Service Management and Orchestration (SMO)
With respect to OAM the SMO implements the O1-interface consumers. According to the O-RAN OAM Architecture and the O-RAN OAM Interface Specification, the SMO implements a NetConf Client for configuration and a HTTP/REST/VES server for receiving all kind of events in a VES format.
The setup contains an OpenDaylight based NetConf client and a VES Collector.
This docker-compose file starts a pre-configured, self-contained SDN-R solution for developer test or demo purposes
Identity ... representing an KeyCloak based identity service for centralized user management. Please note that the implementation does not support IPv6. Therefore, its own network is required called 'DMZ'.
SDN-R single node instance
... representing the NetConf consumer on the Service Management and Orchestration framework (SMO) for the O1 interface based on ODL-Silicon/ONAP-Istanbul
VES collector
... representing the VES (REST) provider at SMO for all kind of events.
DMaaP ... representing SMO DMaaP component, includes message-router
ONAP-Policy ... representing all the components of ONAP policy framework, in particular the policy-apex-pdp which executes the apex policies deployed in the framework when a certain event occurs.
Non-RT-RIC ... representing all the components of Non-RT-RIC, includes Non-RT-RIC Control Panel, Non-RT-RIC (Spring Cloud) Service Gateway, A1 Policy Management Services, Enrichment Data Coordinator, Non-RT-RIC App Catalogue, "Helloworld" O-RU Fronthaul Recovery use-case, Near-RT RIC A1 Simulator etc.
$ cat /etc/os-release | grep PRETTY_NAME PRETTY_NAME="Ubuntu 20.04.2 LTS" $ docker --version Docker version 20.10.2, build 20.10.2-0ubuntu1~20.04.2 $ docker-compose version docker-compose version 1.29.1, build c34c88b2 docker-py version: 5.0.0 CPython version: 3.7.10 OpenSSL version: OpenSSL 1.1.0l 10 Sep 2019 $ git --version git version 2.25.1
Please modify the /etc/hosts of your system.
: is the hostname of the system, where the browser is started
: is the IP address of the system where the solution will be deployed
For development purposes and may reference the same system.
$ cat /etc/hosts 127.0.0.1 localhost 127.0.1.1 <your-system> <deployment-system-ipv4> sdnc-web <your-system> <deployment-system-ipv4> identity <your-system>
├── network │ ├── .env │ ├── config.py │ ├── docker-compose.yml │ │ │ ├── ntsim-ng-o-du │ └── ntsim-ng-o-ru └── smo ├── common │ ├── .env │ ├── docker-compose.yml │ │ │ ├── dmaap │ ├── docker │ ├── identity │ ├── kafka │ └── zookeeper ├── non-rt-ric │ ├── docker-compose.yml │ │ │ ├── test │ ├── data │ └── config ├── oam │ ├── docker-compose.yml │ │ │ ├── sdnc-web │ ├── sdnr │ └── ves-collector └── onap-policy ├── docker-compose.yml │ ├── config └── wait_for_port.sh
nano smo/common/.env nano smo/non-rt-ric/.env nano smo/oam/.env nano network/.env
Please note that it is necessary to configure first the identity service, before starting further docker images.
The several docker-compose yml files must be started in the right order as listed below:
docker-compose -f smo/common/docker-compose.yml up -d python smo/common/identity/config.py
The python script configure the users within the identity service (keycloak). A system user (%USER) is also created with administration rights.
docker-compose -f smo/onap-policy/docker-compose.yml up -d docker-compose -f smo/oam/docker-compose.yml up -d docker-compose -f smo/non-rt-ric/docker-compose.yml up -d
In order to create/deploy the apex policy for O-RU closed loop recovery usecase, refer to the section named "Create/Deploy apex policy for O-RU & O-DU use case" in this page: https://wiki.o-ran-sc.org/pages/viewpage.action?pageId=35881325
Please note that the above instructions assume that the ToscaPolicy.json file with default config needs to be deployed. However, when there is a need to update the config (for example, to change the O-RU to O-DU mapping), a new ToscaPolicy.json file needs to be created. Refer to the section named "Workflow for updating the policy config" in this page: https://wiki.o-ran-sc.org/pages/viewpage.action?pageId=35881325
Please wait about 2min until all the service are up and running. If you see the login page (https://sdnc-web:8453) you are good to go and can start the (simulated) network.
Full instrucion on how to run Non-RT-RIC can be found in this page: https://wiki.o-ran-sc.org/display/RICNR/Release+D
When containers in Non-RT-RIC are all up, by default, there is no data running inside. Folder non-rt-ric/data/
contains serveral scripts to populate data into Non-RT-RIC for test & demo purpose.
bash prepareDmaapMsg.sh
script prepareDmaapMsg.sh
sends messages to dmaap message router, then Non-RT-RIC policy-agent service polls messages from dmaap, and creates policy instances accordingly.
bash preparePmsData.sh
script preparePmsData.sh
sends http requests to policy-agent service, and creates policy instances accordingly.
bash prepareEcsData.sh
script prepareEcsData.sh
sends http requests to ecs service, and creates data accordingly.
Afterwards, open webpage: http://localhost:8182/ Now we should see some data in the page.
docker-compose -f network/docker-compose.yml up -d
Usually the first ves:event gets lost. Please restart the O-DU docker container(s) to send a second ves:pnfRegistration.
docker-compose -f network/docker-compose.yml restart ntsim-ng-o-du-1122 python network/config.py
The python script configures the simulated O-DU and O-RU according to O-RAN hybrid architecture.
O-DU - NETCONF Call HOME and NETCONF notifications O-RU - ves:pnfRegistration and ves:fault, ves:heartbeat
'True' indicated that the settings through SDN-R to the NETCONF server were successful.
SDN-R reads the fault events from DMaaP and processes them. Finally the fault events are visible in ODLUX.
docker exec -it sdnr tail -f /opt/opendaylight/data/log/karaf.log
ssh karaf@localhost -p 8101
docker logs -f ves-collector
docker logs policy-apex-pdp
docker logs policy-agent docker logs oru-app docker logs ecs
'.env' file contains customizing parameters
https://sdnc-web:8453 User: admin // see .env file Password: Kp8bJ4SXszM0WXlhak3eHlcse2gAw84vaoGGmJvUy2U
In case of trouble, please update the commands with your customized '.env' file.
To stop all container please respect the following order
docker-compose -f network/docker-compose.yml down docker-compose -f smo/oam/docker-compose.yml down docker-compose -f smo/onap-policy/docker-compose.yml down docker-compose -f smo/non-rt-ric/docker-compose.yml down docker-compose -f smo/common/docker-compose.yml down
!!! be careful if other stopped containers are on the same system
docker system prune -a -f
In most cases the .env setting do not fit to the environment and need to be adjusted.
Please make sure that the network settings to not overlap with other networks.
The commands ...
docker ps -a docker-compose ps ## Prerequisites # python3, tmux, libtmux tmux new-session -n workspace -s integration # within tmux session python tmux-logging.py
... are your friends.