fix(core): pin supervisor image tag to gateway version for all drivers

benoitf · benoitf · commit fd09c75f9a78 · 2026-07-01T20:41:15.000+02:00
The Podman and Kubernetes drivers defaulted the supervisor image to `:latest` via DEFAULT_SUPERVISOR_IMAGE, while the Docker driver already resolved a version-pinned tag. Extract the tag resolution logic into openshell-core so all three drivers use the same OPENSHELL_IMAGE_TAG > IMAGE_TAG > CARGO_PKG_VERSION priority chain. Closes #2068 Signed-off-by: Florent Benoit <fbenoit@redhat.com>
diff --git a/crates/openshell-core/src/config.rs b/crates/openshell-core/src/config.rs
@@ -37,8 +37,50 @@ pub const DEFAULT_DOCKER_NETWORK_NAME: &str = "openshell-docker";
 /// Default domain used for browser-facing sandbox service URLs.
 pub const DEFAULT_SERVICE_ROUTING_DOMAIN: &str = "openshell.localhost";
 
-/// Default OCI image for the openshell-sandbox supervisor binary.
-pub const DEFAULT_SUPERVISOR_IMAGE: &str = "ghcr.io/nvidia/openshell/supervisor:latest";
+/// Default OCI repository for the supervisor image (no tag).
+pub const DEFAULT_SUPERVISOR_IMAGE_REPO: &str = "ghcr.io/nvidia/openshell/supervisor";
+
+/// Return the default supervisor image reference with a version-pinned tag.
+#[must_use]
+pub fn default_supervisor_image() -> String {
+    format!(
+        "{DEFAULT_SUPERVISOR_IMAGE_REPO}:{}",
+        default_supervisor_image_tag()
+    )
+}
+
+fn default_supervisor_image_tag() -> String {
+    resolve_supervisor_image_tag(
+        option_env!("OPENSHELL_IMAGE_TAG"),
+        option_env!("IMAGE_TAG"),
+        env!("CARGO_PKG_VERSION"),
+    )
+}
+
+/// Resolve the supervisor image tag from build-time inputs.
+///
+/// Priority: `OPENSHELL_IMAGE_TAG` > `IMAGE_TAG` > `CARGO_PKG_VERSION`.
+/// Falls back to `"dev"` when the Cargo version is empty or `"0.0.0"`.
+/// Replaces `+` with `-` for OCI tag compatibility.
+#[must_use]
+pub fn resolve_supervisor_image_tag(
+    openshell_image_tag: Option<&str>,
+    image_tag: Option<&str>,
+    cargo_pkg_version: &str,
+) -> String {
+    let tag = openshell_image_tag
+        .filter(|tag| !tag.is_empty())
+        .or_else(|| image_tag.filter(|tag| !tag.is_empty()))
+        .unwrap_or_else(|| {
+            if cargo_pkg_version.is_empty() || cargo_pkg_version == "0.0.0" {
+                "dev"
+            } else {
+                cargo_pkg_version
+            }
+        });
+
+    tag.replace('+', "-")
+}
 
 /// CDI device identifier for requesting all NVIDIA GPUs.
 pub const CDI_GPU_DEVICE_ALL: &str = "nvidia.com/gpu=all";
@@ -1064,4 +1106,48 @@ mod tests {
             }
         }
     }
+
+    #[test]
+    fn supervisor_image_tag_prefers_explicit_build_tags() {
+        use super::resolve_supervisor_image_tag;
+        assert_eq!(
+            resolve_supervisor_image_tag(Some("1.2.3"), Some("sha"), "0.0.0"),
+            "1.2.3",
+        );
+        assert_eq!(
+            resolve_supervisor_image_tag(None, Some("sha"), "0.0.0"),
+            "sha",
+        );
+        assert_eq!(resolve_supervisor_image_tag(None, None, "1.2.3"), "1.2.3",);
+        assert_eq!(
+            resolve_supervisor_image_tag(Some(""), Some(""), "0.0.0"),
+            "dev",
+        );
+        assert_eq!(
+            resolve_supervisor_image_tag(Some("latest"), None, "1.2.3"),
+            "latest",
+        );
+    }
+
+    #[test]
+    fn supervisor_image_tag_sanitizes_build_metadata_for_oci() {
+        use super::resolve_supervisor_image_tag;
+        assert_eq!(
+            resolve_supervisor_image_tag(None, None, "0.0.37-dev.156+g1d3b741ee"),
+            "0.0.37-dev.156-g1d3b741ee",
+        );
+        assert_eq!(
+            resolve_supervisor_image_tag(Some("0.0.37-dev.156+g1d3b741ee"), None, "0.0.0"),
+            "0.0.37-dev.156-g1d3b741ee",
+        );
+    }
+
+    #[test]
+    fn default_supervisor_image_is_version_pinned() {
+        use super::default_supervisor_image;
+        let image = default_supervisor_image();
+        assert!(image.starts_with("ghcr.io/nvidia/openshell/supervisor:"));
+        let tag = image.rsplit_once(':').unwrap().1;
+        assert!(!tag.is_empty());
+    }
 }
diff --git a/crates/openshell-driver-docker/src/lib.rs b/crates/openshell-driver-docker/src/lib.rs
@@ -78,55 +78,24 @@ const HOST_OPENSHELL_INTERNAL: &str = "host.openshell.internal";
 const HOST_DOCKER_INTERNAL: &str = "host.docker.internal";
 const DOCKER_NETWORK_DRIVER: &str = "bridge";
 
