// Code generated by protoc-gen-go. DO NOT EDIT. // versions: // protoc-gen-go v1.36.12 // protoc (unknown) // source: llmgw/v1/agent.proto // Package llmgw.v1 is the protocol between a node agent and the control plane // (plan §5.6). // // The agent dials out and holds a bidirectional stream open. That direction is // deliberate: GPU nodes sit behind firewalls and NAT far more often than a // control plane does, and an outbound connection needs no inbound rule on every // node in the fleet. package llmgwv1 import ( protoreflect "google.golang.org/protobuf/reflect/protoreflect" protoimpl "google.golang.org/protobuf/runtime/protoimpl" reflect "reflect" sync "sync" unsafe "unsafe" ) const ( // Verify that this generated code is sufficiently up-to-date. _ = protoimpl.EnforceVersion(20 - protoimpl.MinVersion) // Verify that runtime/protoimpl is sufficiently up-to-date. _ = protoimpl.EnforceVersion(protoimpl.MaxVersion - 20) ) type ShardStatus_State int32 const ( ShardStatus_STATE_UNSPECIFIED ShardStatus_State = 0 ShardStatus_STATE_PENDING ShardStatus_State = 1 ShardStatus_STATE_STARTING ShardStatus_State = 2 ShardStatus_STATE_LOADING ShardStatus_State = 3 ShardStatus_STATE_READY ShardStatus_State = 4 ShardStatus_STATE_DRAINING ShardStatus_State = 5 ShardStatus_STATE_STOPPED ShardStatus_State = 6 ShardStatus_STATE_FAILED ShardStatus_State = 7 ) // Enum value maps for ShardStatus_State. var ( ShardStatus_State_name = map[int32]string{ 0: "STATE_UNSPECIFIED", 1: "STATE_PENDING", 2: "STATE_STARTING", 3: "STATE_LOADING", 4: "STATE_READY", 5: "STATE_DRAINING", 6: "STATE_STOPPED", 7: "STATE_FAILED", } ShardStatus_State_value = map[string]int32{ "STATE_UNSPECIFIED": 0, "STATE_PENDING": 1, "STATE_STARTING": 2, "STATE_LOADING": 3, "STATE_READY": 4, "STATE_DRAINING": 5, "STATE_STOPPED": 6, "STATE_FAILED": 7, } ) func (x ShardStatus_State) Enum() *ShardStatus_State { p := new(ShardStatus_State) *p = x return p } func (x ShardStatus_State) String() string { return protoimpl.X.EnumStringOf(x.Descriptor(), protoreflect.EnumNumber(x)) } func (ShardStatus_State) Descriptor() protoreflect.EnumDescriptor { return file_llmgw_v1_agent_proto_enumTypes[0].Descriptor() } func (ShardStatus_State) Type() protoreflect.EnumType { return &file_llmgw_v1_agent_proto_enumTypes[0] } func (x ShardStatus_State) Number() protoreflect.EnumNumber { return protoreflect.EnumNumber(x) } // Deprecated: Use ShardStatus_State.Descriptor instead. func (ShardStatus_State) EnumDescriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{27, 0} } // AgentMessage is anything a node tells the control plane. type AgentMessage struct { state protoimpl.MessageState `protogen:"open.v1"` // Types that are valid to be assigned to Msg: // // *AgentMessage_Register // *AgentMessage_Heartbeat // *AgentMessage_Shard // *AgentMessage_Logs // *AgentMessage_FabricScan // *AgentMessage_FabricTest // *AgentMessage_Upgrade // *AgentMessage_KvEvents // *AgentMessage_TrainStatus // *AgentMessage_TrainArtifact Msg isAgentMessage_Msg `protobuf_oneof:"msg"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *AgentMessage) Reset() { *x = AgentMessage{} mi := &file_llmgw_v1_agent_proto_msgTypes[0] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *AgentMessage) String() string { return protoimpl.X.MessageStringOf(x) } func (*AgentMessage) ProtoMessage() {} func (x *AgentMessage) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[0] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use AgentMessage.ProtoReflect.Descriptor instead. func (*AgentMessage) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{0} } func (x *AgentMessage) GetMsg() isAgentMessage_Msg { if x != nil { return x.Msg } return nil } func (x *AgentMessage) GetRegister() *Register { if x != nil { if x, ok := x.Msg.(*AgentMessage_Register); ok { return x.Register } } return nil } func (x *AgentMessage) GetHeartbeat() *Heartbeat { if x != nil { if x, ok := x.Msg.(*AgentMessage_Heartbeat); ok { return x.Heartbeat } } return nil } func (x *AgentMessage) GetShard() *ShardStatus { if x != nil { if x, ok := x.Msg.(*AgentMessage_Shard); ok { return x.Shard } } return nil } func (x *AgentMessage) GetLogs() *LogsReply { if x != nil { if x, ok := x.Msg.(*AgentMessage_Logs); ok { return x.Logs } } return nil } func (x *AgentMessage) GetFabricScan() *FabricScanReply { if x != nil { if x, ok := x.Msg.(*AgentMessage_FabricScan); ok { return x.FabricScan } } return nil } func (x *AgentMessage) GetFabricTest() *FabricTestReply { if x != nil { if x, ok := x.Msg.(*AgentMessage_FabricTest); ok { return x.FabricTest } } return nil } func (x *AgentMessage) GetUpgrade() *UpgradeReply { if x != nil { if x, ok := x.Msg.(*AgentMessage_Upgrade); ok { return x.Upgrade } } return nil } func (x *AgentMessage) GetKvEvents() *KVEvents { if x != nil { if x, ok := x.Msg.(*AgentMessage_KvEvents); ok { return x.KvEvents } } return nil } func (x *AgentMessage) GetTrainStatus() *TrainStatus { if x != nil { if x, ok := x.Msg.(*AgentMessage_TrainStatus); ok { return x.TrainStatus } } return nil } func (x *AgentMessage) GetTrainArtifact() *TrainArtifact { if x != nil { if x, ok := x.Msg.(*AgentMessage_TrainArtifact); ok { return x.TrainArtifact } } return nil } type isAgentMessage_Msg interface { isAgentMessage_Msg() } type AgentMessage_Register struct { Register *Register `protobuf:"bytes,1,opt,name=register,proto3,oneof"` } type AgentMessage_Heartbeat struct { Heartbeat *Heartbeat `protobuf:"bytes,2,opt,name=heartbeat,proto3,oneof"` } type AgentMessage_Shard struct { Shard *ShardStatus `protobuf:"bytes,3,opt,name=shard,proto3,oneof"` } type AgentMessage_Logs struct { Logs *LogsReply `protobuf:"bytes,4,opt,name=logs,proto3,oneof"` } type AgentMessage_FabricScan struct { FabricScan *FabricScanReply `protobuf:"bytes,5,opt,name=fabric_scan,json=fabricScan,proto3,oneof"` } type AgentMessage_FabricTest struct { FabricTest *FabricTestReply `protobuf:"bytes,6,opt,name=fabric_test,json=fabricTest,proto3,oneof"` } type AgentMessage_Upgrade struct { Upgrade *UpgradeReply `protobuf:"bytes,7,opt,name=upgrade,proto3,oneof"` } type AgentMessage_KvEvents struct { KvEvents *KVEvents `protobuf:"bytes,8,opt,name=kv_events,json=kvEvents,proto3,oneof"` } type AgentMessage_TrainStatus struct { TrainStatus *TrainStatus `protobuf:"bytes,9,opt,name=train_status,json=trainStatus,proto3,oneof"` } type AgentMessage_TrainArtifact struct { TrainArtifact *TrainArtifact `protobuf:"bytes,10,opt,name=train_artifact,json=trainArtifact,proto3,oneof"` } func (*AgentMessage_Register) isAgentMessage_Msg() {} func (*AgentMessage_Heartbeat) isAgentMessage_Msg() {} func (*AgentMessage_Shard) isAgentMessage_Msg() {} func (*AgentMessage_Logs) isAgentMessage_Msg() {} func (*AgentMessage_FabricScan) isAgentMessage_Msg() {} func (*AgentMessage_FabricTest) isAgentMessage_Msg() {} func (*AgentMessage_Upgrade) isAgentMessage_Msg() {} func (*AgentMessage_KvEvents) isAgentMessage_Msg() {} func (*AgentMessage_TrainStatus) isAgentMessage_Msg() {} func (*AgentMessage_TrainArtifact) isAgentMessage_Msg() {} // KVEvents is a change to what one vLLM replica's prefix cache holds, read // by the agent from the engine's KV cache events and sent on for the router's // precise index (plan §7.5 tier 2). // // A net effect since the last one, not the events themselves: a block stored // and evicted between two sends is not worth the bytes. On every new stream // the agent first sends each replica in full, with clear_first set, because a // control plane that has just started or taken over knows nothing. type KVEvents struct { state protoimpl.MessageState `protogen:"open.v1"` ReplicaId string `protobuf:"bytes,1,opt,name=replica_id,json=replicaId,proto3" json:"replica_id,omitempty"` // block_size is how many tokens make one of the replica's blocks. BlockSize int32 `protobuf:"varint,2,opt,name=block_size,json=blockSize,proto3" json:"block_size,omitempty"` // clear_first says to forget what was known of the replica before applying the // rest: its cache was cleared, its engine restarted, events were lost, or // this is the first piece of a full copy. ClearFirst bool `protobuf:"varint,3,opt,name=clear_first,json=clearFirst,proto3" json:"clear_first,omitempty"` // stored and removed are block hashes as vLLM publishes them: the low 64 // bits of the sha256_cbor digest. Stored []uint64 `protobuf:"fixed64,4,rep,packed,name=stored,proto3" json:"stored,omitempty"` Removed []uint64 `protobuf:"fixed64,5,rep,packed,name=removed,proto3" json:"removed,omitempty"` // problem, when set, is why the replica's events cannot be used — most // likely an engine hashing another way than llmgw configured. Its index is // empty while it stands. Problem string `protobuf:"bytes,6,opt,name=problem,proto3" json:"problem,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *KVEvents) Reset() { *x = KVEvents{} mi := &file_llmgw_v1_agent_proto_msgTypes[1] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *KVEvents) String() string { return protoimpl.X.MessageStringOf(x) } func (*KVEvents) ProtoMessage() {} func (x *KVEvents) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[1] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use KVEvents.ProtoReflect.Descriptor instead. func (*KVEvents) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{1} } func (x *KVEvents) GetReplicaId() string { if x != nil { return x.ReplicaId } return "" } func (x *KVEvents) GetBlockSize() int32 { if x != nil { return x.BlockSize } return 0 } func (x *KVEvents) GetClearFirst() bool { if x != nil { return x.ClearFirst } return false } func (x *KVEvents) GetStored() []uint64 { if x != nil { return x.Stored } return nil } func (x *KVEvents) GetRemoved() []uint64 { if x != nil { return x.Removed } return nil } func (x *KVEvents) GetProblem() string { if x != nil { return x.Problem } return "" } // ControlMessage is anything the control plane tells a node. type ControlMessage struct { state protoimpl.MessageState `protogen:"open.v1"` // Types that are valid to be assigned to Msg: // // *ControlMessage_Start // *ControlMessage_Stop // *ControlMessage_Drain // *ControlMessage_Ack // *ControlMessage_RemoveWeights // *ControlMessage_Provision // *ControlMessage_Logs // *ControlMessage_FabricScan // *ControlMessage_FabricTest // *ControlMessage_Upgrade // *ControlMessage_ControlPlanes // *ControlMessage_Firewall // *ControlMessage_RestartAgent // *ControlMessage_AgentSettings // *ControlMessage_TrainStart // *ControlMessage_TrainCancel Msg isControlMessage_Msg `protobuf_oneof:"msg"` // trace is the W3C trace context (traceparent, and tracestate when there // is one) of the control plane's span that sent this command, so what the // agent does for it — fetching weights, launching the engine, waiting for // it to answer — lands in the same trace. Empty when the control plane is // not tracing, and ignored by an agent that is not. Trace map[string]string `protobuf:"bytes,12,rep,name=trace,proto3" json:"trace,omitempty" protobuf_key:"bytes,1,opt,name=key" protobuf_val:"bytes,2,opt,name=value"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *ControlMessage) Reset() { *x = ControlMessage{} mi := &file_llmgw_v1_agent_proto_msgTypes[2] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *ControlMessage) String() string { return protoimpl.X.MessageStringOf(x) } func (*ControlMessage) ProtoMessage() {} func (x *ControlMessage) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[2] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use ControlMessage.ProtoReflect.Descriptor instead. func (*ControlMessage) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{2} } func (x *ControlMessage) GetMsg() isControlMessage_Msg { if x != nil { return x.Msg } return nil } func (x *ControlMessage) GetStart() *StartShard { if x != nil { if x, ok := x.Msg.