Files
LocalAI/core/http/endpoints/localai/system.go
T
localai-org-maint-botandEttore Di Giacinto c0b7e64973 feat(ui): show running models and host gauges on single-node installs (#12189)
On a single-node install nothing in Operate listed the models loaded on
this machine or let an admin stop one. The System page that did was
retired in #11548, and its replacements (the Nodes workbench) only work
in distributed mode. The Nodes page also mis-detected single-node mode:
the cluster routes are not registered there, so /api/nodes answers 404,
but only 503 was treated as "distributed off", which sent every
single-node install to the empty worker-registration card. The rail hid
the entry anyway.

Nodes route on a single node becomes "This machine":
- the Nodes page's VRAM / RAM / CPU / models-disk gauges, fed from this
  host by mapping /api/resources onto the worker heartbeat fields
- a memory bar splitting host RAM by running model
- a running-models table (backend, RSS, CPU share, uptime, PID) with
  search, sorting, logs and a confirmed Stop
- the distributed setup behind an "Add machines" button

The Operate overview gains a "Running now" preview (heaviest five, with
Stop) on single node and a pointer to Nodes > Running models on a
cluster. The rail shows "This machine" in Runtime with a running count.

Backend, additive only:
- /system: each loaded model carries a `process` block (pid, rss_bytes,
  memory_percent, cpu_percent, started_at). A sampler keeps one gopsutil
  handle per PID so CPU is the share since the previous poll rather than
  the lifetime average; it is omitted on the first reading.
- /api/resources: host `cpu` and models-path `disk`, the same readings
  workers send in their heartbeat.

Also fixes the fleet tables widening the page on phones: the headers'
absolutely positioned sr-only labels escaped the scroll wrapper.

Assisted-by: Claude:claude-opus-5 [Playwright]

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Co-authored-by: Ettore Di Giacinto <mudler@localai.io>
2026-09-21 19:12:35 +02:00

72 lines
2.1 KiB
Go

package localai
import (
"strconv"
"github.com/labstack/echo/v4"
"github.com/mudler/LocalAI/core/config"
"github.com/mudler/LocalAI/core/schema"
"github.com/mudler/LocalAI/core/services/monitoring"
"github.com/mudler/LocalAI/pkg/model"
)
// SystemInformations returns the system informations
// @Summary Show the LocalAI instance information
// @Tags monitoring
// @Success 200 {object} schema.SystemInformationResponse "Response"
// @Router /system [get]
func SystemInformations(cl *config.ModelConfigLoader, ml *model.ModelLoader, appConfig *config.ApplicationConfig, sampler *monitoring.LocalProcessSampler) echo.HandlerFunc {
return func(c echo.Context) error {
availableBackends := []string{}
loadedModels := ml.ListLoadedModels()
for b := range appConfig.ExternalGRPCBackends {
availableBackends = append(availableBackends, b)
}
for b := range ml.GetAllExternalBackends(nil) {
availableBackends = append(availableBackends, b)
}
sysmodels := []schema.SysInfoModel{}
live := map[int32]struct{}{}
for _, m := range loadedModels {
entry := schema.SysInfoModel{ID: m.ID}
// The loader tracks only the ID. Which engine is serving a model is
// the first thing an operator wants beside its name, and it is one
// config lookup away.
if cfg, ok := cl.GetModelConfig(m.ID); ok {
entry.Backend = cfg.Backend
}
if pid, ok := localPID(m); ok && sampler != nil {
live[pid] = struct{}{}
if proc, err := sampler.Sample(pid); err == nil {
entry.Process = proc
}
}
sysmodels = append(sysmodels, entry)
}
if sampler != nil {
sampler.Retain(live)
}
return c.JSON(200,
schema.SystemInformationResponse{
Backends: availableBackends,
Models: sysmodels,
},
)
}
}
// localPID is the PID of the backend process this host started for m. A model
// served by a remote worker, or through an external gRPC address, has none.
func localPID(m *model.Model) (int32, bool) {
p := m.Process()
if p == nil {
return 0, false
}
pid, err := strconv.ParseInt(p.CurrentPID(), 10, 32)
if err != nil || pid <= 0 {
return 0, false
}
return int32(pid), true
}