-/// Default image holding the Linux `openshell-sandbox` binary. The gateway
-/// pulls this image and extracts the binary to a host-side cache when no
-/// explicit `supervisor_bin`, configured `supervisor_image`, sibling binary,
-/// or local build is available.
-const DEFAULT_DOCKER_SUPERVISOR_IMAGE_REPO: &str = "ghcr.io/nvidia/openshell/supervisor";
-
 /// Return the default `ghcr.io/nvidia/openshell/supervisor:<tag>` reference
 /// used when no supervisor binary override is provided.
 pub fn default_docker_supervisor_image() -> String {
     format!(
-        "{DEFAULT_DOCKER_SUPERVISOR_IMAGE_REPO}:{}",
+        "{}:{}",
+        openshell_core::config::DEFAULT_SUPERVISOR_IMAGE_REPO,
         default_docker_supervisor_image_tag()
     )
 }
 
-/// Image tag baked in at compile time to pair the gateway with a matching
-/// supervisor image.
-///
-/// Build pipelines pass `OPENSHELL_IMAGE_TAG` explicitly. The `IMAGE_TAG`
-/// fallback covers image build wrappers that already tag the gateway and
-/// supervisor together. Standalone release binaries also patch the Cargo
-/// package version, so use it when it has been set to a real release value.
 fn default_docker_supervisor_image_tag() -> String {
-    resolve_default_docker_supervisor_image_tag(
+    openshell_core::config::resolve_supervisor_image_tag(
         option_env!("OPENSHELL_IMAGE_TAG"),
         option_env!("IMAGE_TAG"),
         env!("CARGO_PKG_VERSION"),
     )
 }
 
-fn resolve_default_docker_supervisor_image_tag(
-    openshell_image_tag: Option<&'static str>,
-    image_tag: Option<&'static str>,
-    cargo_pkg_version: &'static str,
-) -> String {
-    let tag = openshell_image_tag
-        .filter(|tag| !tag.is_empty())
-        .or_else(|| image_tag.filter(|tag| !tag.is_empty()))
-        .unwrap_or_else(|| {
-            if cargo_pkg_version.is_empty() || cargo_pkg_version == "0.0.0" {
-                "dev"
-            } else {
-                cargo_pkg_version
-            }
-        });
-
-    tag.replace('+', "-")
-}
-
 /// Queried by the Docker driver to decide when a sandbox's supervisor
 /// relay is live. Implementations return `true` once a sandbox has an
 /// active `ConnectSupervisor` session registered.
diff --git a/crates/openshell-driver-docker/src/tests.rs b/crates/openshell-driver-docker/src/tests.rs
@@ -1897,36 +1897,31 @@ fn configured_supervisor_image_takes_precedence_over_local_binaries() {
 
 #[test]
 fn docker_supervisor_image_tag_prefers_explicit_build_tags() {
+    use openshell_core::config::resolve_supervisor_image_tag;
     assert_eq!(
-        resolve_default_docker_supervisor_image_tag(Some("1.2.3"), Some("sha"), "0.0.0"),
+        resolve_supervisor_image_tag(Some("1.2.3"), Some("sha"), "0.0.0"),
         "1.2.3",
     );
     assert_eq!(
-        resolve_default_docker_supervisor_image_tag(None, Some("sha"), "0.0.0"),
+        resolve_supervisor_image_tag(None, Some("sha"), "0.0.0"),
         "sha",
     );
+    assert_eq!(resolve_supervisor_image_tag(None, None, "1.2.3"), "1.2.3",);
     assert_eq!(
-        resolve_default_docker_supervisor_image_tag(None, None, "1.2.3"),
-        "1.2.3",
-    );
-    assert_eq!(
-        resolve_default_docker_supervisor_image_tag(Some(""), Some(""), "0.0.0"),
+        resolve_supervisor_image_tag(Some(""), Some(""), "0.0.0"),
         "dev",
     );
 }
 