(*ControlMessage_Start); ok { return x.Start } } return nil } func (x *ControlMessage) GetStop() *StopShard { if x != nil { if x, ok := x.Msg.(*ControlMessage_Stop); ok { return x.Stop } } return nil } func (x *ControlMessage) GetDrain() *DrainShard { if x != nil { if x, ok := x.Msg.(*ControlMessage_Drain); ok { return x.Drain } } return nil } func (x *ControlMessage) GetAck() *Ack { if x != nil { if x, ok := x.Msg.(*ControlMessage_Ack); ok { return x.Ack } } return nil } func (x *ControlMessage) GetRemoveWeights() *RemoveWeights { if x != nil { if x, ok := x.Msg.(*ControlMessage_RemoveWeights); ok { return x.RemoveWeights } } return nil } func (x *ControlMessage) GetProvision() *Provision { if x != nil { if x, ok := x.Msg.(*ControlMessage_Provision); ok { return x.Provision } } return nil } func (x *ControlMessage) GetLogs() *LogsRequest { if x != nil { if x, ok := x.Msg.(*ControlMessage_Logs); ok { return x.Logs } } return nil } func (x *ControlMessage) GetFabricScan() *FabricScanRequest { if x != nil { if x, ok := x.Msg.(*ControlMessage_FabricScan); ok { return x.FabricScan } } return nil } func (x *ControlMessage) GetFabricTest() *FabricTestRequest { if x != nil { if x, ok := x.Msg.(*ControlMessage_FabricTest); ok { return x.FabricTest } } return nil } func (x *ControlMessage) GetUpgrade() *UpgradeAgent { if x != nil { if x, ok := x.Msg.(*ControlMessage_Upgrade); ok { return x.Upgrade } } return nil } func (x *ControlMessage) GetControlPlanes() *ControlPlanes { if x != nil { if x, ok := x.Msg.(*ControlMessage_ControlPlanes); ok { return x.ControlPlanes } } return nil } func (x *ControlMessage) GetFirewall() *Firewall { if x != nil { if x, ok := x.Msg.(*ControlMessage_Firewall); ok { return x.Firewall } } return nil } func (x *ControlMessage) GetRestartAgent() *RestartAgent { if x != nil { if x, ok := x.Msg.(*ControlMessage_RestartAgent); ok { return x.RestartAgent } } return nil } func (x *ControlMessage) GetAgentSettings() *AgentSettings { if x != nil { if x, ok := x.Msg.(*ControlMessage_AgentSettings); ok { return x.AgentSettings } } return nil } func (x *ControlMessage) GetTrainStart() *TrainStart { if x != nil { if x, ok := x.Msg.(*ControlMessage_TrainStart); ok { return x.TrainStart } } return nil } func (x *ControlMessage) GetTrainCancel() *TrainCancel { if x != nil { if x, ok := x.Msg.(*ControlMessage_TrainCancel); ok { return x.TrainCancel } } return nil } func (x *ControlMessage) GetTrace() map[string]string { if x != nil { return x.Trace } return nil } type isControlMessage_Msg interface { isControlMessage_Msg() } type ControlMessage_Start struct { Start *StartShard `protobuf:"bytes,1,opt,name=start,proto3,oneof"` } type ControlMessage_Stop struct { Stop *StopShard `protobuf:"bytes,2,opt,name=stop,proto3,oneof"` } type ControlMessage_Drain struct { Drain *DrainShard `protobuf:"bytes,3,opt,name=drain,proto3,oneof"` } type ControlMessage_Ack struct { Ack *Ack `protobuf:"bytes,4,opt,name=ack,proto3,oneof"` } type ControlMessage_RemoveWeights struct { RemoveWeights *RemoveWeights `protobuf:"bytes,5,opt,name=remove_weights,json=removeWeights,proto3,oneof"` } type ControlMessage_Provision struct { Provision *Provision `protobuf:"bytes,6,opt,name=provision,proto3,oneof"` } type ControlMessage_Logs struct { Logs *LogsRequest `protobuf:"bytes,7,opt,name=logs,proto3,oneof"` } type ControlMessage_FabricScan struct { FabricScan *FabricScanRequest `protobuf:"bytes,8,opt,name=fabric_scan,json=fabricScan,proto3,oneof"` } type ControlMessage_FabricTest struct { FabricTest *FabricTestRequest `protobuf:"bytes,9,opt,name=fabric_test,json=fabricTest,proto3,oneof"` } type ControlMessage_Upgrade struct { Upgrade *UpgradeAgent `protobuf:"bytes,10,opt,name=upgrade,proto3,oneof"` } type ControlMessage_ControlPlanes struct { ControlPlanes *ControlPlanes `protobuf:"bytes,11,opt,name=control_planes,json=controlPlanes,proto3,oneof"` } type ControlMessage_Firewall struct { Firewall *Firewall `protobuf:"bytes,13,opt,name=firewall,proto3,oneof"` } type ControlMessage_RestartAgent struct { RestartAgent *RestartAgent `protobuf:"bytes,14,opt,name=restart_agent,json=restartAgent,proto3,oneof"` } type ControlMessage_AgentSettings struct { AgentSettings *AgentSettings `protobuf:"bytes,15,opt,name=agent_settings,json=agentSettings,proto3,oneof"` } type ControlMessage_TrainStart struct { TrainStart *TrainStart `protobuf:"bytes,16,opt,name=train_start,json=trainStart,proto3,oneof"` } type ControlMessage_TrainCancel struct { TrainCancel *TrainCancel `protobuf:"bytes,17,opt,name=train_cancel,json=trainCancel,proto3,oneof"` } func (*ControlMessage_Start) isControlMessage_Msg() {} func (*ControlMessage_Stop) isControlMessage_Msg() {} func (*ControlMessage_Drain) isControlMessage_Msg() {} func (*ControlMessage_Ack) isControlMessage_Msg() {} func (*ControlMessage_RemoveWeights) isControlMessage_Msg() {} func (*ControlMessage_Provision) isControlMessage_Msg() {} func (*ControlMessage_Logs) isControlMessage_Msg() {} func (*ControlMessage_FabricScan) isControlMessage_Msg() {} func (*ControlMessage_FabricTest) isControlMessage_Msg() {} func (*ControlMessage_Upgrade) isControlMessage_Msg() {} func (*ControlMessage_ControlPlanes) isControlMessage_Msg() {} func (*ControlMessage_Firewall) isControlMessage_Msg() {} func (*ControlMessage_RestartAgent) isControlMessage_Msg() {} func (*ControlMessage_AgentSettings) isControlMessage_Msg() {} func (*ControlMessage_TrainStart) isControlMessage_Msg() {} func (*ControlMessage_TrainCancel) isControlMessage_Msg() {} // RestartAgent has the agent start again, as llmgw restart asks of every // node: its engines keep serving and are taken over again. type RestartAgent struct { state protoimpl.MessageState `protogen:"open.v1"` Reason string `protobuf:"bytes,1,opt,name=reason,proto3" json:"reason,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *RestartAgent) Reset() { *x = RestartAgent{} mi := &file_llmgw_v1_agent_proto_msgTypes[3] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *RestartAgent) String() string { return protoimpl.X.MessageStringOf(x) } func (*RestartAgent) ProtoMessage() {} func (x *RestartAgent) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[3] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use RestartAgent.ProtoReflect.Descriptor instead. func (*RestartAgent) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{3} } func (x *RestartAgent) GetReason() string { if x != nil { return x.Reason } return "" } // Firewall is whether llmgw's firewall is kept on the node — its nftables // table, which lets the cluster's networks alone reach the engines' ports and // the node exporter — sent when an agent registers and whenever the // cluster's settings change it (firewall.enabled). type Firewall struct { state protoimpl.MessageState `protogen:"open.v1"` Enabled bool `protobuf:"varint,1,opt,name=enabled,proto3" json:"enabled,omitempty"` // networks are let in besides those the node reaches the control planes // on: firewall_networks, networks or single addresses. Networks []string `protobuf:"bytes,2,rep,name=networks,proto3" json:"networks,omitempty"` // engine_ports_from and engine_ports_to are the engines' ports // (control.port_range). EnginePortsFrom int32 `protobuf:"varint,3,opt,name=engine_ports_from,json=enginePortsFrom,proto3" json:"engine_ports_from,omitempty"` EnginePortsTo int32 `protobuf:"varint,4,opt,name=engine_ports_to,json=enginePortsTo,proto3" json:"engine_ports_to,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *Firewall) Reset() { *x = Firewall{} mi := &file_llmgw_v1_agent_proto_msgTypes[4] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *Firewall) String() string { return protoimpl.X.MessageStringOf(x) } func (*Firewall) ProtoMessage() {} func (x *Firewall) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[4] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use Firewall.ProtoReflect.Descriptor instead. func (*Firewall) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{4} } func (x *Firewall) GetEnabled() bool { if x != nil { return x.Enabled } return false } func (x *Firewall) GetNetworks() []string { if x != nil { return x.Networks } return nil } func (x *Firewall) GetEnginePortsFrom() int32 { if x != nil { return x.EnginePortsFrom } return 0 } func (x *Firewall) GetEnginePortsTo() int32 { if x != nil { return x.EnginePortsTo } return 0 } // AgentSettings is what the cluster's settings say of this node's agent // (agents.default, with agents.nodes. over it), sent when it registers // and whenever they change. The agent keeps it beside agent.yaml, which // holds only what it needs to start; over that file, it is what the agent // runs on. A change from what it runs on has the agent start again on it, // its engines serving throughout. type AgentSettings struct { state protoimpl.MessageState `protogen:"open.v1"` // yaml is agent.yaml's keys, as YAML; empty for none. Yaml string `protobuf:"bytes,1,opt,name=yaml,proto3" json:"yaml,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *AgentSettings) Reset() { *x = AgentSettings{} mi := &file_llmgw_v1_agent_proto_msgTypes[5] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *AgentSettings) String() string { return protoimpl.X.MessageStringOf(x) } func (*AgentSettings) ProtoMessage() {} func (x *AgentSettings) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[5] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use AgentSettings.ProtoReflect.Descriptor instead. func (*AgentSettings) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{5} } func (x *AgentSettings) GetYaml() string { if x != nil { return x.Yaml } return "" } // ControlPlanes is every control plane an agent may dial, sent by the leader // when an agent registers and whenever the list changes. The agent keeps it // beside its configuration, so a second control plane added after the node // was onboarded is one it fails over to without anybody editing the node. type ControlPlanes struct { state protoimpl.MessageState `protogen:"open.v1"` Servers []string `protobuf:"bytes,1,rep,name=servers,proto3" json:"servers,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *ControlPlanes) Reset() { *x = ControlPlanes{} mi := &file_llmgw_v1_agent_proto_msgTypes[6] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *ControlPlanes) String() string { return protoimpl.X.MessageStringOf(x) } func (*ControlPlanes) ProtoMessage() {} func (x *ControlPlanes) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[6] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use ControlPlanes.ProtoReflect.Descriptor instead. func (*ControlPlanes) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{6} } func (x *ControlPlanes) GetServers() []string { if x != nil { return x.Servers } return nil } // Register is the first message on every stream. It carries the node's full // hardware inventory, so the control plane needs no separate discovery call and // a reconnecting agent automatically refreshes what the scheduler knows. type Register struct { state protoimpl.MessageState `protogen:"open.v1"` NodeId string `protobuf:"bytes,1,opt,name=node_id,json=nodeId,proto3" json:"node_id,omitempty"` Hostname