Verify that the Multus networking in configured in your Fusion Data Foundation cluster.
About this task
Based on your network configuration choices, the Fusion Data Foundation operator will do one of the following:
- If only a single NetworkAttachmentDefinition (for example,
ocs-public-cluster)
was selected for the Public Network Interface, then the traffic between the application pods and the
Fusion Data Foundation cluster will happen on this network.
Additionally the cluster will be self configured to also use this network for the replication and
rebalancing traffic between OSDs.
- If both NetworkAttachmentDefinitions (for example,
ocs-public and
ocs-cluster) were selected for the Public Network Interface and the Cluster Network
Interface respectively during the Storage Cluster installation, then client storage traffic will be
on the public network and cluster network for the replication and rebalancing traffic between
OSDs.
To verify the network configuration is correct, complete the following:
Procedure
-
In the Red Hat® OpenShift® Container Platform web console, go to
.
- From the Action menu, select Edit Storage
System.
- In the YAML tab, search for
network in the
spec section and ensure the configuration is correct for your network interface
choices.
This example is for separating the client storage traffic from the storage replication
traffic.
Sample output:
[..]
spec:
[..]
network:
ipFamily: IPv4
provider: multus
selectors:
cluster: openshift-storage/ocs-cluster
public: openshift-storage/ocs-public
[..]
To verify the network configuration is correct using the command line interface, run the
following commands:
oc get storagecluster ocs-storagecluster \
-n openshift-storage \
-o=jsonpath='{.spec.network}{"\n"}'
Sample
output:
{"ipFamily":"IPv4","provider":"multus","selectors":{"cluster":"openshift-storage/ocs-cluster","public":"openshift-storage/ocs-public"}}
- Confirm the OSD pods are using correct network.
In the openshift-storage namespace use one of the OSD pods to verify the pod has
connectivity to the correct networks.
Note: Only the OSD pods will connect to both Multus public and cluster networks if both are created.
All other OCS pods will connect to the Multus public network.
This example is for separating the client storage traffic from the storage replication
traffic.
oc get -n openshift-storage $(oc get pods -n openshift-storage -o name -l app=rook-ceph-osd | grep 'osd-0') -o=jsonpath='{.metadata.annotations.k8s\.v1\.cni\.cncf\.io/network-status}{"\n"}'
Sample output:
[{
"name": "openshift-sdn",
"interface": "eth0",
"ips": [
"10.129.2.30"
],
"default": true,
"dns": {}
},{
"name": "openshift-storage/ocs-cluster",
"interface": "net1",
"ips": [
"192.168.2.1"
],
"mac": "e2:04:c6:81:52:f1",
"dns": {}
},{
"name": "openshift-storage/ocs-public",
"interface": "net2",
"ips": [
"192.168.1.1"
],
"mac": "ee:a0:b6:a4:07:94",
"dns": {}
}]
To confirm the OSD pods are using correct network using the command line interface, run the
following command (requires the jq
utility):
oc get -n openshift-storage $(oc get pods -n openshift-storage -o name -l app=rook-ceph-osd | grep 'osd-0') -o=jsonpath='{.metadata.annotations.k8s\.v1\.cni\.cncf\.io/network-status}{"\n"}' | jq -r '.[].name'
Sample output:
openshift-sdn
openshift-storage/ocs-cluster
openshift-storage/ocs-public