 #[test]
 fn docker_supervisor_image_tag_sanitizes_build_metadata_for_docker() {
+    use openshell_core::config::resolve_supervisor_image_tag;
     assert_eq!(
-        resolve_default_docker_supervisor_image_tag(None, None, "0.0.37-dev.156+g1d3b741ee"),
+        resolve_supervisor_image_tag(None, None, "0.0.37-dev.156+g1d3b741ee"),
         "0.0.37-dev.156-g1d3b741ee",
     );
     assert_eq!(
-        resolve_default_docker_supervisor_image_tag(
-            Some("0.0.37-dev.156+g1d3b741ee"),
-            None,
-            "0.0.0",
-        ),
+        resolve_supervisor_image_tag(Some("0.0.37-dev.156+g1d3b741ee"), None, "0.0.0",),
         "0.0.37-dev.156-g1d3b741ee",
     );
 }
diff --git a/crates/openshell-driver-kubernetes/src/config.rs b/crates/openshell-driver-kubernetes/src/config.rs
@@ -1,7 +1,7 @@
 // SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 // SPDX-License-Identifier: Apache-2.0
 
-use openshell_core::config::DEFAULT_SUPERVISOR_IMAGE;
+use openshell_core::config;
 use serde::{Deserialize, Deserializer, Serialize};
 use std::path::Path;
 use std::str::FromStr;
@@ -263,7 +263,7 @@ impl Default for KubernetesComputeConfig {
             // is Podman vocabulary and is not a valid Kubernetes value.
             image_pull_policy: String::new(),
             image_pull_secrets: Vec::new(),
-            supervisor_image: DEFAULT_SUPERVISOR_IMAGE.to_string(),
+            supervisor_image: config::default_supervisor_image(),
             supervisor_image_pull_policy: String::new(),
             supervisor_sideload_method: SupervisorSideloadMethod::default(),
             supervisor_topology: SupervisorTopology::default(),
diff --git a/crates/openshell-driver-kubernetes/src/main.rs b/crates/openshell-driver-kubernetes/src/main.rs
@@ -121,7 +121,7 @@ async fn main() -> Result<()> {
         image_pull_secrets: args.sandbox_image_pull_secrets,
         supervisor_image: args
             .supervisor_image
-            .unwrap_or_else(|| openshell_core::config::DEFAULT_SUPERVISOR_IMAGE.to_string()),
+            .unwrap_or_else(openshell_core::config::default_supervisor_image),
         supervisor_image_pull_policy: args.supervisor_image_pull_policy.unwrap_or_default(),
         supervisor_sideload_method: args.supervisor_sideload_method,
         supervisor_topology: args.supervisor_topology,
diff --git a/crates/openshell-driver-podman/src/config.rs b/crates/openshell-driver-podman/src/config.rs
@@ -1,7 +1,7 @@
 // SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 // SPDX-License-Identifier: Apache-2.0
 
-use openshell_core::config::{DEFAULT_STOP_TIMEOUT_SECS, DEFAULT_SUPERVISOR_IMAGE};
+use openshell_core::config::DEFAULT_STOP_TIMEOUT_SECS;
 use std::net::IpAddr;
 use std::path::PathBuf;
 use std::str::FromStr;
@@ -255,7 +255,7 @@ impl Default for PodmanComputeConfig {
             network_name: DEFAULT_NETWORK_NAME.to_string(),
             host_gateway_ip: Self::default_host_gateway_ip(),
             stop_timeout_secs: DEFAULT_STOP_TIMEOUT_SECS,
-            supervisor_image: DEFAULT_SUPERVISOR_IMAGE.to_string(),
+            supervisor_image: openshell_core::config::default_supervisor_image(),
             guest_tls_ca: None,
             guest_tls_cert: None,
             guest_tls_key: None,
diff --git a/crates/openshell-driver-podman/src/container.rs b/crates/openshell-driver-podman/src/container.rs
@@ -1798,7 +1798,7 @@ mod tests {
         let vol = &image_volumes[0];
         assert_eq!(
             vol["source"].as_str(),
-            Some("ghcr.io/nvidia/openshell/supervisor:latest"),
+            Some(openshell_core::config::default_supervisor_image().as_str()),
             "image volume source should be the supervisor image"
         );
         assert_eq!(
@@ -1884,8 +1884,9 @@ mod tests {
         let image_volumes = spec["image_volumes"]
             .as_array()
             .expect("image_volumes should be an array");
+        let expected_supervisor = openshell_core::config::default_supervisor_image();
         assert!(image_volumes.iter().any(|volume| {
-            volume["source"].as_str() == Some("ghcr.io/nvidia/openshell/supervisor:latest")
+            volume["source"].as_str() == Some(expected_supervisor.as_str())
                 && volume["destination"].as_str() == Some("/opt/openshell/bin")
         }));
         assert!(image_volumes.iter().any(|volume| {
diff --git a/crates/openshell-driver-podman/src/main.rs b/crates/openshell-driver-podman/src/main.rs
@@ -90,7 +90,7 @@ struct Args {
 
