fix(compute): recover Error-phase sandboxes on gateway startup

crates/openshell-core/src/driver_utils.rs

@@ -27,6 +27,18 @@ pub const LABEL_SANDBOX_NAME: &str = "openshell.ai/sandbox-name";
 /// Container/pod label carrying the sandbox namespace.
 pub const LABEL_SANDBOX_NAMESPACE: &str = "openshell.ai/sandbox-namespace";
 
+// ---------------------------------------------------------------------------
+// Sandbox condition reason strings set by compute drivers.
+// ---------------------------------------------------------------------------
+
+/// Ready-condition reason when a container exits on its own (e.g. SIGTERM
+/// from a machine restart, OOM kill, application crash).
+pub const CONDITION_EXITED: &str = "ContainerExited";
+
+/// Ready-condition reason when a container is explicitly stopped via the
+/// runtime API (e.g. `podman stop`, gateway-initiated shutdown).
+pub const CONDITION_STOPPED: &str = "ContainerStopped";
+
 // ---------------------------------------------------------------------------
 
 /// Path to the sandbox supervisor binary inside the container image.

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

@@ -2801,7 +2801,7 @@ fn container_ready_condition(
             ("False", "ContainerPaused", "Container is paused", false)
         }
         ContainerSummaryStateEnum::EXITED => {
-            ("False", "ContainerExited", "Container exited", false)
+            ("False", openshell_core::driver_utils::CONDITION_EXITED, "Container exited", false)
         }
         ContainerSummaryStateEnum::DEAD => ("False", "ContainerDead", "Container is dead", false),
     }

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

@@ -693,6 +693,25 @@ impl PodmanComputeDriver {
         }
     }
 
+    /// Resume an exited sandbox container. Returns `Ok(true)` if the
+    /// container exists and was (re)started, `Ok(false)` if it is gone.
+    pub async fn resume_sandbox(&self, sandbox_name: &str) -> Result<bool, ComputeDriverError> {
+        let name = container::container_name(sandbox_name);
+        let inspect = match self.client.inspect_container(&name).await {
+            Ok(i) => i,
+            Err(PodmanApiError::NotFound(_)) => return Ok(false),
+            Err(e) => return Err(ComputeDriverError::from(e)),
+        };
+        if inspect.state.running {
+            return Ok(true);
+        }
+        match self.client.start_container(&name).await {
+            Ok(()) => Ok(true),
+            Err(PodmanApiError::NotFound(_)) => Ok(false),
+            Err(e) => Err(ComputeDriverError::from(e)),
+        }
+    }
+
     /// Fetch a single sandbox by name.
     pub async fn get_sandbox(
         &self,

crates/openshell-driver-podman/src/watcher.rs

@@ -22,8 +22,7 @@ use tracing::{debug, info, warn};
 // Condition reason constants shared across event-building paths.
 const CONDITION_RUNNING: &str = "ContainerRunning";
 const CONDITION_STARTING: &str = "ContainerStarting";
-const CONDITION_EXITED: &str = "ContainerExited";
-const CONDITION_STOPPED: &str = "ContainerStopped";
+use openshell_core::driver_utils::{CONDITION_EXITED, CONDITION_STOPPED};
 
 pub type WatchStream =
     Pin<Box<dyn Stream<Item = Result<WatchSandboxesEvent, ComputeDriverError>> + Send>>;