Files
llmgw/internal/control/reconciler/replica.go
T
Ilker ManapandClaude Opus 5 7b4acd9284
CI / test (push) Successful in 1m23s
A replica missing a shard is not ready
Restarting the control plane over a running two-node replica adopted half of
it. Every shard it still had was healthy, so it was reported Ready — with one
shard of two and no endpoint at all, because the missing one was rank 0. The
router was handed a replica nothing could route to.

The cause is that a replica had no idea how many shards it should have. It
could tell "every shard I have is ready" and nothing else, which is the same
sentence whether or not half of it is gone. It now carries the count from its
profile, set at placement and again at adoption, and a replica short of shards
fails rather than pretending.

Onboarding also wrote no engine paths, so a node onboarded by the script could
not find llama-server and failed every shard placed on it with "executable
file not found". The script now takes them, and says so when none are given —
a node with no engine paths can run the sim engine and nothing else.

The cluster now runs both at once: two-node simulated replicas on each fabric,
and real llama.cpp inference on the CPU, which is a better test bed than
either alone.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 12:43:51 +02:00

192 lines
5.2 KiB
Go

// Package reconciler drives the cluster toward the catalog's desired state
// (plan §6.3).
package reconciler
import (
"fmt"
"slices"
"time"
"gitea.mynodes.xyz/ilker/llmgw/internal/agent/supervisor"
)
// ReplicaState is where a replica stands.
type ReplicaState int
const (
// Placing means it has been scheduled but no shard has started.
Placing ReplicaState = iota
// Starting means shards are launching or loading.
Starting
// Ready means it serves traffic.
Ready
// Draining means it takes no new requests and will be stopped.
Draining
// Stopped means it was shut down on purpose.
Stopped
// Failed means it died or never became ready.
Failed
)
func (s ReplicaState) String() string {
switch s {
case Placing:
return "placing"
case Starting:
return "starting"
case Ready:
return "ready"
case Draining:
return "draining"
case Stopped:
return "stopped"
case Failed:
return "failed"
default:
return "unknown"
}
}
// Terminal reports whether the replica has finished for good.
func (s ReplicaState) Terminal() bool { return s == Stopped || s == Failed }
// Counts reports whether the replica should count toward the desired total.
// Draining replicas do not: they are on the way out, and counting them would
// stop a replacement from being started.
func (s ReplicaState) Counts() bool { return s == Placing || s == Starting || s == Ready }
// Shard is one node's part of a replica.
type Shard struct {
ID string
NodeID string
Rank int
GPUIndices []int
HCAs []string
HTTPPort int
State supervisor.State
Err string
}
// Leader reports whether this shard serves HTTP for the replica.
func (s Shard) Leader() bool { return s.Rank == 0 }
// Replica is one running copy of a model. It is the unit the router sees.
type Replica struct {
ID string
Model string
Profile string
// Fabric is set only for multi-node replicas.
Fabric string
Pool string
Shards []Shard
State ReplicaState
// Endpoint is rank 0's base URL.
Endpoint string
// Weight is the profile's capacity weight, passed to the router.
Weight float64
// WantShards is how many shards this replica should have, from its profile.
//
// Without it a replica cannot tell "every shard I have is ready" from
// "every shard I have left is ready". A control plane that adopted half of
// a two-node replica reported it Ready with one shard and no endpoint, and
// the router was handed a replica nothing could route to.
WantShards int
CreatedAt time.Time
ReadyAt time.Time
// DrainingAt is when the replica stopped taking new work, so the wait for a
// replacement can be bounded.
DrainingAt time.Time
// Attempt counts how many times this model has failed to come up in a row,
// carried across rebuilds so the backoff does not reset on every attempt.
Attempt int
Err string
}
// NodeIDs lists the nodes this replica occupies.
func (r Replica) NodeIDs() []string {
out := make([]string, 0, len(r.Shards))
for _, s := range r.Shards {
out = append(out, s.NodeID)
}
slices.Sort(out)
return out
}
// OnNode reports whether any shard runs on a node.
func (r Replica) OnNode(id string) bool {
return slices.ContainsFunc(r.Shards, func(s Shard) bool { return s.NodeID == id })
}
// deriveState recomputes the replica's state from its shards.
//
// A replica is only as healthy as its worst shard: one failed shard makes the
// whole thing useless, because the others are waiting on collectives that will
// never complete. That is why there is no path here that keeps a replica
// partially alive (plan §5.4).
// hasLiveShard reports whether any shard has actually got somewhere.
//
// Starting and Pending are not evidence of anything: they are what a replica
// looks like both while an engine loads a model and when nothing was ever
// started at all. Loading means the port is open, which only a real process
// does.
func (r *Replica) hasLiveShard() bool {
for _, s := range r.Shards {
if s.State == supervisor.Ready || s.State == supervisor.Loading {
return true
}
}
return false
}
func (r *Replica) deriveState() {
if r.State == Draining || r.State == Stopped {
return
}
if len(r.Shards) == 0 {
r.State = Placing
return
}
ready := 0
for _, s := range r.Shards {
switch s.State {
case supervisor.Failed:
r.State = Failed
if r.Err == "" {
r.Err = fmt.Sprintf("shard %s failed: %s", s.ID, s.Err)
}
return
case supervisor.Stopped:
// A shard stopping on its own, while the replica was not draining,
// is a failure of the replica.
r.State = Failed
if r.Err == "" {
r.Err = fmt.Sprintf("shard %s stopped unexpectedly", s.ID)
}
return
case supervisor.Ready:
ready++
}
}
// Both conditions. A replica missing a shard is not ready however healthy
// the shards it still has look — and rank 0's absence in particular means
// there is no endpoint to route to at all.
if ready == len(r.Shards) && (r.WantShards == 0 || len(r.Shards) == r.WantShards) {
if r.State != Ready {
r.ReadyAt = time.Now()
}
r.State = Ready
return
}
if r.WantShards > 0 && len(r.Shards) < r.WantShards {
r.State = Failed
if r.Err == "" {
r.Err = fmt.Sprintf("replica has %d of %d shards", len(r.Shards), r.WantShards)
}
return
}
r.State = Starting
}