From 716489eafd8d46645884f4d682d5cc439429602c Mon Sep 17 00:00:00 2001
From: Chris Lu <chris.lu@gmail.com>
Date: Tue, 16 Jun 2026 18:34:39 -0700
Subject: [PATCH] Deployment to Kubernetes: document DaemonSet + hostPath
 volume servers

---
 Deployment-to-Kubernetes-and-Minikube.md | 42 ++++++++++++++++++++++++
 1 file changed, 42 insertions(+)

diff --git a/Deployment-to-Kubernetes-and-Minikube.md b/Deployment-to-Kubernetes-and-Minikube.md
index 8ce1a85..314744d 100644
--- a/Deployment-to-Kubernetes-and-Minikube.md
+++ b/Deployment-to-Kubernetes-and-Minikube.md
@@ -58,6 +58,48 @@ spec:
 
 For more details, see the [IAM Support](https://github.com/seaweedfs/seaweedfs-operator/blob/master/IAM_SUPPORT.md) documentation in the operator repository.
 
+## Bare-metal volume servers (DaemonSet + node-local disks)
+
+By default the Operator runs volume servers as a `StatefulSet` backed by
+dynamically-provisioned `PersistentVolumeClaim`s. On bare-metal / on-prem
+clusters you often instead want **one volume server per node, writing directly
+to that node's physical disks**. Set `volume.kind: DaemonSet` and list the
+node-local directories in `volume.hostPath`:
+
+```yaml
+apiVersion: seaweed.seaweedfs.com/v1
+kind: Seaweed
+metadata:
+  name: seaweed1
+spec:
+  image: chrislusf/seaweedfs:latest
+  master:
+    replicas: 3
+  volume:
+    kind: DaemonSet        # one volume server per matching node (replicas is ignored)
+    hostPath:              # node-local data dirs, one per physical disk
+      - path: /mnt/disks/ssd0
+        maxVolumeCount: 100
+      - path: /mnt/disks/ssd1
+    nodeSelector:          # restrict to labelled storage nodes
+      seaweedfs.com/storage: "true"
+    tolerations:
+      - key: seaweedfs.com/storage
+        operator: Exists
+        effect: NoSchedule
+  filer:
+    replicas: 1
+```
+
+Each `hostPath` entry is mounted at `/data<index>` and passed to
+`weed volume -dir`, so a single server can span several disks; the optional
+`maxVolumeCount` caps volumes per directory. `hostPath` is required in
+DaemonSet mode (DaemonSets cannot use `volumeClaimTemplates`), but it also
+works with the default `StatefulSet` — pair it with node anti-affinity so two
+replicas never share a host directory. For rack/datacenter-aware placement
+across multiple volume groups, see
+[Topology Support](https://github.com/seaweedfs/seaweedfs-operator/blob/master/TOPOLOGY_SUPPORT.md).
+
 ## Declarative Buckets and IAM
 
 Besides provisioning the cluster itself, the Operator can manage S3 buckets,