feat(k8s): make default workspace PVC storage size configurable

The Kubernetes driver hardcoded the workspace PVC size to 2Gi. Add a
workspace_default_storage_size field to KubernetesComputeConfig so
operators can tune it via TOML config, Helm values, or the
OPENSHELL_K8S_WORKSPACE_DEFAULT_STORAGE_SIZE environment variable.

Author: sjenning

crates/openshell-driver-kubernetes/src/config.rs

@@ -7,6 +7,9 @@ use serde::{Deserialize, Serialize};
 /// Default Kubernetes namespace for sandbox resources.
 pub const DEFAULT_K8S_NAMESPACE: &str = "openshell";
 
+/// Default storage size for the workspace PVC.
+pub const DEFAULT_WORKSPACE_STORAGE_SIZE: &str = "2Gi";
+
 /// How the supervisor binary is delivered into sandbox pods.
 #[derive(Debug, Clone, Copy, PartialEq, Eq, Default, Serialize, Deserialize)]
 #[serde(rename_all = "kebab-case")]
@@ -64,6 +67,7 @@ pub struct KubernetesComputeConfig {
     pub client_tls_secret_name: String,
     pub host_gateway_ip: String,
     pub enable_user_namespaces: bool,
+    pub workspace_default_storage_size: String,
 }
 
 impl Default for KubernetesComputeConfig {
@@ -84,6 +88,7 @@ impl Default for KubernetesComputeConfig {
             client_tls_secret_name: String::new(),
             host_gateway_ip: String::new(),
             enable_user_namespaces: false,
+            workspace_default_storage_size: DEFAULT_WORKSPACE_STORAGE_SIZE.to_string(),
         }
     }
 }
@@ -94,3 +99,23 @@ fn default_sandbox_image() -> String {
         openshell_core::image::DEFAULT_COMMUNITY_REGISTRY
     )
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn default_workspace_storage_size_is_2gi() {
+        let cfg = KubernetesComputeConfig::default();
+        assert_eq!(cfg.workspace_default_storage_size, DEFAULT_WORKSPACE_STORAGE_SIZE);
+    }
+
+    #[test]
+    fn serde_override_workspace_storage_size() {
+        let json = serde_json::json!({
+            "workspace_default_storage_size": "10Gi"
+        });
+        let cfg: KubernetesComputeConfig = serde_json::from_value(json).unwrap();
+        assert_eq!(cfg.workspace_default_storage_size, "10Gi");
+    }
+}

crates/openshell-driver-kubernetes/src/driver.rs

@@ -3,7 +3,9 @@
 
 //! Kubernetes compute driver.
 
-use crate::config::{KubernetesComputeConfig, SupervisorSideloadMethod};
+use crate::config::{
+    DEFAULT_WORKSPACE_STORAGE_SIZE, KubernetesComputeConfig, SupervisorSideloadMethod,
+};
 use futures::{Stream, StreamExt, TryStreamExt};
 use k8s_openapi::api::core::v1::{Event as KubeEventObj, Node};
 use kube::api::{Api, ApiResource, DeleteParams, ListParams, PostParams};
@@ -106,8 +108,6 @@ const WORKSPACE_INIT_MOUNT_PATH: &str = "/workspace-pvc";
 /// Name of the init container that seeds the workspace PVC.
 const WORKSPACE_INIT_CONTAINER_NAME: &str = "workspace-init";
 
-/// Default storage request for the workspace PVC.
-const WORKSPACE_DEFAULT_STORAGE: &str = "2Gi";
 
 /// Sentinel file written by the init container after copying the image's
 /// `/sandbox` contents.  Subsequent pod starts skip the copy.
@@ -327,6 +327,7 @@ impl KubernetesComputeDriver {
             client_tls_secret_name: &self.config.client_tls_secret_name,
             host_gateway_ip: &self.config.host_gateway_ip,
             enable_user_namespaces: self.config.enable_user_namespaces,
+            workspace_default_storage_size: &self.config.workspace_default_storage_size,
         };
         obj.data = sandbox_to_k8s_spec(sandbox.spec.as_ref(), &params);
         let api = self.api();
@@ -1025,7 +1026,12 @@ fn apply_workspace_persistence(
 ///
 /// Provides a single PVC named "workspace" that backs the `/sandbox`
 /// directory.  The init container seeds it from the image on first use.
-fn default_workspace_volume_claim_templates() -> serde_json::Value {
+fn default_workspace_volume_claim_templates(storage_size: &str) -> serde_json::Value {
+    let size = if storage_size.is_empty() {
+        DEFAULT_WORKSPACE_STORAGE_SIZE
+    } else {
+        storage_size
+    };
     serde_json::json!([{
         "metadata": {
             "name": WORKSPACE_VOLUME_NAME
@@ -1034,7 +1040,7 @@ fn default_workspace_volume_claim_templates() -> serde_json::Value {
             "accessModes": ["ReadWriteOnce"],
             "resources": {
                 "requests": {
-                    "storage": WORKSPACE_DEFAULT_STORAGE
+                    "storage": size
                 }
             }
         }
@@ -1056,6 +1062,7 @@ struct SandboxPodParams<'a> {
     client_tls_secret_name: &'a str,
     host_gateway_ip: &'a str,
     enable_user_namespaces: bool,
+    workspace_default_storage_size: &'a str,
 }
 