string `protobuf:"bytes,2,opt,name=hostname,proto3" json:"hostname,omitempty"` // mgmt_addr is the address engines are reachable on, and the address NCCL // bootstraps over. Never a fabric address: NCCL rendezvouses over TCP before // it moves to RDMA. MgmtAddr string `protobuf:"bytes,3,opt,name=mgmt_addr,json=mgmtAddr,proto3" json:"mgmt_addr,omitempty"` Labels map[string]string `protobuf:"bytes,4,rep,name=labels,proto3" json:"labels,omitempty" protobuf_key:"bytes,1,opt,name=key" protobuf_val:"bytes,2,opt,name=value"` Gpus []*GPU `protobuf:"bytes,5,rep,name=gpus,proto3" json:"gpus,omitempty"` // fabrics is empty on a node without InfiniBand. Such a node is its own // single-node island and can never host a multi-node replica. Fabrics []*FabricAttachment `protobuf:"bytes,6,rep,name=fabrics,proto3" json:"fabrics,omitempty"` AgentVersion string `protobuf:"bytes,7,opt,name=agent_version,json=agentVersion,proto3" json:"agent_version,omitempty"` // engine_versions decides which CLI flags and metric names apply; both drift // between engine releases (plan §14). EngineVersions map[string]string `protobuf:"bytes,8,rep,name=engine_versions,json=engineVersions,proto3" json:"engine_versions,omitempty" protobuf_key:"bytes,1,opt,name=key" protobuf_val:"bytes,2,opt,name=value"` // running_shards are the engines this agent is already supervising, including // ones adopted from a previous agent. // // This is what lets a control plane restart recover instead of orphaning live // engines. Without it the control plane forgets every replica, places new // ones, and the old engines keep holding GPUs the inventory believes are free // — so the new engines fail to allocate, with an error naming the wrong // process (plan §6.5). RunningShards []*RunningShard `protobuf:"bytes,9,rep,name=running_shards,json=runningShards,proto3" json:"running_shards,omitempty"` // memory_total_mb is the machine's RAM, and 0 means the agent did not say — // which is what every agent older than this field reports, and is treated as // no constraint rather than as a node with no memory. // // Placement needs it because a CPU profile has no card to be free, so every // ready node looks like a candidate: a 14B model whose working set is 24 GB // would be placed just as happily on a node with 8 GB. MemoryTotalMb int64 `protobuf:"varint,10,opt,name=memory_total_mb,json=memoryTotalMb,proto3" json:"memory_total_mb,omitempty"` // weights are the model files this node can see, from the directories its // configuration names. // // How they got there is the operator's business and deliberately outside // llmgw's: a shared filesystem, an image exported over the fabric, rsync, or // a copy made by hand all look identical here. What matters to placement is // only whether they are present, because a node without them either cannot // start the replica at all or must first move tens of gigabytes. // // Empty means the node was not asked to look, which is not the same as having // nothing — an agent with no model directories configured behaves exactly as // every agent did before this existed, and constrains nothing. Weights []*LocalWeights `protobuf:"bytes,11,rep,name=weights,proto3" json:"weights,omitempty"` // storage is the disk behind those weights, one entry per filesystem. // // Reported so an operator can see a node filling up before a fetch dies at // ninety percent, and so placement can refuse a node that cannot hold a // model rather than discovering it during the download. // // Empty means the node was not asked to look, exactly as for weights, and // constrains nothing. Storage []*Storage `protobuf:"bytes,12,rep,name=storage,proto3" json:"storage,omitempty"` // host is what the machine is, so the control plane can have the mirror // fetch software built for it rather than assuming every node matches the // first. Host *Host `protobuf:"bytes,13,opt,name=host,proto3" json:"host,omitempty"` // engines_probed says engine_versions is an answer. Without it an empty map // meant "an agent too old to say" and constrained nothing — so a freshly // built node with no engine installed yet was a candidate for every model. EnginesProbed bool `protobuf:"varint,14,opt,name=engines_probed,json=enginesProbed,proto3" json:"engines_probed,omitempty"` // cpu_count is the CPUs the agent may run on. Placement reserves cores // for CPU replicas out of it, the way it reserves memory out of // memory_total_mb; zero, from an older agent, reserves nothing. CpuCount int32 `protobuf:"varint,18,opt,name=cpu_count,json=cpuCount,proto3" json:"cpu_count,omitempty"` // imex is the service that lets a multi-node NVLink domain share memory. Imex *IMEX `protobuf:"bytes,15,opt,name=imex,proto3" json:"imex,omitempty"` // installable_engines are the engines this agent knows how to install // itself. The control plane asks it for no others. Empty is an agent older // than the field, which could install llama.cpp and nothing else: asked for // vLLM, it answered "unsupported", and every node of a cluster half way // through an upgrade showed as failing. InstallableEngines []string `protobuf:"bytes,16,rep,name=installable_engines,json=installableEngines,proto3" json:"installable_engines,omitempty"` // features are the requests this agent answers beyond the ones every agent // does: "fabric-test". An agent that does not know a request ignores it, so // without this the control plane would wait minutes for an answer that is // never coming. Features []string `protobuf:"bytes,17,rep,name=features,proto3" json:"features,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *Register) Reset() { *x = Register{} mi := &file_llmgw_v1_agent_proto_msgTypes[7] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *Register) String() string { return protoimpl.X.MessageStringOf(x) } func (*Register) ProtoMessage() {} func (x *Register) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[7] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use Register.ProtoReflect.Descriptor instead. func (*Register) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{7} } func (x *Register) GetNodeId() string { if x != nil { return x.NodeId } return "" } func (x *Register) GetHostname() string { if x != nil { return x.Hostname } return "" } func (x *Register) GetMgmtAddr() string { if x != nil { return x.MgmtAddr } return "" } func (x *Register) GetLabels() map[string]string { if x != nil { return x.Labels } return nil } func (x *Register) GetGpus() []*GPU { if x != nil { return x.Gpus } return nil } func (x *Register) GetFabrics() []*FabricAttachment { if x != nil { return x.Fabrics } return nil } func (x *Register) GetAgentVersion() string { if x != nil { return x.AgentVersion } return "" } func (x *Register) GetEngineVersions() map[string]string { if x != nil { return x.EngineVersions } return nil } func (x *Register) GetRunningShards() []*RunningShard { if x != nil { return x.RunningShards } return nil } func (x *Register) GetMemoryTotalMb() int64 { if x != nil { return x.MemoryTotalMb } return 0 } func (x *Register) GetWeights() []*LocalWeights { if x != nil { return x.Weights } return nil } func (x *Register) GetStorage() []*Storage { if x != nil { return x.Storage } return nil } func (x *Register) GetHost() *Host { if x != nil { return x.Host } return nil } func (x *Register) GetEnginesProbed() bool { if x != nil { return x.EnginesProbed } return false } func (x *Register) GetCpuCount() int32 { if x != nil { return x.CpuCount } return 0 } func (x *Register) GetImex() *IMEX { if x != nil { return x.Imex } return nil } func (x *Register) GetInstallableEngines() []string { if x != nil { return x.InstallableEngines } return nil } func (x *Register) GetFeatures() []string { if x != nil { return x.Features } return nil } // Host describes the operating system and processor of a node. type Host struct { state protoimpl.MessageState `protogen:"open.v1"` // From /etc/os-release: "ubuntu", "24.04", "noble". OsId string `protobuf:"bytes,1,opt,name=os_id,json=osId,proto3" json:"os_id,omitempty"` OsVersion string `protobuf:"bytes,2,opt,name=os_version,json=osVersion,proto3" json:"os_version,omitempty"` OsCodename string `protobuf:"bytes,3,opt,name=os_codename,json=osCodename,proto3" json:"os_codename,omitempty"` // Debian's spelling, which is what package indexes are keyed by: "amd64", // "arm64". Arch string `protobuf:"bytes,4,opt,name=arch,proto3" json:"arch,omitempty"` // The vector extensions inference cares about, of those the CPU has. A VM // given QEMU's default CPU model has none of them, and runs, and is five // times slower, and nothing says why. CpuFlags []string `protobuf:"bytes,5,rep,name=cpu_flags,json=cpuFlags,proto3" json:"cpu_flags,omitempty"` // nvidia_driver is the driver version, empty when there is none. A card // without a driver is not a GPU anything can use. NvidiaDriver string `protobuf:"bytes,6,opt,name=nvidia_driver,json=nvidiaDriver,proto3" json:"nvidia_driver,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *Host) Reset() { *x = Host{} mi := &file_llmgw_v1_agent_proto_msgTypes[8] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *Host) String() string { return protoimpl.X.MessageStringOf(x) } func (*Host) ProtoMessage() {} func (x *Host) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[8] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use Host.ProtoReflect.Descriptor instead. func (*Host) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{8} } func (x *Host) GetOsId() string { if x != nil { return x.OsId } return "" } func (x *Host) GetOsVersion() string { if x != nil { return x.OsVersion } return "" } func (x *Host) GetOsCodename() string { if x != nil { return x.OsCodename } return "" } func (x *Host) GetArch() string { if x != nil { return x.Arch } return "" } func (x *Host) GetCpuFlags() []string { if x != nil { return x.CpuFlags } return nil } func (x *Host) GetNvidiaDriver() string { if x != nil { return x.NvidiaDriver } return "" } // Storage is one filesystem a node keeps weights on. // // Per filesystem rather than per configured directory: free space belongs to // the disk, and two model directories on one disk are one pool of bytes. type Storage struct { state protoimpl.MessageState `protogen:"open.v1"` // path is one directory on this filesystem, so the operator recognises it. Path string `protobuf:"bytes,1,opt,name=path,proto3" json:"path,omitempty"` // free_bytes is what a non-root process can write — not the root reserve, // which is space the agent cannot have. FreeBytes int64 `protobuf:"varint,2,opt,name=free_bytes,json=freeBytes,proto3" json:"free_bytes,omitempty"` TotalBytes int64 `protobuf:"varint,3,opt,name=total_bytes,json=totalBytes,proto3" json:"total_bytes,omitempty"` // shared marks a filesystem the node did not fill. Free space on one is // shared with every other node using it, so two nodes each reporting 500 GB // free on the same filer have 500 GB between them and not a terabyte. Shared bool `protobuf:"varint,4,opt,name=shared,proto3" json:"shared,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *Storage) Reset() { *x = Storage{} mi := &file_llmgw_v1_agent_proto_msgTypes[9] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *Storage) String() string { return protoimpl.X.MessageStringOf(x) } func (*Storage) ProtoMessage() {} func (x *Storage) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[9] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use Storage.ProtoReflect.Descriptor instead. func (*Storage) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{9} } func (x *Storage) GetPath() string { if x != nil { return x.Path } return "" } func (x *Storage) GetFreeBytes() int64 { if x != nil { return x.FreeBytes } return 0 } func (x *Storage) GetTotalBytes() int64 { if x != nil { return x.TotalBytes } return 0 } func (x *Storage) GetShared() bool { if x != nil { return x.Shared } return false } // RunningShard describes an engine already running on a node. // // It carries enough for the control plane to rebuild the replica it belongs to, // which is more than a shard id: the replica's identity, its pool and fabric, // and the leader endpoint clients reach it through. type RunningShard struct { state protoimpl.MessageState `protogen:"open.v1"` ReplicaId string `protobuf:"bytes,1,opt,name=replica_id,json=replicaId,proto3" json:"replica_id,omitempty"` ShardId string `protobuf:"bytes,2,opt,name=shard_id,json=shardId,proto3" json:"shard_id,omitempty"` Rank int32 `protobuf:"varint,3,opt,name=rank,proto3" json:"rank,omitempty"` Nnodes int32 `protobuf:"varint,4,opt,name=nnodes,proto3" json:"nnodes,omitempty"` GpuIndices []int32 `protobuf:"varint,5,rep,packed,name=gpu_indices,json=gpuIndices,proto3" json:"gpu_indices,omitempty"` Hcas []string `protobuf:"bytes,6,rep,name=hcas,proto3" json:"hcas,omitempty"` HttpPort int32 `protobuf:"varint,7,opt,name=http_port,json=httpPort,proto3" json:"http_port,omitempty"` State ShardStatus_State `protobuf:"varint,8,opt,name=state,proto3,enum=llmgw.v1.ShardStatus_State" json:"state,omitempty"` // model and profile name the catalog entry this shard was started from. Model string `protobuf:"bytes,9,opt,name=model,proto3" json:"model,omitempty"` Profile string `protobuf:"bytes,10,opt,name=profile,proto3" json:"profile,omitempty"` // pool and fabric are echoed back from the StartShard that created it; the // agent has no way to derive them. Pool string `protobuf:"bytes,11,opt,name=pool,proto3" json:"pool,omitempty"` Fabric string `protobuf:"bytes,12,opt,name=fabric,proto3" json:"fabric,omitempty"` LeaderAddr string `protobuf:"bytes,13,opt,name=leader_addr,json=leaderAddr,proto3" json:"leader_addr,omitempty"` LeaderPort int32 `protobuf:"varint,14,opt,name=leader_port,json=leaderPort,proto3" json:"leader_port,omitempty"` // profile_fp is the fingerprint of the profile this shard was started from, // echoed back from StartShard. The control plane keeps no record of a running // replica across its own restart, so without this a shard adopted after a // restart could not be compared with the catalog — which is exactly when the // catalog is most likely to have been edited. ProfileFp string `protobuf:"bytes,15,opt,name=profile_fp,json=profileFp,proto3" json:"profile_fp,omitempty"` // compute is where its engine said it put the model when it became ready, // kept by the agent across its own restarts: a control plane that takes // over, or restarts, learns it here. Compute *Compute `protobuf:"bytes,16,opt,name=compute,proto3" json:"compute,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *RunningShard) Reset() { *x = RunningShard{} mi := &file_llmgw_v1_agent_proto_msgTypes[10] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *RunningShard) String() string { return protoimpl.X.MessageStringOf(x) } func (*RunningShard) ProtoMessage() {} func (x *RunningShard) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[10] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use RunningShard.ProtoReflect.Descriptor instead. func (*RunningShard) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{10} } func (x *RunningShard) GetReplicaId() string { if x != nil { return x.ReplicaId } return "" } func (x *RunningShard) GetShardId() string { if x != nil { return x.ShardId } return "" } func (x *RunningShard) GetRank() int32 { if x != nil { return x.Rank } return 0 } func (x *RunningShard) GetNnodes() int32 { if x != nil { return x.Nnodes } return 0 } func (x *RunningShard) GetGpuIndices() []int32 { if x != nil { return x.GpuIndices } return nil } func (x *RunningShard) GetHcas() []string { if x != nil { return x.Hcas } return nil } func (x *RunningShard) GetHttpPort() int32 { if x != nil { return x.HttpPort } return 0 } func (x *RunningShard) GetState() ShardStatus_State { if x != nil { return x.State } return ShardStatus_STATE_UNSPECIFIED } func (x *RunningShard) GetModel() string { if x != nil { return x.Model } return "" } func (x *RunningShard) GetProfile() string { if x != nil { return x.Profile } return "" } func (x *RunningShard) GetPool() string { if x != nil { return x.Pool } return "" } func (x *RunningShard) GetFabric() string { if x != nil { return x.Fabric } return "" } func (x *RunningShard) GetLeaderAddr() string { if x != nil { return x.LeaderAddr } return "" } func (x *RunningShard) GetLeaderPort() int32 { if x != nil { return x.LeaderPort } return 0 } func (x *RunningShard) GetProfileFp() string { if x != nil { return x.ProfileFp } return "" } func (x *RunningShard) GetCompute() *Compute { if x != nil { return x.Compute } return nil } type GPU struct { state protoimpl.MessageState `protogen:"open.v1"` Index int32 `protobuf:"varint,1,opt,name=index,proto3" json:"index,omitempty"` Uuid string `protobuf:"bytes,2,opt,name=uuid,proto3" json:"uuid,omitempty"` Name string `protobuf:"bytes,3,opt,name=name,proto3" json:"name,omitempty"` MemoryTotalMb int32 `protobuf:"varint,4,opt,name=memory_total_mb,json=memoryTotalMb,proto3" json:"memory_total_mb,omitempty"` MemoryUsedMb int32 `protobuf:"varint,5,opt,name=memory_used_mb,json=memoryUsedMb,proto3" json:"memory_used_mb,omitempty"` ComputeCapability string `protobuf:"bytes,6,opt,name=compute_capability,json=computeCapability,proto3" json:"compute_capability,omitempty"` NumaNode int32 `protobuf:"varint,7,opt,name=numa_node,json=numaNode,proto3" json:"numa_node,omitempty"` PciBusId string `protobuf:"bytes,8,opt,name=pci_bus_id,json=pciBusId,proto3" json:"pci_bus_id,omitempty"` DriverVersion string `protobuf:"bytes,9,opt,name=driver_version,json=driverVersion,proto3" json:"driver_version,omitempty"` CudaVersion string `protobuf:"bytes,10,opt,name=cuda_version,json=cudaVersion,proto3" json:"cuda_version,omitempty"` // nvlink_peers is only populated by agents built with the nvml tag. NvlinkPeers []int32 `protobuf:"varint,11,rep,packed,name=nvlink_peers,json=nvlinkPeers,proto3" json:"nvlink_peers,omitempty"` // problems keeps a GPU out of placement: ECC faults, or memory already held // by another process. Problems []string `protobuf:"bytes,12,rep,name=problems,proto3" json:"problems,omitempty"` TemperatureC int32 `protobuf:"varint,13,opt,name=temperature_c,json=temperatureC,proto3" json:"temperature_c,omitempty"` EccErrors int32 `protobuf:"varint,14,opt,name=ecc_errors,json=eccErrors,proto3" json:"ecc_errors,omitempty"` // utilization_percent is how busy the card was over the last sample period. // Zero when the driver is too old to report it, which is not the same as // idle: the metric is simply absent rather than wrong. UtilizationPercent int32 `protobuf:"varint,15,opt,name=utilization_percent,json=utilizationPercent,proto3" json:"utilization_percent,omitempty"` // nic_paths is how far each network adapter is from this card, by HCA // name: PIX, PXB, PHB, NODE or SYS, from nvidia-smi topo. NicPaths map[string]string `protobuf:"bytes,16,rep,name=nic_paths,json=nicPaths,proto3" json:"nic_paths,omitempty" protobuf_key:"bytes,1,opt,name=key" protobuf_val:"bytes,2,opt,name=value"` // The card's multi-node NVLink membership, from nvidia-smi -q: state is // Completed once the fabric manager has joined it; cluster_uuid and // clique_id name the domain it joined. FabricState string `protobuf:"bytes,17,opt,name=fabric_state,json=fabricState,proto3" json:"fabric_state,omitempty"` FabricStatus string `protobuf:"bytes,18,opt,name=fabric_status,json=fabricStatus,proto3" json:"fabric_status,omitempty"` ClusterUuid string `protobuf:"bytes,19,opt,name=cluster_uuid,json=clusterUuid,proto3" json:"cluster_uuid,omitempty"` CliqueId string `protobuf:"bytes,20,opt,name=clique_id,json=cliqueId,proto3" json:"clique_id,omitempty"` FabricBandwidth string `protobuf:"bytes,21,opt,name=fabric_bandwidth,json=fabricBandwidth,proto3" json:"fabric_bandwidth,omitempty"` // mig is set when this is a MIG instance rather than a whole card. index // is then 1000 + 100 × card + device, uuid the instance's (MIG-…), and // memory the instance's share; utilization_percent stays zero, because no // driver measures it per instance. Mig *MIGDevice `protobuf:"bytes,22,opt,name=mig,proto3" json:"mig,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *GPU) Reset() { *x = GPU{} mi := &file_llmgw_v1_agent_proto_msgTypes[11] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *GPU) String() string { return protoimpl.X.MessageStringOf(x) } func (*GPU) ProtoMessage() {} func (x *GPU) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[11] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use GPU.ProtoReflect.Descriptor instead. func (*GPU) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{11} } func (x *GPU) GetIndex() int32 { if x != nil { return x.Index } return 0 } func (x *GPU) GetUuid() string { if x != nil { return x.Uuid } return "" } func (x *GPU) GetName() string { if x != nil { return x.Name } return "" } func (x *GPU) GetMemoryTotalMb() int32 { if x != nil { return x.MemoryTotalMb } return 0 } func (x *GPU) GetMemoryUsedMb() int32 { if x != nil { return x.MemoryUsedMb } return 0 } func (x *GPU) GetComputeCapability() string { if x != nil { return x.ComputeCapability } return "" } func (x *GPU) GetNumaNode() int32 { if x != nil { return x.NumaNode } return 0 } func (x *GPU) GetPciBusId() string { if x != nil { return x.PciBusId } return "" } func (x *GPU) GetDriverVersion() string { if x != nil { return x.DriverVersion } return "" } func (x *GPU) GetCudaVersion() string { if x != nil { return x.CudaVersion } return "" } func (x *GPU) GetNvlinkPeers() []int32 { if x != nil { return x.NvlinkPeers } return nil } func (x *GPU) GetProblems() []string { if x != nil { return x.Problems } return nil } func (x *GPU) GetTemperatureC() int32 { if x != nil { return x.TemperatureC } return 0 } func (x *GPU) GetEccErrors() int32 { if x != nil { return x.EccErrors } return 0 } func (x *GPU) GetUtilizationPercent() int32 { if x != nil { return x.UtilizationPercent } return 0 } func (x *GPU) GetNicPaths() map[string]string { if x != nil { return x.NicPaths } return nil } func (x *GPU) GetFabricState() string { if x != nil { return x.FabricState } return "" } func (x *GPU) GetFabricStatus() string { if x != nil { return x.FabricStatus } return "" } func (x *GPU) GetClusterUuid() string { if x != nil { return x.ClusterUuid } return "" } func (x *GPU) GetCliqueId() string { if x != nil { return x.CliqueId } return "" } func (x *GPU) GetFabricBandwidth() string { if x != nil { return x.FabricBandwidth } return "" } func (x *GPU) GetMig() *MIGDevice { if x != nil { return x.Mig } return nil } // MIGDevice is which card a MIG instance is carved from, and how. type MIGDevice struct { state protoimpl.MessageState `protogen:"open.v1"` ParentIndex int32 `protobuf:"varint,1,opt,name=parent_index,json=parentIndex,proto3" json:"parent_index,omitempty"` ParentUuid string `protobuf:"bytes,2,opt,name=parent_uuid,json=parentUuid,proto3" json:"parent_uuid,omitempty"` // parent_memory_total_mb is the whole card's: the class is named from