Status: Closed (View Workflow)
Affects Version/s: 1.10.0, 1.11.0
ONOS in a docker container with physical accton switches. ONOS is not clustered. docker host is ubuntu.
Using a CORD setup (a bit customized as it is not being deployed by CORD deployer).
ONOS being deployed under k8s, but this same behavior has been seen with onos just as a native container.
onos is started with the following apps
onos> apps -s -a
- 4 org.onosproject.netcfglinksprovider 1.10.1.SNAPSHOT Network Config Link Provider
- 5 org.onosproject.fibinstaller 1.10.1.SNAPSHOT FIB installer app
- 6 org.onosproject.cpr 1.10.1.SNAPSHOT Control plane redirect
- 7 org.onosproject.fpm 1.10.1.SNAPSHOT FIB Push Manager (FPM) route reciever
- 8 org.onosproject.vrouter 1.10.1.SNAPSHOT Virtual Router App
- 9 org.onosproject.drivers 1.10.1.SNAPSHOT Default device drivers
- 10 org.onosproject.optical-model 1.10.1.SNAPSHOT Optical information model
- 11 org.onosproject.openflow-base 1.10.1.SNAPSHOT OpenFlow Provider
- 12 org.onosproject.netcfghostprovider 1.10.1.SNAPSHOT Network Config Host Provider
- 13 org.onosproject.hostprovider 1.10.1.SNAPSHOT Host Location Provider
- 14 org.onosproject.segmentrouting 1.10.1.SNAPSHOT Segment Routing App
connect 4 switches (two leaf, two spine) to onos, onos does not have any configuration pushed.
execute `ping -I fabric 126.96.36.199` where the fabric interface is the interface used to connect the switch to ONOS. the host does not show up in onos.
push the segment routing config to onos (w/devies, hosts, ports) and then ping will show host in onos.