From bf73efdb7fde95149d5d1c9a9bf0b66968245f1c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Maximilian=20H=C3=BCter?= Date: Mon, 16 Feb 2026 19:50:44 +0100 Subject: [PATCH] develop/deploy: change crun/youki wording in intro MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The previous wording stated crun/youki would bypass Linux Containers if WASM was used. This is not true. Signed-off-by: Maximilian Hüter --- docs/develop/deploy/intro.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/docs/develop/deploy/intro.md b/docs/develop/deploy/intro.md index 3673d28d0..9702ce5b3 100644 --- a/docs/develop/deploy/intro.md +++ b/docs/develop/deploy/intro.md @@ -10,7 +10,7 @@ Several options exist to manage WASM apps as “containers” under Kubernetes. ## With containerd-shim -**Option #1:** is to use a containerd-shim to start WASM "containers" via runwasi. Basically containerd could look at the image’s target platform. It uses runwasi if the image is wasm32 and runc if it is x86 / arm. Currently, Docker and Microsoft prefer this approach, which is also the basis for the [Docker + WASM preview](https://www.docker.com/blog/docker-wasm-technical-preview/). Based on containerd +Option #1 is to use a containerd-shim to start WASM "containers" via runwasi. Basically containerd could look at the image’s target platform. It uses runwasi if the image is wasm32 and runc if it is x86 / arm. Currently, Docker and Microsoft prefer this approach, which is also the basis for the [Docker + WASM preview](https://www.docker.com/blog/docker-wasm-technical-preview/). Based on containerd The following image shows how it works. @@ -18,11 +18,11 @@ The following image shows how it works. ## With crun -Option #2 is to use an OCI runtime called crun (the C version of runc, mainly supported by Red Hat). crun decides whether an OCI image is wasm or Linux based on image annotations. If the image is annotated as wasm32, crun will bypass Linux container setup and just use WasmEdge to run it. Based on crun, we can get the entire Kubernetes stack CRI-O, containerd, Podman, kind, micro k8s, k8s etc to work with WASM images. +Option #2 is to use an OCI runtime called crun (the C version of runc, mainly supported by Red Hat). crun decides whether an OCI image is wasm or Linux based on image annotations. If the image is annotated as wasm32, crun will create a container and use an embedded WasmEdge to run it. Based on crun, we can get the entire Kubernetes stack CRI-O, containerd, Podman, kind, micro k8s, k8s etc to work with WASM images. ## With youki -Option #3 is to use an OCI runtime called youki (the Rust version of runc). Like crun, youki decides whether an OCI image is wasm or Linux based on image annotations. If the image is annotated as wasm32, youki will bypass Linux container setup and just use WasmEdge to run it. Based on youki, we can get the entire Kubernetes stack CRI-O, containerd, Podman, kind, micro k8s, k8s etc to work with WASM images. +Option #3 is to use an OCI runtime called youki (the Rust version of runc). Like crun, youki decides whether an OCI image is wasm or Linux based on image annotations. If the image is annotated as wasm32, youki will create a container and use an embedded WasmEdge to run it. Based on youki, we can get the entire Kubernetes stack CRI-O, containerd, Podman, kind, micro k8s, k8s etc to work with WASM images. Both crun and youki could use the following image to show how it works.