it. ParentMemoryTotalMb int32 `protobuf:"varint,3,opt,name=parent_memory_total_mb,json=parentMemoryTotalMb,proto3" json:"parent_memory_total_mb,omitempty"` // device is the instance's number on its card, as nvidia-smi -L lists it. Device int32 `protobuf:"varint,4,opt,name=device,proto3" json:"device,omitempty"` // profile is NVIDIA's name for the slice, like 3g.40gb. Profile string `protobuf:"bytes,5,opt,name=profile,proto3" json:"profile,omitempty"` GpuInstance int32 `protobuf:"varint,6,opt,name=gpu_instance,json=gpuInstance,proto3" json:"gpu_instance,omitempty"` ComputeInstance int32 `protobuf:"varint,7,opt,name=compute_instance,json=computeInstance,proto3" json:"compute_instance,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *MIGDevice) Reset() { *x = MIGDevice{} mi := &file_llmgw_v1_agent_proto_msgTypes[12] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *MIGDevice) String() string { return protoimpl.X.MessageStringOf(x) } func (*MIGDevice) ProtoMessage() {} func (x *MIGDevice) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[12] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use MIGDevice.ProtoReflect.Descriptor instead. func (*MIGDevice) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{12} } func (x *MIGDevice) GetParentIndex() int32 { if x != nil { return x.ParentIndex } return 0 } func (x *MIGDevice) GetParentUuid() string { if x != nil { return x.ParentUuid } return "" } func (x *MIGDevice) GetParentMemoryTotalMb() int32 { if x != nil { return x.ParentMemoryTotalMb } return 0 } func (x *MIGDevice) GetDevice() int32 { if x != nil { return x.Device } return 0 } func (x *MIGDevice) GetProfile() string { if x != nil { return x.Profile } return "" } func (x *MIGDevice) GetGpuInstance() int32 { if x != nil { return x.GpuInstance } return 0 } func (x *MIGDevice) GetComputeInstance() int32 { if x != nil { return x.ComputeInstance } return 0 } type FabricAttachment struct { state protoimpl.MessageState `protogen:"open.v1"` // fabric is the operator-assigned name. It is not derived from the GID // prefix: OpenSM defaults to fe80:: on every fabric, so a matching prefix // proves nothing (plan §5.2). Fabric string `protobuf:"bytes,1,opt,name=fabric,proto3" json:"fabric,omitempty"` Hcas []string `protobuf:"bytes,2,rep,name=hcas,proto3" json:"hcas,omitempty"` RateGbps int32 `protobuf:"varint,3,opt,name=rate_gbps,json=rateGbps,proto3" json:"rate_gbps,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *FabricAttachment) Reset() { *x = FabricAttachment{} mi := &file_llmgw_v1_agent_proto_msgTypes[13] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *FabricAttachment) String() string { return protoimpl.X.MessageStringOf(x) } func (*FabricAttachment) ProtoMessage() {} func (x *FabricAttachment) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[13] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use FabricAttachment.ProtoReflect.Descriptor instead. func (*FabricAttachment) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{13} } func (x *FabricAttachment) GetFabric() string { if x != nil { return x.Fabric } return "" } func (x *FabricAttachment) GetHcas() []string { if x != nil { return x.Hcas } return nil } func (x *FabricAttachment) GetRateGbps() int32 { if x != nil { return x.RateGbps } return 0 } // Heartbeat says the node is alive and reports anything that changed cheaply. // LocalWeights is one model file or directory a node can see. type LocalWeights struct { state protoimpl.MessageState `protogen:"open.v1"` // path is what an engine would be given, so it can be compared against a // model's source without interpretation. Path string `protobuf:"bytes,1,opt,name=path,proto3" json:"path,omitempty"` SizeBytes int64 `protobuf:"varint,2,opt,name=size_bytes,json=sizeBytes,proto3" json:"size_bytes,omitempty"` // shared marks weights on a filesystem the node did not have to fill — // reported so a fabric with a shared filer is not mistaken for one where // every node happens to have fetched its own copy. Shared bool `protobuf:"varint,3,opt,name=shared,proto3" json:"shared,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *LocalWeights) Reset() { *x = LocalWeights{} mi := &file_llmgw_v1_agent_proto_msgTypes[14] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *LocalWeights) String() string { return protoimpl.X.MessageStringOf(x) } func (*LocalWeights) ProtoMessage() {} func (x *LocalWeights) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[14] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use LocalWeights.ProtoReflect.Descriptor instead. func (*LocalWeights) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{14} } func (x *LocalWeights) GetPath() string { if x != nil { return x.Path } return "" } func (x *LocalWeights) GetSizeBytes() int64 { if x != nil { return x.SizeBytes } return 0 } func (x *LocalWeights) GetShared() bool { if x != nil { return x.Shared } return false } // NodeLoad is a node's use of its own CPU and memory, and each engine's share. // // Carried on the heartbeat rather than scraped from the node: the control // plane already exports the fleet from one place, and a node exporter per // machine would be a scrape target, a port and a firewall rule each. type NodeLoad struct { state protoimpl.MessageState `protogen:"open.v1"` // cpu_percent is the whole machine, 0 to 100 however many cores it has. CpuPercent float64 `protobuf:"fixed64,1,opt,name=cpu_percent,json=cpuPercent,proto3" json:"cpu_percent,omitempty"` CpuCount int32 `protobuf:"varint,2,opt,name=cpu_count,json=cpuCount,proto3" json:"cpu_count,omitempty"` // The kernel's run-queue averages over 1, 5 and 15 minutes. Load1 float64 `protobuf:"fixed64,3,opt,name=load1,proto3" json:"load1,omitempty"` Load5 float64 `protobuf:"fixed64,4,opt,name=load5,proto3" json:"load5,omitempty"` Load15 float64 `protobuf:"fixed64,5,opt,name=load15,proto3" json:"load15,omitempty"` // memory_used_bytes is total less available: what the machine could not // hand to a new process without evicting something. MemoryUsedBytes int64 `protobuf:"varint,6,opt,name=memory_used_bytes,json=memoryUsedBytes,proto3" json:"memory_used_bytes,omitempty"` MemoryAvailableBytes int64 `protobuf:"varint,7,opt,name=memory_available_bytes,json=memoryAvailableBytes,proto3" json:"memory_available_bytes,omitempty"` Shards []*ShardLoad `protobuf:"bytes,8,rep,name=shards,proto3" json:"shards,omitempty"` // What the machine's own sensors say; zero is not measured. cpu_temp_c is // the hottest CPU sensor, cpu_power_w the CPU packages' draw from RAPL, // inlet_temp_c the board's inlet or ambient sensor where it has one. CpuTempC float64 `protobuf:"fixed64,9,opt,name=cpu_temp_c,json=cpuTempC,proto3" json:"cpu_temp_c,omitempty"` CpuPowerW float64 `protobuf:"fixed64,10,opt,name=cpu_power_w,json=cpuPowerW,proto3" json:"cpu_power_w,omitempty"` InletTempC float64 `protobuf:"fixed64,11,opt,name=inlet_temp_c,json=inletTempC,proto3" json:"inlet_temp_c,omitempty"` Gpus []*GPUSensor `protobuf:"bytes,12,rep,name=gpus,proto3" json:"gpus,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *NodeLoad) Reset() { *x = NodeLoad{} mi := &file_llmgw_v1_agent_proto_msgTypes[15] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *NodeLoad) String() string { return protoimpl.X.MessageStringOf(x) } func (*NodeLoad) ProtoMessage() {} func (x *NodeLoad) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[15] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use NodeLoad.ProtoReflect.Descriptor instead. func (*NodeLoad) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{15} } func (x *NodeLoad) GetCpuPercent() float64 { if x != nil { return x.CpuPercent } return 0 } func (x *NodeLoad) GetCpuCount() int32 { if x != nil { return x.CpuCount } return 0 } func (x *NodeLoad) GetLoad1() float64 { if x != nil { return x.Load1 } return 0 } func (x *NodeLoad) GetLoad5() float64 { if x != nil { return x.Load5 } return 0 } func (x *NodeLoad) GetLoad15() float64 { if x != nil { return x.Load15 } return 0 } func (x *NodeLoad) GetMemoryUsedBytes() int64 { if x != nil { return x.MemoryUsedBytes } return 0 } func (x *NodeLoad) GetMemoryAvailableBytes() int64 { if x != nil { return x.MemoryAvailableBytes } return 0 } func (x *NodeLoad) GetShards() []*ShardLoad { if x != nil { return x.Shards } return nil } func (x *NodeLoad) GetCpuTempC() float64 { if x != nil { return x.CpuTempC } return 0 } func (x *NodeLoad) GetCpuPowerW() float64 { if x != nil { return x.CpuPowerW } return 0 } func (x *NodeLoad) GetInletTempC() float64 { if x != nil { return x.InletTempC } return 0 } func (x *NodeLoad) GetGpus() []*GPUSensor { if x != nil { return x.Gpus } return nil } // GPUSensor is one card's heat, power, and what is holding it back. type GPUSensor struct { state protoimpl.MessageState `protogen:"open.v1"` Index int32 `protobuf:"varint,1,opt,name=index,proto3" json:"index,omitempty"` TempC float64 `protobuf:"fixed64,2,opt,name=temp_c,json=tempC,proto3" json:"temp_c,omitempty"` MemoryTempC float64 `protobuf:"fixed64,3,opt,name=memory_temp_c,json=memoryTempC,proto3" json:"memory_temp_c,omitempty"` PowerW float64 `protobuf:"fixed64,4,opt,name=power_w,json=powerW,proto3" json:"power_w,omitempty"` PowerLimitW float64 `protobuf:"fixed64,5,opt,name=power_limit_w,json=powerLimitW,proto3" json:"power_limit_w,omitempty"` // throttle names what is holding the clocks down now: sw_power_cap, // hw_slowdown, sw_thermal, hw_thermal, power_brake. Throttle []string `protobuf:"bytes,6,rep,name=throttle,proto3" json:"throttle,omitempty"` // How busy the card is, sampled with the rest, rather than read once at // registration. -1 is not reported by the driver. memory_total_mb is set // by every agent that sends these, so zero there is an older agent. UtilizationPercent float64 `protobuf:"fixed64,7,opt,name=utilization_percent,json=utilizationPercent,proto3" json:"utilization_percent,omitempty"` MemoryUtilizationPercent float64 `protobuf:"fixed64,8,opt,name=memory_utilization_percent,json=memoryUtilizationPercent,proto3" json:"memory_utilization_percent,omitempty"` MemoryUsedMb float64 `protobuf:"fixed64,9,opt,name=memory_used_mb,json=memoryUsedMb,proto3" json:"memory_used_mb,omitempty"` MemoryTotalMb float64 `protobuf:"fixed64,10,opt,name=memory_total_mb,json=memoryTotalMb,proto3" json:"memory_total_mb,omitempty"` SmClockMhz float64 `protobuf:"fixed64,11,opt,name=sm_clock_mhz,json=smClockMhz,proto3" json:"sm_clock_mhz,omitempty"` // ecc_uncorrected counts uncorrected ECC errors since the driver loaded. EccUncorrected float64 `protobuf:"fixed64,12,opt,name=ecc_uncorrected,json=eccUncorrected,proto3" json:"ecc_uncorrected,omitempty"` // What wears the card out, read every minute; has_wear says these are // present. Counts of -1 are not reported for this card. HasWear bool `protobuf:"varint,13,opt,name=has_wear,json=hasWear,proto3" json:"has_wear,omitempty"` PcieReplays int64 `protobuf:"varint,14,opt,name=pcie_replays,json=pcieReplays,proto3" json:"pcie_replays,omitempty"` RetiredPagesSbe int64 `protobuf:"varint,15,opt,name=retired_pages_sbe,json=retiredPagesSbe,proto3" json:"retired_pages_sbe,omitempty"` RetiredPagesDbe int64 `protobuf:"varint,16,opt,name=retired_pages_dbe,json=retiredPagesDbe,proto3" json:"retired_pages_dbe,omitempty"` RetirementPending bool `protobuf:"varint,17,opt,name=retirement_pending,json=retirementPending,proto3" json:"retirement_pending,omitempty"` RemappedRowsCorrectable int64 `protobuf:"varint,18,opt,name=remapped_rows_correctable,json=remappedRowsCorrectable,proto3" json:"remapped_rows_correctable,omitempty"` RemappedRowsUncorrectable int64 `protobuf:"varint,19,opt,name=remapped_rows_uncorrectable,json=remappedRowsUncorrectable,proto3" json:"remapped_rows_uncorrectable,omitempty"` RemapPending bool `protobuf:"varint,20,opt,name=remap_pending,json=remapPending,proto3" json:"remap_pending,omitempty"` RemapFailure bool `protobuf:"varint,21,opt,name=remap_failure,json=remapFailure,proto3" json:"remap_failure,omitempty"` NvlinkCrcErrors int64 `protobuf:"varint,22,opt,name=nvlink_crc_errors,json=nvlinkCrcErrors,proto3" json:"nvlink_crc_errors,omitempty"` NvlinkReplayErrors int64 `protobuf:"varint,23,opt,name=nvlink_replay_errors,json=nvlinkReplayErrors,proto3" json:"nvlink_replay_errors,omitempty"` NvlinkRecoveryErrors int64 `protobuf:"varint,24,opt,name=nvlink_recovery_errors,json=nvlinkRecoveryErrors,proto3" json:"nvlink_recovery_errors,omitempty"` // xid_errors are the driver's error events since the agent started. XidErrors int64 `protobuf:"varint,25,opt,name=xid_errors,json=xidErrors,proto3" json:"xid_errors,omitempty"` LastXid int32 `protobuf:"varint,26,opt,name=last_xid,json=lastXid,proto3" json:"last_xid,omitempty"` LastXidText string `protobuf:"bytes,27,opt,name=last_xid_text,json=lastXidText,proto3" json:"last_xid_text,omitempty"` LastXidUnix int64 `protobuf:"varint,28,opt,name=last_xid_unix,json=lastXidUnix,proto3" json:"last_xid_unix,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *GPUSensor) Reset() { *x = GPUSensor{} mi := &file_llmgw_v1_agent_proto_msgTypes[16] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *GPUSensor) String() string { return protoimpl.X.MessageStringOf(x) } func (*GPUSensor) ProtoMessage() {} func (x *GPUSensor) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[16] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use GPUSensor.ProtoReflect.Descriptor instead. func (*GPUSensor) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{16} } func (x *GPUSensor) GetIndex() int32 { if x != nil { return x.Index } return 0 } func (x *GPUSensor) GetTempC() float64 { if x != nil { return x.TempC } return 0 } func (x *GPUSensor) GetMemoryTempC() float64 { if x != nil { return x.MemoryTempC } return 0 } func (x *GPUSensor) GetPowerW() float64 { if x != nil { return x.PowerW } return 0 } func (x *GPUSensor) GetPowerLimitW() float64 { if x != nil { return x.PowerLimitW } return 0 } func (x *GPUSensor) GetThrottle() []string { if x != nil { return x.Throttle } return nil } func (x *GPUSensor) GetUtilizationPercent() float64 { if x != nil { return x.UtilizationPercent } return 0 } func (x *GPUSensor) GetMemoryUtilizationPercent() float64 { if x != nil { return x.MemoryUtilizationPercent } return 0 } func (x *GPUSensor) GetMemoryUsedMb() float64 { if x != nil { return x.MemoryUsedMb } return 0 } func (x *GPUSensor) GetMemoryTotalMb() float64 { if x != nil { return x.MemoryTotalMb } return 0 } func (x *GPUSensor) GetSmClockMhz() float64 { if x != nil { return x.SmClockMhz } return 0 } func (x *GPUSensor) GetEccUncorrected() float64 { if x != nil { return x.EccUncorrected } return 0 } func (x *GPUSensor) GetHasWear() bool { if x != nil { return x.HasWear } return false } func (x *GPUSensor) GetPcieReplays() int64 { if x != nil { return x.PcieReplays } return 0 } func (x *GPUSensor) GetRetiredPagesSbe() int64 { if x != nil { return x.RetiredPagesSbe } return 0 } func (x *GPUSensor) GetRetiredPagesDbe() int64 { if x != nil { return x.RetiredPagesDbe } return 0 } func (x *GPUSensor) GetRetirementPending() bool { if x != nil { return x.RetirementPending } return false } func (x *GPUSensor) GetRemappedRowsCorrectable() int64 { if x != nil { return x.RemappedRowsCorrectable } return 0 } func (x *GPUSensor) GetRemappedRowsUncorrectable() int64 { if x != nil { return x.RemappedRowsUncorrectable } return 0 } func (x *GPUSensor) GetRemapPending() bool { if x != nil { return x.RemapPending } return false } func (x *GPUSensor) GetRemapFailure() bool { if x != nil { return x.RemapFailure } return false } func (x *GPUSensor) GetNvlinkCrcErrors() int64 { if x != nil { return x.NvlinkCrcErrors } return 0 } func (x *GPUSensor) GetNvlinkReplayErrors() int64 { if x != nil { return x.NvlinkReplayErrors } return 0 } func (x *GPUSensor) GetNvlinkRecoveryErrors() int64 { if x != nil { return x.NvlinkRecoveryErrors } return 0 } func (x *GPUSensor) GetXidErrors() int64 { if x != nil { return x.XidErrors } return 0 } func (x *GPUSensor) GetLastXid() int32 { if x != nil { return x.LastXid } return 0 } func (x *GPUSensor) GetLastXidText() string { if x != nil { return x.LastXidText } return "" } func (x *GPUSensor) GetLastXidUnix() int64 { if x != nil { return x.LastXidUnix } return 0 } // ShardLoad is one engine process group's use. type ShardLoad struct { state protoimpl.MessageState `protogen:"open.v1"` ShardId string `protobuf:"bytes,1,opt,name=shard_id,json=shardId,proto3" json:"shard_id,omitempty"` // cpu_cores is in cores, the way top counts a process: 2.5 is two and a // half cores busy, which a percentage of the machine would hide on a large // one. CpuCores float64 `protobuf:"fixed64,2,opt,name=cpu_cores,json=cpuCores,proto3" json:"cpu_cores,omitempty"` MemoryRssBytes int64 `protobuf:"varint,3,opt,name=memory_rss_bytes,json=memoryRssBytes,proto3" json:"memory_rss_bytes,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *ShardLoad) Reset() { *x = ShardLoad{} mi := &file_llmgw_v1_agent_proto_msgTypes[17] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *ShardLoad) String() string { return protoimpl.X.MessageStringOf(x) } func (*ShardLoad) ProtoMessage() {} func (x *ShardLoad) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[17] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use ShardLoad.ProtoReflect.Descriptor instead. func (*ShardLoad) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{17} } func (x *ShardLoad) GetShardId() string { if x != nil { return x.ShardId } return "" } func (x *ShardLoad) GetCpuCores() float64 { if x != nil { return x.CpuCores } return 0 } func (x *ShardLoad) GetMemoryRssBytes() int64 { if x != nil { return x.MemoryRssBytes } return 0 } // IBPort is one InfiniBand or RoCE port's state and counters. type IBPort struct { state protoimpl.MessageState `protogen:"open.v1"` Hca string `protobuf:"bytes,1,opt,name=hca,proto3" json:"hca,omitempty"` Port int32 `protobuf:"varint,2,opt,name=port,proto3" json:"port,omitempty"` // state is ACTIVE, INIT, DOWN...; phys_state tells a cable unplugged // (Polling, Disabled) from a link up with no subnet manager (LinkUp and INIT). State string `protobuf:"bytes,3,opt,name=state,proto3" json:"state,omitempty"` PhysState string `protobuf:"bytes,4,opt,name=phys_state,json=physState,proto3" json:"phys_state,omitempty"` RateGbps int32 `protobuf:"varint,5,opt,name=rate_gbps,json=rateGbps,proto3" json:"rate_gbps,omitempty"` // link_layer is InfiniBand or Ethernet (RoCE). LinkLayer string `protobuf:"bytes,6,opt,name=link_layer,json=linkLayer,proto3" json:"link_layer,omitempty"` // counters are cumulative, by sysfs name; data counters in bytes. Counters map[string]uint64 `protobuf:"bytes,7,rep,name=counters,proto3" json:"counters,omitempty" protobuf_key:"bytes,1,opt,name=key" protobuf_val:"varint,2,opt,name=value"` // Identity: the adapter's node GUID and the port's own, as the fabric // knows them. NodeGuid string `protobuf:"bytes,8,opt,name=node_guid,json=nodeGuid,proto3" json:"node_guid,omitempty"` PortGuid string `protobuf:"bytes,9,opt,name=port_guid,json=portGuid,proto3" json:"port_guid,omitempty"` // The subnet manager this port sees as master, on InfiniBand. Every port on // one physical fabric sees the same one, which is what tells separate // fabrics apart when their names are only what somebody typed. SmGuid string `protobuf:"bytes,10,opt,name=sm_guid,json=smGuid,proto3" json:"sm_guid,omitempty"` SmLid int32 `protobuf:"varint,11,opt,name=sm_lid,json=smLid,proto3" json:"sm_lid,omitempty"` SmPriority int32 `protobuf:"varint,12,opt,name=sm_priority,json=smPriority,proto3" json:"sm_priority,omitempty"` SmState string `protobuf:"bytes,13,opt,name=sm_state,json=smState,proto3" json:"sm_state,omitempty"` // sm_error says why sm_guid is empty on an active InfiniBand port: // infiniband-diags missing, or no subnet manager answering. SmError string `protobuf:"bytes,14,opt,name=sm_error,json=smError,proto3" json:"sm_error,omitempty"` // network is, on RoCE, the IP network of the RoCE v2 address: RoCE has no // subnet manager, and the network is what its fabric is. Network string `protobuf:"bytes,15,opt,name=network,proto3" json:"network,omitempty"` GidPrefix string `protobuf:"bytes,16,opt,name=gid_prefix,json=gidPrefix,proto3" json:"gid_prefix,omitempty"` // subnet_managers is every subnet manager the fabric's subnet // administrator lists — master and standbys — asked through one port per // fabric, so it is on that port only. It is what keeps a fabric one fabric // through a failover: the master changes, the list does not. SubnetManagers []*SubnetManager `protobuf:"bytes,17,rep,name=subnet_managers,json=subnetManagers,proto3" json:"subnet_managers,omitempty"` SubnetManagersError string `protobuf:"bytes,18,opt,name=subnet_managers_error,json=subnetManagersError,proto3" json:"subnet_managers_error,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *IBPort) Reset() { *x = IBPort{} mi := &file_llmgw_v1_agent_proto_msgTypes[18] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *IBPort) String() string { return protoimpl.X.MessageStringOf(x) } func (*IBPort) ProtoMessage() {} func (x *IBPort) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[18] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use IBPort.ProtoReflect.Descriptor instead. func (*IBPort) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{18} } func (x *IBPort) GetHca() string { if x != nil { return x.Hca } return "" } func (x *IBPort) GetPort() int32 { if x != nil { return x.Port } return 0 } func (x *IBPort) GetState() string { if x != nil { return x.State } return "" } func (x *IBPort) GetPhysState() string { if x != nil { return x.PhysState } return "" } func (x *IBPort) GetRateGbps() int32 { if x != nil { return x.RateGbps } return 0 } func (x *IBPort) GetLinkLayer() string { if x != nil { return x.LinkLayer } return "" } func (x *IBPort) GetCounters() map[string]uint64 { if x != nil { return x.Counters } return nil } func (x *IBPort) GetNodeGuid() string { if x != nil { return x.NodeGuid } return "" } func (x *IBPort) GetPortGuid() string { if x != nil { return x.PortGuid } return "" } func (x *IBPort) GetSmGuid() string { if x != nil { return x.SmGuid } return "" } func (x *IBPort) GetSmLid() int32 { if x != nil { return x.SmLid } return 0 } func (x *IBPort) GetSmPriority() int32 { if x != nil { return x.SmPriority } return 0 } func (x *IBPort) GetSmState() string { if x != nil { return x.SmState } return "" } func (x *IBPort) GetSmError() string { if x != nil { return x.SmError } return "" } func (x *IBPort) GetNetwork() string { if x != nil { return x.Network } return "" } func (x *IBPort) GetGidPrefix() string { if x != nil { return x.GidPrefix } return "" } func (x *IBPort) GetSubnetManagers() []*SubnetManager { if x != nil { return x.SubnetManagers } return nil } func (x *IBPort) GetSubnetManagersError() string { if x != nil { return x.SubnetManagersError } return "" } // SubnetManager is one subnet manager on a fabric. type SubnetManager