 fn spec_pod_env(spec: Option<&SandboxSpec>) -> std::collections::HashMap<String, String> {
@@ -1112,7 +1119,7 @@ fn sandbox_to_k8s_spec(
     if inject_workspace {
         root.insert(
             "volumeClaimTemplates".to_string(),
-            default_workspace_volume_claim_templates(),
+            default_workspace_volume_claim_templates(params.workspace_default_storage_size),
         );
     }
 
@@ -2572,4 +2579,18 @@ mod tests {
         assert_eq!(tolerations[0]["operator"], "Exists");
         assert_eq!(tolerations[0]["effect"], "NoSchedule");
     }
+
+    #[test]
+    fn default_workspace_vct_uses_provided_storage_size() {
+        let vct = default_workspace_volume_claim_templates("5Gi");
+        let storage = &vct[0]["spec"]["resources"]["requests"]["storage"];
+        assert_eq!(storage, "5Gi");
+    }
+
+    #[test]
+    fn default_workspace_vct_falls_back_to_const_when_empty() {
+        let vct = default_workspace_volume_claim_templates("");
+        let storage = &vct[0]["spec"]["resources"]["requests"]["storage"];
+        assert_eq!(storage, DEFAULT_WORKSPACE_STORAGE_SIZE);
+    }
 }

crates/openshell-driver-kubernetes/src/lib.rs

@@ -5,6 +5,6 @@ pub mod config;
 pub mod driver;
 pub mod grpc;
 
-pub use config::{KubernetesComputeConfig, SupervisorSideloadMethod};
+pub use config::{DEFAULT_WORKSPACE_STORAGE_SIZE, KubernetesComputeConfig, SupervisorSideloadMethod};
 pub use driver::{KubernetesComputeDriver, KubernetesDriverError};
 pub use grpc::ComputeDriverService;

crates/openshell-driver-kubernetes/src/main.rs

@@ -93,6 +93,8 @@ async fn main() -> Result<()> {
         client_tls_secret_name: args.client_tls_secret_name.unwrap_or_default(),
         host_gateway_ip: args.host_gateway_ip.unwrap_or_default(),
         enable_user_namespaces: args.enable_user_namespaces,
+        workspace_default_storage_size: std::env::var("OPENSHELL_K8S_WORKSPACE_DEFAULT_STORAGE_SIZE")
+            .unwrap_or_else(|_| openshell_driver_kubernetes::DEFAULT_WORKSPACE_STORAGE_SIZE.to_string()),
     })
     .await
     .into_diagnostic()?;

crates/openshell-server/src/lib.rs

@@ -583,7 +583,10 @@ async fn build_compute_runtime(
 
     match driver {
         ComputeDriverKind::Kubernetes => {
-            let k8s = kubernetes_config_from_file(file)?;
+            let mut k8s = kubernetes_config_from_file(file)?;
+            if let Ok(size) = std::env::var("OPENSHELL_K8S_WORKSPACE_DEFAULT_STORAGE_SIZE") {
+                k8s.workspace_default_storage_size = size;
+            }
             ComputeRuntime::new_kubernetes(
                 k8s,
                 store,

deploy/helm/openshell/templates/gateway-config.yaml

@@ -90,6 +90,9 @@ data:
     {{- if .Values.server.sandboxImagePullPolicy }}
     image_pull_policy            = {{ .Values.server.sandboxImagePullPolicy | quote }}
     {{- end }}
+    {{- if .Values.server.workspaceDefaultStorageSize }}
+    workspace_default_storage_size = {{ .Values.server.workspaceDefaultStorageSize | quote }}
+    {{- end }}
     {{- if .Values.supervisor.image.pullPolicy }}
     supervisor_image_pull_policy = {{ .Values.supervisor.image.pullPolicy | quote }}
     {{- end }}

deploy/helm/openshell/values.yaml

@@ -95,6 +95,10 @@ server:
   # (Always for :latest, IfNotPresent otherwise).  Set to "Always" for dev
   # clusters so new images are picked up without manual eviction.
   sandboxImagePullPolicy: ""
+  # Default storage size for the workspace PVC in sandbox pods.
+  # Uses Kubernetes quantity syntax (e.g. "2Gi", "10Gi", "500Mi").
+  # Empty = built-in default (2Gi).
+  workspaceDefaultStorageSize: ""
   # gRPC endpoint sandboxes call back into the gateway. Leave empty to derive
   # it from the chart fullname, release namespace, service port, and
   # disableTls flag (i.e. <scheme>://<fullname>.<namespace>.svc.cluster.local:<port>).