demo

gistfile1.txt

Demo


https://github.com/syntasso/kratix/tree/main/demo
Done
* Download raw images

Pre-demo TODO
* Set up screen to be legible
* Run setup script
    * Ensure you’re logged in to Lastpass CLI
    * Kind load all images
    * Install Kratix
    * Config platform cluster
    * Creating Slack secret
    * 

In demo
You’ll be sharing screen!
* Just in case: ensure you’re logged in to Lastpass CLI
* Go into ‘focus’ mode to turn off notifications
* Put Slack, Chrome, Terminal into one screen
* Ensure fonts are legible
* Quit unnecessary applications

Reminders:
* You need to create not apply the Promise
* cd into /app-as-a-service
* Install Promise
* Promise overview
    * Encapsulates what Colin spoke about: deploying an app using Knative, Postgres, Redis
    * Deals with some business context by 
        * What configurations are required for the services
        * Notifying via Slack when a request comes in and an app gets deployed
    * The Promise we’ve defined is App as a Service
        * We’ll see this a few times
    * High level
        * How do teams request this Promise?
        * What does the system need to do fulfil the Promise?
        * What extra configuration is required for the business?
* Under the hood
        * Platform: teach the API about the new resource
        * Worker: prepare for when an instance is requested
* It
    * Installed the Promise
        * k get promises
    * Installed the high-level API for the application teams, as a crd
        * We can see that with k get crds | grep app
* Make the resource request
    * When we installed the Promise, the API we defined in document got installed 
    * We can see that with k get crds | grep app
    * We can see the slack message to show that the request has been made
    * We can see pipelines running
* Switch context to worker
    * We can see the pods coming up
    * This is server-less so the TODO app might not be running- that’s okay
    * Lets setup port-forwarding
    * Make a request to http://todo.default.local.gd:8081/
* Summarise
* Hand back to Colin for slides on PCI clusters
    * Delete the existing resource request while no ones looking to simplify the next part of the demo
* PCI compliant part
    * Show `resource-request-cc.yaml` and how credCadData:true is set, talk about business logic required by that
    * Make the resource request
    * See nothing happens on the worker cluster
    * Show existing clusters `kubectl get clusters —show-labels`
    * Label the cluster `kubectl label cluster worker-cluster-1 pci=true`
    * See the pod get scheduled onto worker cluster now that its compliant
    * Summarise
* Hand back to colin