image in Nexus has been updated Thu Feb 21 2019 and belongs to Casablanca (?!) - ticket raised: DMAAP-1502 - Getting issue details... STATUS | TBD | It's unclear in some of these suites if datarouter and messagerouter are dependencies for buscontroller or components under test? We seem to have multi-repository problem here that needs to be solved somehow |
| multicloud | multicloud-master-csit-functionality1 | Ticket raised about invalid trigger jobs: MULTICLOUD-1231 - Getting issue details... STATUS | nexus3.onap.org:10001/onap/multicloud/vio:1.4.1 nexus3.onap.org:10001/onap/multicloud/framework:1.4.1 nexus3.onap.org:10001/onap/multicloud/openstack-starlingx nexus3.onap.org:10001/onap/multicloud/vio (all hardcoded, default latest for those that do not define version explicitly) | Latest in Nexus: onap/multicloud/vio:1.4.1 is from Sep 12 2019 onap/multicloud/framework:1.4.1 is from Sep 06 2019 ...these seem to be some early El-Alto versions? Ticker raised: MULTICLOUD-1229 - Getting issue details... STATUS https://jenkins.onap.org/view/multicloud/job/multicloud-openstack-maven-docker-stage-elalto produces version 1.4.3 of onap/multicloud/openstack-starlingx https://jenkins.onap.org/view/multicloud/job/multicloud-framework-maven-docker-stage-master produces version 1.6.0 (and also latest) of onap/multicloud/framework multicloud-openstack-maven-docker-stage-master produces version 1.5.6 (and also latest) of onap/multicloud/openstack-starlingx multicloud-openstack-vmware-maven-docker-stage-master produces version 1.4.2 (and also latest) of onap/multicloud/vio | - | |
| music (disabled) | music-master-verify-csit-music-distributed-kv-store-test-plan | music-distributed-kv-store-master-docker-golang-shell-daily | nexus3.onap.org:10001/onap/music/distributed-kv-store (hardcoded, latest by default) | music-distributed-kv-store-master-docker-golang-shell-daily | - | In maintenance mode, CSIT disabled: MUSIC-613 - Getting issue details... STATUS |
| oom-platform-cert-service | oom-platform-cert-service-master-merge-java (triggers two CSITs but does not produce or trigger docker image? Ticket raised: OOM-2594 - Getting issue details... STATUS ) | nexus3.onap.org:10001/onap/org.onap.oom.platform.cert-service.oom-certservice-api:latest nexus3.onap.org:10001/onap/org.onap.oom.platform.cert-service.oom-certservice-post-processor:latest (but started in Robot test and defined in Robot variable so it would be theoretically overridable) | oom-platform-cert-service-maven-docker-stage-master (for both) | primekey/ejbca-ce:6.15.2.5 | ||
| optf | Ticket raised for removing invalid triggers: OPTFRA-858 - Getting issue details... STATUS | nexus3.onap.org:10001/onap/optf-cmso-service:latest nexus3.onap.org:10001/onap/optf-cmso-dbinit:latest nexus3.onap.org:10001/onap/optf-cmso-topology:latest nexus3.onap.org:10001/onap/optf-cmso-ticketmgt:latest nexus3.onap.org:10001/onap/optf-cmso-optimizer:latest nexus3.onap.org:10001/onap/optf-cmso-robot:latest (!! The actual Robot tests run here in setup phase instead of as separate Robot tests from CSIT repo) nexus3.onap.org:10001/onap/optf-has:2.0.2-SNAPSHOT-latest nexus3.onap.org:10001/onap/optf-osdf:2.0.3-SNAPSHOT-latest Repositories: | optf-cmso-maven-docker-stage-master optf-has-maven-docker-stage-master optf-osdf-maven-docker-stage-master (word of warning: the frankfurt versions of the above jobs are also pushing the same images with "latest"! Probably worth a ticket) | |||
| policy | ||||||
| sdc (special case) | ||||||
| sdnc | ||||||
| so | ||||||
| usecases (disabled?) | ||||||
| vfc | ||||||
| vid | ||||||
| vfsdk |