mirror of
https://github.com/netbirdio/docs.git
synced 2026-04-21 18:06:38 +00:00
Fix typos and enhance clarity in documentation (#471)
* Fix typos and enhance clarity in documentation Corrected typos and improved clarity in the documentation. * Update src/pages/use-cases/distributed-multi-cloud-ai-argocd-microk8s-vllm.mdx --------- Co-authored-by: Eduard Gert <kontakt@eduardgert.de>
This commit is contained in:
@@ -5,7 +5,7 @@ import {Note} from "@/components/mdx";
|
|||||||
We are building a distributed AI infrastructure project that connects GPU clusters across many different cloud providers using Kubernetes orchestration. This setup uses ArgoCD as a GitOps control plane to manage workloads across multiple MicroK8s clusters, with NetBird providing secure zero-trust networking between all components.
|
We are building a distributed AI infrastructure project that connects GPU clusters across many different cloud providers using Kubernetes orchestration. This setup uses ArgoCD as a GitOps control plane to manage workloads across multiple MicroK8s clusters, with NetBird providing secure zero-trust networking between all components.
|
||||||
|
|
||||||
<Note>
|
<Note>
|
||||||
This is document is activily being changed and tested. Please see 'Known Issues and Future Improvements' at the bottom of the page.
|
This document is actively being changed and tested. Please see 'Known Issues and Future Improvements' at the bottom of the page.
|
||||||
</Note>
|
</Note>
|
||||||
|
|
||||||
<p>
|
<p>
|
||||||
@@ -284,7 +284,7 @@ After the VM boots and MicroK8s is running, we need to register this cluster wit
|
|||||||
**Future Optimization Notes:**
|
**Future Optimization Notes:**
|
||||||
|
|
||||||
- Setup keys: Currently using one-shot keys per node; needs automation for scale
|
- Setup keys: Currently using one-shot keys per node; needs automation for scale
|
||||||
- Pre-baked images: Creating custom VM images with software pre-installed would speed up provisioning significantly
|
- Pre-baked images: Creating custom VM images, with software pre-installed would speed up provisioning significantly
|
||||||
|
|
||||||
### Create Service Account and RBAC
|
### Create Service Account and RBAC
|
||||||
|
|
||||||
@@ -333,7 +333,7 @@ microk8s.kubectl apply -f svc.yml
|
|||||||
- **Secret**: Automatically generated token for authentication
|
- **Secret**: Automatically generated token for authentication
|
||||||
- **ClusterRoleBinding**: Grants cluster-admin permissions to ArgoCD (full control)
|
- **ClusterRoleBinding**: Grants cluster-admin permissions to ArgoCD (full control)
|
||||||
|
|
||||||
*Note: In production, you'd use more restricted RBAC permissions based on the principle of least privilege.*
|
*Note: In production, you should use more restricted RBAC permissions based on the principle of least privilege.*
|
||||||
|
|
||||||
### Extract the Authentication Token
|
### Extract the Authentication Token
|
||||||
|
|
||||||
@@ -584,7 +584,7 @@ spec:
|
|||||||
- `{{server}}`: Cluster API endpoint
|
- `{{server}}`: Cluster API endpoint
|
||||||
- `{{path}}`: Directory path in Git (e.g., "apps/vllm")
|
- `{{path}}`: Directory path in Git (e.g., "apps/vllm")
|
||||||
- `{{path.basename}}`: Just the directory name (e.g., "vllm")
|
- `{{path.basename}}`: Just the directory name (e.g., "vllm")
|
||||||
3. **Resulting Applications**:
|
3. **Applications Created**:
|
||||||
- For each cluster with `megamesh: "true"` label
|
- For each cluster with `megamesh: "true"` label
|
||||||
- For each app directory in the repo
|
- For each app directory in the repo
|
||||||
- Creates an Application named like: `aws-germany-cluster-vllm`
|
- Creates an Application named like: `aws-germany-cluster-vllm`
|
||||||
|
|||||||
Reference in New Issue
Block a user