Add unikernel runtime: run services as Nanos VMs under QEMU/KVM

Implements the hypervisor design's Phase 1: a second runtime.Runtime
backend (QEMU) that runs each service component as a Nanos unikernel VM
instead of a podman container, selected per-component via a new
runtime = "unikernel" service-def field.

- internal/runtime/qemu.go: QEMURuntime. Pull extracts the ELF from the
  OCI image; Run does `ops build` + boots qemu-system-x86_64 with KVM,
  user-mode net port-forwards, QMP control socket and serial console log;
  Stop/Remove/Inspect/List/Logs map onto VM lifecycle + state dir.
- proto/registry/servicedef: add runtime, memory_mb, vcpus fields
  (registry migration 5).
- agent: holds both runtimes; runtimeFor() selects per component;
  listAllContainers() merges containers + VMs so drift/status see both.
  Unikernel runtime auto-enables on nodes with /dev/kvm + ops.

Validated end-to-end on straylight: a test service deploys via
`mcp deploy --direct`, boots as a Nanos unikernel, serves HTTP through
the agent port-forward, and reports running via `mcp status`/`mcp logs`.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Kyle Isom
2026-06-11 00:54:49 -07:00
parent 3b08caaa0a
commit d56f224359
30 changed files with 949 additions and 152 deletions

View File

@@ -22,7 +22,7 @@ func (a *Agent) Recover(ctx context.Context) error {
}
// Get the list of currently running containers from podman.
running, err := a.Runtime.List(ctx)
running, err := a.listAllContainers(ctx)
if err != nil {
a.Logger.Warn("cannot list containers, assuming none running", "err", err)
running = nil
@@ -67,18 +67,22 @@ func (a *Agent) Recover(ctx context.Context) error {
"image", comp.Image,
)
rt := a.runtimeFor(comp.Runtime)
// Remove any stale container with the same name.
_ = a.Runtime.Remove(ctx, containerName)
_ = rt.Remove(ctx, containerName)
// Build the container spec from the registry.
spec := runtime.ContainerSpec{
Name: containerName,
Image: comp.Image,
Network: comp.Network,
User: comp.UserSpec,
Restart: comp.Restart,
Volumes: comp.Volumes,
Cmd: comp.Cmd,
Name: containerName,
Image: comp.Image,
Network: comp.Network,
User: comp.UserSpec,
Restart: comp.Restart,
Volumes: comp.Volumes,
Cmd: comp.Cmd,
MemoryMB: comp.MemoryMB,
VCPUs: comp.VCPUs,
}
// Allocate ports from routes if the component has routes.
@@ -95,7 +99,7 @@ func (a *Agent) Recover(ctx context.Context) error {
spec.Ports = comp.Ports
}
if err := a.Runtime.Run(ctx, spec); err != nil {
if err := rt.Run(ctx, spec); err != nil {
a.Logger.Error("recover container failed",
"container", containerName,
"err", err,