Add the KVM health check: reachability and storage pool per host

Each configured host is checked concurrently over the pooled SSH+libvirt
connection: libvirt version and hostname prove the daemon answers, then
the configured pool must exist and be running, with its free space
reported. Dial failures are now tagged with their stage (SSH material,
TCP, SSH handshake, libvirt connect) by a dialError that keeps the
message and unwraps to the cause, so classification is unchanged and the
check can say why a host is unreachable. A pool verdict is decided after
the call returns so it never drops a working connection.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
This commit is contained in:
2026-09-11 20:17:35 +00:00
co-authored by Claude Opus 5
parent 1962365f44
commit 04dd2c0284
9 changed files with 694 additions and 5 deletions
+54
View File
@@ -28,6 +28,12 @@ type fakeAPI struct {
// uploadHook, when set, sees every StorageVolUpload before the fake
// reads the stream; it can return an error to simulate a broken upload.
uploadHook func(vol string, r io.Reader) error
// hostname, libVersion, poolCapacity and poolStates answer the health
// check's RPCs. A pool missing from poolStates is running.
hostname string
libVersion uint64
poolCapacity uint64
poolStates map[string]libvirt.StoragePoolState
}
type fakeVol struct {
@@ -53,6 +59,11 @@ func newFakeAPI(pools ...string) *fakeAPI {
domains: map[string]*fakeDomain{},
fail: map[string]error{},
failOnce: map[string]error{},
hostname: "fake-host",
libVersion: 10002000,
poolCapacity: 100 << 30,
poolStates: map[string]libvirt.StoragePoolState{},
}
for _, p := range pools {
f.pools[p] = map[string]*fakeVol{}
@@ -409,3 +420,46 @@ func (f *fakeAPI) StorageVolDelete(vol libvirt.StorageVol, _ libvirt.StorageVolD
delete(f.pools[vol.Pool], vol.Name)
return nil
}
// --- host ---
func (f *fakeAPI) ConnectGetLibVersion() (uint64, error) {
f.mu.Lock()
defer f.mu.Unlock()
if err := f.enter("ConnectGetLibVersion"); err != nil {
return 0, err
}
return f.libVersion, nil
}
func (f *fakeAPI) ConnectGetHostname() (string, error) {
f.mu.Lock()
defer f.mu.Unlock()
if err := f.enter("ConnectGetHostname"); err != nil {
return "", err
}
return f.hostname, nil
}
// StoragePoolGetInfo reports the pool's state and poolCapacity minus the
// volumes' capacities as available space.
func (f *fakeAPI) StoragePoolGetInfo(pool libvirt.StoragePool) (uint8, uint64, uint64, uint64, error) {
f.mu.Lock()
defer f.mu.Unlock()
if err := f.enter("StoragePoolGetInfo"); err != nil {
return 0, 0, 0, 0, err
}
vols, ok := f.pools[pool.Name]
if !ok {
return 0, 0, 0, 0, lvError(libvirt.ErrNoStoragePool, "no pool %s", pool.Name)
}
var used uint64
for _, v := range vols {
used += uint64(v.capacity)
}
state := libvirt.StoragePoolRunning
if s, ok := f.poolStates[pool.Name]; ok {
state = s
}
return uint8(state), f.poolCapacity, used, f.poolCapacity - used, nil
}