osm_host_subnet_length
This variable specifies the size of the per host subnet allocated for pod IPs by OpenShift Container Platform SDN. Defaults to 9 which means that a subnet of size /23 is allocated to each host; for example, given the default 10.128.0.0/14 cluster network, this will allocate 10.128.0.0/23, 10.128.2.0/23, 10.128.4.0/23, and so on. This cannot be re-configured after deployment.
osm_cluster_network_cidr
This variable overrides the SDN cluster network CIDR block. This is the network from which pod IPs are assigned. Specify a private block that does not conflict with existing network blocks in your infrastructure to which pods, nodes, or the master might require access. Defaults to 10.128.0.0/14 and cannot be arbitrarily re-configured after deployment, although certain changes to it can be made in the SDN master configuration.
openshift_portal_net
This variable configures the subnet in which services will be created in the OpenShift Container Platform SDN. Specify a private block that does not conflict with any existing network blocks in your infrastructure to which pods, nodes, or the master might require access to, or the installation will fail. Defaults to 172.30.0.0/16, and cannot be re-configured after deployment. If changing from the default, avoid 172.17.0.0/16, which the docker0 network bridge uses by default, or modify the docker0 network.
Ex:
[OSE3:vars]
osm_cluster_network_cidr=10.128.0.0/14
openshift_portal_net=10.172.0.0/16
osm_host_subnet_length=9Max pods per node = (2 ^ osm_host_subnet_length) - 2
Max IPs available for PODs = 2 ^ (32 - osm_cluster_network_cidr) - 2
Max amount of nodes = (2 ^ (32 - osm_cluster_network_cidr) - 2) / ((2 ^ osm_host_subnet_length) - 2 )
On Python (using default values):
osm_cluster_network_cidr = 14
osm_host_subnet_length = 9
maxnodes = (pow(2,(32-osm_cluster_network_cidr))-2) / (pow(2,osm_host_subnet_length)-2)
print(maxnodes)Using /14 for cluster and /9 for nodes (510 pods/node):
>>> (pow(2,(32-14))-2) / (pow(2,9)-2)
514
Using /14 for cluster and /8 for nodes (254 pods/node):
>>> (pow(2,(32-14))-2) / (pow(2,8)-2)
1032
Using /16 for cluster and /8 for nodes (254 pods/node):
>>> (pow(2,(32-16))-2) / (pow(2,8)-2)
258
Using /16 for cluster and /9 for nodes (510 pods/node):
>>> (pow(2,(32-16))-2) / (pow(2,9)-2)
128
Many of the names are old (even for 3.11). They should be replaced for the more current equivalent terms (pod network, cluster network...)
But what is more important: Max nodes number calculation is wrong.
The right formula would be: 2^(32-
host_subnet_length-pod_network_prefix_length).In the example, this would be:
2^(32-14-9)=2^9=512You can double-check this easily with the
ipcalccommand available in Fedora, like this:This one will show you the networks. If you just want to count them, try