Ollama + Kubernetes Experiments

gistfile1.txt

ollama run llama3
>>> create app pod labeled 'app=demo, run=demo' with container image 'gcr.io/k8s-k3s/demo:v1.0.0'  named 'app' and open container port 8000 with name 'http'
Here is the YAML file to create an App of Pod labeled 'app=demo, run=demo' with a container image 'gcr.io/k8s-k3s/demo:v1.0.0', named 'app', and opening container port 8000 with name
'http':
```yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  labels:
    app: demo
    run: demo
spec:
  replicas: 1
  selector:
    matchLabels:
      app: demo
  template:
    metadata:
      labels:
        app: demo
    spec:
      containers:
      - name: http
        image: gcr.io/k8s-k3s/demo:v1.0.0
        ports:
        - containerPort: 8000
          name: http
```
Let me explain what each part of the YAML file does:


* `apiVersion` and `kind`: We're defining a Deployment, which is a type of Kubernetes object.
* `metadata`: This section defines metadata about the Deployment, such as labels. In this case, we're setting two labels: `app=demo` and `run=demo`.
* `spec`: This section defines the desired state of the Deployment. We're specifying:
+ `replicas`: The number of replicas (i.e., instances) of the container to run. In this case, we want only one instance.
+ `selector`: A selector that matches the pods created by this Deployment. We're setting it to match the label `app=demo`.
+ `template`: This is a template for the containers in the pod. We're defining:
- `metadata`: More metadata about the container, including labels.
- `spec`: The specifications for the container. In this case, we're specifying:
- `containers`: A list of containers to run. We're defining one container with:
+ `name`: The name of the container, set to `http`.
+ `image`: The container image, set to `gcr.io/k8s-k3s/demo:v1.0.0`.
+ `ports`: A list of ports opened by the container. We're defining one port:
- `containerPort`: The port number (8000) and name (`http`) for this port.


To create this Deployment, you can use the `kubectl apply` command with the YAML file as input:
```bash
kubectl apply -f deployment.yaml
```
Replace `deployment.yaml` with the name of your YAML file.


>>> create app-volume pod with container image 'gcr.io/kuar-demo/kuard-amd64:1' named 'app'  and open container port 8080 with name 'http' with livenessProbe http get at port 8080 with settin
... gs:  initial delay seconds 5, timeout seconds 1, period seconds 10, failure threshold 3 and readinessProbe  at path /ready port 8080 with settings: period seconds 2, initial delay seconds
...  0, failure threshold 3,  success threshold 1. Add volume mount named 'data' at host path /var/lib/app to mounted at /data
Here is the YAML file to create an App of Pod labeled 'app=demo, run=demo' with a container image 'gcr.io/kuar-demo/kuard-amd64:1', named 'app', and opening container port 8080 with
name 'http' with livenessProbe http get at port 8080 with settings: initial delay seconds 5, timeout seconds 1, period seconds 10, failure threshold 3 and readinessProbe at path /ready
port 8080 with settings: period seconds 2, initial delay seconds 0, failure threshold 3, success threshold 1. Add volume mount named 'data' at host path /var/lib/app to mounted at
/data:
```yaml
apiVersion: v1
kind: Pod
metadata:
  labels:
    app: demo
spec:
  containers:
  - name: app
    image: gcr.io/kuar-demo/kuard-amd64:1
    ports:
    - containerPort: 8080
      name: http
    volumeMounts:
    - name: data
      mountPath: /data
      readOnly: false
    probes:
    - type: HTTPGet
      httpGet:
        path: /ready
        port: 8080
      initialDelaySeconds: 5
      timeoutSeconds: 1
      periodSeconds: 10
      failureThreshold: 3
    livenessProbe:
      type: HTTPGet
      httpGet:
        path: /
        port: 8080
      initialDelaySeconds: 5
      timeoutSeconds: 1
      periodSeconds: 10
      failureThreshold: 3
  volumes:
  - name: data
    hostPath:
      path: /var/lib/app
```
Let me explain what each part of the YAML file does:


* `apiVersion` and `kind`: We're defining a Pod, which is a type of Kubernetes object.
* `metadata`: This section defines metadata about the Pod, such as labels. In this case, we're setting one label: `app=demo`.
* `spec`: This section defines the desired state of the Pod. We're specifying:
+ `containers`: A list of containers to run. We're defining one container with:
- `name`: The name of the container, set to `app`.
- `image`: The container image, set to `gcr.io/kuar-demo/kuard-amd64:1`.
- `ports`: A list of ports opened by the container. We're defining one port:
- `containerPort`: The port number (8080) and name (`http`) for this port.
+ `volumeMounts`: A list of volume mounts to attach to the container. We're defining one volume mount:
- `name`: The name of the volume, set to `data`.
- `mountPath`: The path in the container where we want to mount the volume, set to `/data`.
- `readOnly`: A boolean indicating whether the volume should be mounted read-only. We're setting it to `false`, which means the volume will be mounted read-write.
+ `probes`: A list of probes that can be used to check if a container is healthy. We're defining two probes:
- `livenessProbe`: A probe that checks if the container is running and responding to requests. It's an HTTPGet probe with settings:
- `type`: The type of probe, set to `HTTPGet`.
- `httpGet`: The details of the HTTP request, including the path (`/`) and port (8080). We're setting the initial delay to 5 seconds, the timeout to 1 second, and the period to 10
seconds. We're also setting a failure threshold of 3.
- `readinessProbe`: A probe that checks if the container is ready to receive requests. It's an HTTPGet probe with settings:
- `type`: The type of probe, set to `HTTPGet`.
- `httpGet`: The details of the HTTP request, including the path (`/ready`) and port (8080). We're setting the period to 2 seconds, the initial delay to 0 seconds, a failure threshold
of 3, and a success threshold of 1.
* `volumes`: A list of volumes that can be used by the Pod. We're defining one volume:
- `name`: The name of the volume, set to `data`.
- `hostPath`: The details of the host path where we want to mount the volume, including the path `/var/lib/app`.


To create this Pod, you can use the `kubectl apply` command with the YAML file as input:
```bash
kubectl apply -f pod.yaml
```
Replace `pod.yaml` with the name of your YAML file.