     /// OCI image containing the openshell-sandbox supervisor binary.
     #[arg(long, env = "OPENSHELL_SUPERVISOR_IMAGE")]
-    supervisor_image: String,
+    supervisor_image: Option<String>,
 
     /// Host path to the CA certificate for sandbox mTLS.
     #[arg(long, env = "OPENSHELL_PODMAN_TLS_CA")]
@@ -130,7 +130,9 @@ async fn main() -> Result<()> {
         sandbox_ssh_socket_path: args.sandbox_ssh_socket_path,
         network_name: args.network_name,
         stop_timeout_secs: args.stop_timeout,
-        supervisor_image: args.supervisor_image,
+        supervisor_image: args
+            .supervisor_image
+            .unwrap_or_else(openshell_core::config::default_supervisor_image),
         guest_tls_ca: args.podman_tls_ca,
         guest_tls_cert: args.podman_tls_cert,
         guest_tls_key: args.podman_tls_key,
diff --git a/docs/reference/gateway-config.mdx b/docs/reference/gateway-config.mdx
@@ -88,7 +88,8 @@ disable_tls = false
 # Shared driver defaults. These inherit into [openshell.drivers.<name>] tables
 # when the driver-specific table does not override them.
 default_image           = "ghcr.io/nvidia/openshell/sandbox:latest"
-supervisor_image        = "ghcr.io/nvidia/openshell/supervisor:latest"
+# Defaults to the gateway version; override to pin a specific build.
+# supervisor_image      = "ghcr.io/nvidia/openshell/supervisor:<version>"
 client_tls_secret_name  = "openshell-client-tls"
 service_account_name    = "openshell-sandbox"
 host_gateway_ip         = "10.0.0.1"
@@ -172,7 +173,8 @@ service_account_name           = "openshell-sandbox"
 default_image                  = "ghcr.io/nvidia/openshell/sandbox:latest"
 image_pull_policy              = "IfNotPresent"
 image_pull_secrets             = ["regcred"]
-supervisor_image               = "ghcr.io/nvidia/openshell/supervisor:latest"
+# Defaults to the gateway version; override to pin a specific build.
+# supervisor_image             = "ghcr.io/nvidia/openshell/supervisor:<version>"
 supervisor_image_pull_policy   = "IfNotPresent"
 # Use the image volume on Kubernetes >= 1.35 (GA in 1.36); switch to "init-container"
 # on older clusters or where the ImageVolume feature gate is off.
@@ -222,7 +224,8 @@ grpc_endpoint      = "https://host.openshell.internal:17670"
 # Skip the image-pull-and-extract step by pointing at a locally built binary.
 supervisor_bin     = "/usr/local/libexec/openshell/openshell-sandbox"
 # When supervisor_bin is omitted, Docker extracts /openshell-sandbox from this image.
-supervisor_image   = "ghcr.io/nvidia/openshell/supervisor:latest"
+# Defaults to the gateway version; override to pin a specific build.
+# supervisor_image = "ghcr.io/nvidia/openshell/supervisor:<version>"
 guest_tls_ca       = "/etc/openshell/certs/ca.pem"
 guest_tls_cert     = "/etc/openshell/certs/client.pem"
 guest_tls_key      = "/etc/openshell/certs/client-key.pem"
@@ -264,7 +267,8 @@ network_name            = "openshell"
 # host_gateway_ip       = "192.168.127.254"
 sandbox_ssh_socket_path = "/run/openshell/ssh.sock"
 stop_timeout_secs       = 10
-supervisor_image        = "ghcr.io/nvidia/openshell/supervisor:latest"
+# Defaults to the gateway version; override to pin a specific build.
+# supervisor_image      = "ghcr.io/nvidia/openshell/supervisor:<version>"
 guest_tls_ca            = "/etc/openshell/certs/ca.pem"
 guest_tls_cert          = "/etc/openshell/certs/client.pem"
 guest_tls_key           = "/etc/openshell/certs/client-key.pem"
