Fix #1050: Handle XPC connection loss to update container state on ex…

[Ronitsabhaya75]

…ternal termination

Type of Change

• Bug fix
• New feature
• Breaking change
• Documentation update

Motivation and Context

Resolves #1050.

When containers are killed externally (via kill or launchctl bootout), the XPC connection loss causes the wait handler to throw an .interrupted error. This prevents cleanup and leaves containers stuck in running state.

Fix: Modified the wait handler in ContainersService.swift to catch .interrupted errors and treat them as container exits (code 137), allowing normal cleanup to proceed.

let waitFunc: ExitMonitor.WaitHandler = {
    log.info("registering container \(id) with exit monitor")
    do {
        let code = try await client.wait(id)
        log.info("container \(id) finished in exit monitor, exit code \(code)")
        return code
    } catch let err as ContainerizationError where err.code == .interrupted {
        // XPC connection was lost (e.g., container killed externally).
        // Treat this as a container exit so cleanup can proceed.
        log.warning("container \(id) XPC connection lost, treating as external termination")
        return ExitStatus(exitCode: 137)  // SIGKILL
    }
}

Testing

• Tested locally
• Added/updated tests
• Added/updated docs