You are viewing an old version of this page. View the current version.
Compare with Current
View Page History
« Previous
Version 2
Next »
Date
Minutes
- Pragash and Yuvaraja presenting slides
- k8s to TF object mapping
- Namespace → single project or shared project
- Pod → VM
- Service, Ingress, Network Policy each have their own analogs in TF
- A single TF controller can manage multiple different types of clusters
- Non-nested, nested (cluster inside of a cluster), standalone
- Namespaces with TF & k8s
- By default in k8s, all pods can talk to all other pods in the cluster
- Can also isolate them via namespace
- Each namespace maps to a Project in k8s
- Pod gets the IP from a controller…
- For every namespace in the cluster, create a virtual network
- Start with two by default (pod network and service network)
- To isolate things with namespaces, can create isolated networks accordingly
- Example of how this looks in the TF GUI
- Syed: When associated networks to a pod with a CRD, how do you ensure that the container in the network gets the appropriate route?
- Multi interfaces in the pod
- Create the networks in the CRD
- Queue manager then uses annotations to give things to the controller, that will set up the correct interfaces for the container
- Syed: But how do you set the routing within the container?
- To steer the package inside the container, add the static routes (manual process)
- Plan to enhance this in the future
- Pod creation workflow
Action items