Auteur/autrice : zigmax

Microsoft MVP Azure 2020-2021 !

[English version below]

Bonjour, 

C’est avec grand plaisir que je vous informe que je viens d’être renouvelé comme Microsoft MVP Azure !

Je tiens tout particulièrement à remercier les équipes produits de chez Microsoft, les communautés Microsoft de Québec, Toronto, ainsi que Betsy, et l’ensemble des personnes avec qui j’ai pu échanger au cours de cette dernière année.

Au plaisir d’échanger avec vous prochainement lors d’un meetup ou d’une conférence !

Maxime.


Hi! 

I have the pleasure to have been rewarded as a Microsoft MVP Azure !

I would like to take time to say « Thank you » to the Microsoft Product team, Microsoft Community (Quebec, Toronto), Betsy, Rochelle and all the people I have met this year. 

Do not hesitate to contact me, I hope to meet you during a next meetup or conference!

Maxime.

AKS | Containerd

Hi,

In this article, I would like to share with you how we can create an AKS cluster with Containerd.

Containerd is an OCI compliant core container runtime designed to be embedded into larger systems. It provides the minimum set of functionality to execute containers and manages images on a node. It was initiated by Docker Inc. and donated to CNCF in March of 2017.

A container runtime is software that executes containers and manages container images on a node. The runtime helps abstract away sys-calls or operating system (OS) specific functionality to run containers on Linux or Windows. Today AKS is using Moby (upstream docker) as its container runtime.

With a containerd-based node and node pools, instead of talking to the dockershim, the kubelet will talk directly to containerd via the CRI (container runtime interface) plugin, removing extra hops on the flow when compared to the Docker CRI implementation. As such, you’ll see better pod startup latency and less resource (CPU and memory) usage.

By using containerd for AKS nodes, pod startup latency improves and node resource consumption by the container runtime decreases. These improvements are enabled by this new architecture where kubelet talks directly to containerd through the CRI plugin while in Moby/docker architecture kubelet would talk to the dockershim and docker engine before reaching containerd, thus having extra hops on the flow.

# - Requirements
az extension add --name aks-preview 
az extension list

az feature register --name UseCustomizedContainerRuntime --namespace Microsoft.ContainerService 
az feature register --name UseCustomizedUbuntuPreview --namespace Microsoft.ContainerService

az feature list -o table --query "[?contains(name, 'Microsoft.ContainerService/UseCustomizedContainerRuntime')].{Name:name,State:properties.state}" 
az feature list -o table --query "[?contains(name, 'Microsoft.ContainerService/UseCustomizedUbuntuPreview')].{Name:name,State:properties.state}"

az provider register --namespace Microsoft.ContainerService

# - Ressource Group + AKS Cluster creation
az group create --name aksmaxime --location eastus

az aks create --name aksclustermax --resource-group aksmaxime --aks-custom-headers CustomizedUbuntu=aks-ubuntu-1804,ContainerRuntime=containerd

az aks get-credentials --resource-group aksmaxime --name aksclustermax --overwrite-existing

kubectl get nodes -o wide

Maxime.