struct { state protoimpl.MessageState `protogen:"open.v1"` Guid string `protobuf:"bytes,1,opt,name=guid,proto3" json:"guid,omitempty"` Lid int32 `protobuf:"varint,2,opt,name=lid,proto3" json:"lid,omitempty"` Priority int32 `protobuf:"varint,3,opt,name=priority,proto3" json:"priority,omitempty"` // state is master, standby, discovering or not active. State string `protobuf:"bytes,4,opt,name=state,proto3" json:"state,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *SubnetManager) Reset() { *x = SubnetManager{} mi := &file_llmgw_v1_agent_proto_msgTypes[19] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *SubnetManager) String() string { return protoimpl.X.MessageStringOf(x) } func (*SubnetManager) ProtoMessage() {} func (x *SubnetManager) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[19] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use SubnetManager.ProtoReflect.Descriptor instead. func (*SubnetManager) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{19} } func (x *SubnetManager) GetGuid() string { if x != nil { return x.Guid } return "" } func (x *SubnetManager) GetLid() int32 { if x != nil { return x.Lid } return 0 } func (x *SubnetManager) GetPriority() int32 { if x != nil { return x.Priority } return 0 } func (x *SubnetManager) GetState() string { if x != nil { return x.State } return "" } // IMEX is the nvidia-imex service and its channels. type IMEX struct { state protoimpl.MessageState `protogen:"open.v1"` // state is systemd's word for it, empty where it is not installed. State string `protobuf:"bytes,1,opt,name=state,proto3" json:"state,omitempty"` Channels int32 `protobuf:"varint,2,opt,name=channels,proto3" json:"channels,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *IMEX) Reset() { *x = IMEX{} mi := &file_llmgw_v1_agent_proto_msgTypes[20] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *IMEX) String() string { return protoimpl.X.MessageStringOf(x) } func (*IMEX) ProtoMessage() {} func (x *IMEX) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[20] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use IMEX.ProtoReflect.Descriptor instead. func (*IMEX) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{20} } func (x *IMEX) GetState() string { if x != nil { return x.State } return "" } func (x *IMEX) GetChannels() int32 { if x != nil { return x.Channels } return 0 } // GPUFabricState is one GPU's NVLink domain membership. type GPUFabricState struct { state protoimpl.MessageState `protogen:"open.v1"` Index int32 `protobuf:"varint,1,opt,name=index,proto3" json:"index,omitempty"` State string `protobuf:"bytes,2,opt,name=state,proto3" json:"state,omitempty"` Status string `protobuf:"bytes,3,opt,name=status,proto3" json:"status,omitempty"` ClusterUuid string `protobuf:"bytes,4,opt,name=cluster_uuid,json=clusterUuid,proto3" json:"cluster_uuid,omitempty"` CliqueId string `protobuf:"bytes,5,opt,name=clique_id,json=cliqueId,proto3" json:"clique_id,omitempty"` Bandwidth string `protobuf:"bytes,6,opt,name=bandwidth,proto3" json:"bandwidth,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *GPUFabricState) Reset() { *x = GPUFabricState{} mi := &file_llmgw_v1_agent_proto_msgTypes[21] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *GPUFabricState) String() string { return protoimpl.X.MessageStringOf(x) } func (*GPUFabricState) ProtoMessage() {} func (x *GPUFabricState) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[21] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use GPUFabricState.ProtoReflect.Descriptor instead. func (*GPUFabricState) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{21} } func (x *GPUFabricState) GetIndex() int32 { if x != nil { return x.Index } return 0 } func (x *GPUFabricState) GetState() string { if x != nil { return x.State } return "" } func (x *GPUFabricState) GetStatus() string { if x != nil { return x.Status } return "" } func (x *GPUFabricState) GetClusterUuid() string { if x != nil { return x.ClusterUuid } return "" } func (x *GPUFabricState) GetCliqueId() string { if x != nil { return x.CliqueId } return "" } func (x *GPUFabricState) GetBandwidth() string { if x != nil { return x.Bandwidth } return "" } type Heartbeat struct { state protoimpl.MessageState `protogen:"open.v1"` NodeId string `protobuf:"bytes,1,opt,name=node_id,json=nodeId,proto3" json:"node_id,omitempty"` // gpu_memory_used_mb, indexed by GPU, catches a foreign process appearing on // a card after registration. GpuMemoryUsedMb []int32 `protobuf:"varint,2,rep,packed,name=gpu_memory_used_mb,json=gpuMemoryUsedMb,proto3" json:"gpu_memory_used_mb,omitempty"` NewProblems []string `protobuf:"bytes,3,rep,name=new_problems,json=newProblems,proto3" json:"new_problems,omitempty"` // storage for the same reason the GPU memory is here: it changes while the // agent runs, and a figure sampled once at registration is a figure that is // wrong on exactly the node worth watching — the one busy filling up. // // Always sent, because it is a few numbers per filesystem. Storage []*Storage `protobuf:"bytes,4,rep,name=storage,proto3" json:"storage,omitempty"` // weights only when they changed, because the list is long enough that // sending it every few seconds would be paying for nothing most of the time. // // weights_changed is what says the list is meant: without it an empty list // would be ambiguous between "this node now holds nothing" and "nothing to // report", and the first would quietly empty a node's inventory. WeightsChanged bool `protobuf:"varint,5,opt,name=weights_changed,json=weightsChanged,proto3" json:"weights_changed,omitempty"` Weights []*LocalWeights `protobuf:"bytes,6,rep,name=weights,proto3" json:"weights,omitempty"` // memory_total_mb for the same reason again: RAM is added to a virtual // machine while it runs, or across a reboot the agent does not notice, and a // figure read once at registration kept a node that had grown to 32 GB // looking like the 7 GB machine it used to be. 0 means not read. MemoryTotalMb int64 `protobuf:"varint,7,opt,name=memory_total_mb,json=memoryTotalMb,proto3" json:"memory_total_mb,omitempty"` // engine_versions only when they changed, like weights. An engine installed // after the agent started is otherwise invisible to placement until // somebody restarts the agent — which is a step nobody knows to take. EnginesChanged bool `protobuf:"varint,8,opt,name=engines_changed,json=enginesChanged,proto3" json:"engines_changed,omitempty"` EngineVersions map[string]string `protobuf:"bytes,9,rep,name=engine_versions,json=engineVersions,proto3" json:"engine_versions,omitempty" protobuf_key:"bytes,1,opt,name=key" protobuf_val:"bytes,2,opt,name=value"` // provision is where installing the node's software stands, sent when it // changed. The flag makes an empty list mean "nothing to install" rather // than "nothing to say", as for weights. ProvisionChanged bool `protobuf:"varint,10,opt,name=provision_changed,json=provisionChanged,proto3" json:"provision_changed,omitempty"` Provision []*ProvisionStatus `protobuf:"bytes,11,rep,name=provision,proto3" json:"provision,omitempty"` // load is how busy the machine is, sampled over the interval since the // previous heartbeat. Absent on the first one, which has nothing to take a // difference from, and from an agent on a system without /proc. Load *NodeLoad `protobuf:"bytes,12,opt,name=load,proto3" json:"load,omitempty"` // ib_ports is every InfiniBand or RoCE port as it stands now, re-read on // each heartbeat: a port that goes down after the agent started used to go // unnoticed until the agent restarted. IbPorts []*IBPort `protobuf:"bytes,13,rep,name=ib_ports,json=ibPorts,proto3" json:"ib_ports,omitempty"` // fabrics is the fabrics this node can now carry traffic on, sent only when // that changed since registration or the last heartbeat that said so — a // port going down takes its fabric away, and one coming back returns it. FabricsChanged bool `protobuf:"varint,14,opt,name=fabrics_changed,json=fabricsChanged,proto3" json:"fabrics_changed,omitempty"` Fabrics []*FabricAttachment `protobuf:"bytes,15,rep,name=fabrics,proto3" json:"fabrics,omitempty"` // gpu_fabric is each GPU's NVLink domain membership as it stands now, sent // once a minute: a fabric manager restarting drops every card out of its // domain, and the node has to stop being offered work that spans it. GpuFabric []*GPUFabricState `protobuf:"bytes,16,rep,name=gpu_fabric,json=gpuFabric,proto3" json:"gpu_fabric,omitempty"` Imex *IMEX `protobuf:"bytes,17,opt,name=imex,proto3" json:"imex,omitempty"` // cpu_count as in Register, for a VM given more CPUs while it runs. CpuCount int32 `protobuf:"varint,18,opt,name=cpu_count,json=cpuCount,proto3" json:"cpu_count,omitempty"` // gpu_foreign_memory_mb, indexed by GPU, is what processes that are not // this agent's shards hold on each card, read afresh every heartbeat. What // keeps a card out of placement for another tenant is decided from this // rather than from registration: a card that had something else on it when // the agent started is not taken forever, and one that gains a tenant later // is not handed out. Empty from an agent that does not measure it, or from // one that could not this time. GpuForeignMemoryMb []int32 `protobuf:"varint,19,rep,packed,name=gpu_foreign_memory_mb,json=gpuForeignMemoryMb,proto3" json:"gpu_foreign_memory_mb,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *Heartbeat) Reset() { *x = Heartbeat{} mi := &file_llmgw_v1_agent_proto_msgTypes[22] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *Heartbeat) String() string { return protoimpl.X.MessageStringOf(x) } func (*Heartbeat) ProtoMessage() {} func (x *Heartbeat) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[22] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use Heartbeat.ProtoReflect.Descriptor instead. func (*Heartbeat) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{22} } func (x *Heartbeat) GetNodeId() string { if x != nil { return x.NodeId } return "" } func (x *Heartbeat) GetGpuMemoryUsedMb() []int32 { if x != nil { return x.GpuMemoryUsedMb } return nil } func (x *Heartbeat) GetNewProblems() []string { if x != nil { return x.NewProblems } return nil } func (x *Heartbeat) GetStorage() []*Storage { if x != nil { return x.Storage } return nil } func (x *Heartbeat) GetWeightsChanged() bool { if x != nil { return x.WeightsChanged } return false } func (x *Heartbeat) GetWeights() []*LocalWeights { if x != nil { return x.Weights } return nil } func (x *Heartbeat) GetMemoryTotalMb() int64 { if x != nil { return x.MemoryTotalMb } return 0 } func (x *Heartbeat) GetEnginesChanged() bool { if x != nil { return x.EnginesChanged } return false } func (x *Heartbeat) GetEngineVersions() map[string]string { if x != nil { return x.EngineVersions } return nil } func (x *Heartbeat) GetProvisionChanged() bool { if x != nil { return x.ProvisionChanged } return false } func (x *Heartbeat) GetProvision() []*ProvisionStatus { if x != nil { return x.Provision } return nil } func (x *Heartbeat) GetLoad() *NodeLoad { if x != nil { return x.Load } return nil } func (x *Heartbeat) GetIbPorts() []*IBPort { if x != nil { return x.IbPorts } return nil } func (x *Heartbeat) GetFabricsChanged() bool { if x != nil { return x.FabricsChanged } return false } func (x *Heartbeat) GetFabrics() []*FabricAttachment { if x != nil { return x.Fabrics } return nil } func (x *Heartbeat) GetGpuFabric() []*GPUFabricState { if x != nil { return x.GpuFabric } return nil } func (x *Heartbeat) GetImex() *IMEX { if x != nil { return x.Imex } return nil } func (x *Heartbeat) GetCpuCount() int32 { if x != nil { return x.CpuCount } return 0 } func (x *Heartbeat) GetGpuForeignMemoryMb() []int32 { if x != nil { return x.GpuForeignMemoryMb } return nil } // Provision is the software a node should have: which engines, at which // versions. Nothing more. // // Everything else is the node's to work out, because only the node knows it: // which build fits its hardware, where that build is published, what its // operating system is missing, and whether the result actually runs. The agent // converges on this — installs what is absent, checks what is there — and // reports each engine through ProvisionStatus. Nothing is taken on trust from // what happens to be on the machine: an engine is offered to placement only // once the agent has installed it and seen it run. type Provision struct { state protoimpl.MessageState `protogen:"open.v1"` Engines []*EngineWant `protobuf:"bytes,1,rep,name=engines,proto3" json:"engines,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *Provision) Reset() { *x = Provision{} mi := &file_llmgw_v1_agent_proto_msgTypes[23] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *Provision) String() string { return protoimpl.X.MessageStringOf(x) } func (*Provision) ProtoMessage() {} func (x *Provision) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[23] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use Provision.ProtoReflect.Descriptor instead. func (*Provision) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{23} } func (x *Provision) GetEngines() []*EngineWant { if x != nil { return x.Engines } return nil } // EngineWant names one engine at one version. type EngineWant struct { state protoimpl.MessageState `protogen:"open.v1"` Engine string `protobuf:"bytes,1,opt,name=engine,proto3" json:"engine,omitempty"` // "llamacpp" Version string `protobuf:"bytes,2,opt,name=version,proto3" json:"version,omitempty"` // "b10991" // variant is the build to install, when one is chosen for the node rather // than left to its hardware: "cu13", "cu129". Empty is the agent's choice. Variant string `protobuf:"bytes,3,opt,name=variant,proto3" json:"variant,omitempty"` // offline_base, on a cluster with no way to the internet, is where this // engine version's files are on its mirror: // http://mirror:8081/_llmgw/engines/llamacpp/b10991. Set, the node never // asks the engine's publisher. OfflineBase string `protobuf:"bytes,4,opt,name=offline_base,json=offlineBase,proto3" json:"offline_base,omitempty"` // offline_sha256 is each of those files' checksum by name, from the // bundle the control plane trusts — not from the mirror, which serves // every node over plain HTTP. OfflineSha256 map[string]string `protobuf:"bytes,5,rep,name=offline_sha256,json=offlineSha256,proto3" json:"offline_sha256,omitempty" protobuf_key:"bytes,1,opt,name=key" protobuf_val:"bytes,2,opt,name=value"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *EngineWant) Reset() { *x = EngineWant{} mi := &file_llmgw_v1_agent_proto_msgTypes[24] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *EngineWant) String() string { return protoimpl.X.MessageStringOf(x) } func (*EngineWant) ProtoMessage() {} func (x *EngineWant) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[24] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use EngineWant.ProtoReflect.Descriptor instead. func (*EngineWant) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{24} } func (x *EngineWant) GetEngine() string { if x != nil { return x.Engine } return "" } func (x *EngineWant) GetVersion() string { if x != nil { return x.Version } return "" } func (x *EngineWant) GetVariant() string { if x != nil { return x.Variant } return "" } func (x *EngineWant) GetOfflineBase() string { if x != nil { return x.OfflineBase } return "" } func (x *EngineWant) GetOfflineSha256() map[string]string { if x != nil { return x.OfflineSha256 } return nil } // ProvisionStatus is where one engine's installation stands on a node. type ProvisionStatus struct { state protoimpl.MessageState `protogen:"open.v1"` Engine string `protobuf:"bytes,1,opt,name=engine,proto3" json:"engine,omitempty"` Version string `protobuf:"bytes,2,opt,name=version,proto3" json:"version,omitempty"` Variant string `protobuf:"bytes,3,opt,name=variant,proto3" json:"variant,omitempty"` // "installing", "ready", "failed", or "unsupported" when no build fits. State string `protobuf:"bytes,4,opt,name=state,proto3" json:"state,omitempty"` // detail says what is happening or what went wrong, in words for a person. Detail string `protobuf:"bytes,5,opt,name=detail,proto3" json:"detail,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *ProvisionStatus) Reset() { *x = ProvisionStatus{} mi := &file_llmgw_v1_agent_proto_msgTypes[25] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *ProvisionStatus) String() string { return protoimpl.X.MessageStringOf(x) } func (*ProvisionStatus) ProtoMessage() {} func (x *ProvisionStatus) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[25] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use ProvisionStatus.ProtoReflect.Descriptor instead. func (*ProvisionStatus) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{25} } func (x *ProvisionStatus) GetEngine() string { if x != nil { return x.Engine } return "" } func (x *ProvisionStatus) GetVersion() string { if x != nil { return x.Version } return "" } func (x *ProvisionStatus) GetVariant() string { if x != nil { return x.Variant } return "" } func (x *ProvisionStatus) GetState() string { if x != nil { return x.State } return "" } func (x *ProvisionStatus) GetDetail() string { if x != nil { return x.Detail } return "" } // Compute is where an engine runs its model, as it said when it loaded it. type Compute struct { state protoimpl.MessageState `protogen:"open.v1"` // backend is "CUDA", "Vulkan", "Metal" or "CPU". Backend string `protobuf:"bytes,1,opt,name=backend,proto3" json:"backend,omitempty"` // devices are the devices it named: "CUDA0: Quadro RTX 6000 (7.5)". Devices []string `protobuf:"bytes,2,rep,name=devices,proto3" json:"devices,omitempty"` // layers_offloaded of layers_total went to the devices; both 0 when the // engine did not say. LayersOffloaded int32 `protobuf:"varint,3,opt,name=layers_offloaded,json=layersOffloaded,proto3" json:"layers_offloaded,omitempty"` LayersTotal int32 `protobuf:"varint,4,opt,name=layers_total,json=layersTotal,proto3" json:"layers_total,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *Compute) Reset() { *x = Compute{} mi := &file_llmgw_v1_agent_proto_msgTypes[26] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *Compute) String() string { return protoimpl.X.MessageStringOf(x) } func (*Compute) ProtoMessage() {} func (x *Compute) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[26] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use Compute.ProtoReflect.Descriptor instead. func (*Compute) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{26} } func (x *Compute) GetBackend() string { if x != nil { return x.Backend } return "" } func (x *Compute) GetDevices() []string { if x != nil { return x.Devices } return nil } func (x *Compute) GetLayersOffloaded() int32 { if x != nil { return x.LayersOffloaded } return 0 } func (x *Compute) GetLayersTotal() int32 { if x != nil { return x.LayersTotal } return 0 } // ShardStatus reports a shard's lifecycle transition. type ShardStatus struct { state protoimpl.MessageState `protogen:"open.v1"` NodeId string `protobuf:"bytes,1,opt,name=node_id,json=nodeId,proto3" json:"node_id,omitempty"` ReplicaId string `protobuf:"bytes,2,opt,name=replica_id,json=replicaId,proto3" json:"replica_id,omitempty"` ShardId string `protobuf:"bytes,3,opt,name=shard_id,json=shardId,proto3" json:"shard_id,omitempty"` State ShardStatus_State `protobuf:"varint,4,opt,name=state,proto3,enum=llmgw.v1.ShardStatus_State" json:"state,omitempty"` Error string `protobuf:"bytes,5,opt,name=error,proto3" json:"error,omitempty"` ExitCode int32 `protobuf:"varint,6,opt,name=exit_code,json=exitCode,proto3" json:"exit_code,omitempty"` // last_output carries the tail of the engine log on a failure. When a model // fails to load this is the only diagnosis available. LastOutput []string `protobuf:"bytes,7,rep,name=last_output,json=lastOutput,proto3" json:"last_output,omitempty"` // fetch is how far a download has got, when the shard is waiting on one. // // The agent logs this already, on the node, where nobody is looking. From // the control plane a replica pulling a hundred gigabytes and one that is // wedged are the same thing — a shard that has been starting for ten // minutes — and telling them apart is the difference between waiting and // going to look. Fetch *FetchProgress `protobuf:"bytes,8,opt,name=fetch,proto3" json:"fetch,omitempty"` // compute is what the engine says it computes on, read from its own output // as it became ready: sent with STATE_READY when the engine says. Compute *Compute `protobuf:"bytes,9,opt,name=compute,proto3" json:"compute,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *ShardStatus) Reset() { *x = ShardStatus{} mi := &file_llmgw_v1_agent_proto_msgTypes[27] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *ShardStatus) String() string { return protoimpl.X.MessageStringOf(x) } func (*ShardStatus) ProtoMessage() {} func (x *ShardStatus) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[27] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use ShardStatus.ProtoReflect.Descriptor instead. func (*ShardStatus) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{27} } func (x *ShardStatus) GetNodeId() string { if x != nil { return x.NodeId } return "" } func (x *ShardStatus) GetReplicaId() string { if x != nil { return x.ReplicaId } return "" } func (x *ShardStatus) GetShardId() string { if x != nil { return x.ShardId } return "" } func (x *ShardStatus) GetState() ShardStatus_State { if x != nil { return x.State } return ShardStatus_STATE_UNSPECIFIED } func (x *ShardStatus) GetError() string { if x != nil { return x.Error } return "" } func (x *ShardStatus) GetExitCode() int32 { if x != nil { return x.ExitCode } return 0 } func (x *ShardStatus) GetLastOutput() []string { if x != nil { return x.LastOutput } return nil } func (x *ShardStatus) GetFetch() *FetchProgress { if x != nil { return x.Fetch } return nil } func (x *ShardStatus) GetCompute() *Compute { if x != nil { return x.Compute } return nil } // FetchProgress is a download in flight. type FetchProgress struct { state protoimpl.MessageState `protogen:"open.v1"` Source string `protobuf:"bytes,1,opt,name=source,proto3" json:"source,omitempty"` BytesDone int64 `protobuf:"varint,2,opt,name=bytes_done,json=bytesDone,proto3" json:"bytes_done,omitempty"` // bytes_total is 0 when the server did not say how large the file is, which // is a real answer: a percentage would then be invented, and an invented // percentage that stops moving is worse than a byte count that keeps rising. BytesTotal int64 `protobuf:"varint,3,opt,name=bytes_total,json=bytesTotal,proto3" json:"bytes_total,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *FetchProgress) Reset() { *x = FetchProgress{} mi := &file_llmgw_v1_agent_proto_msgTypes[28] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *FetchProgress) String() string { return protoimpl.X.MessageStringOf(x) } func (*FetchProgress) ProtoMessage() {} func (x *FetchProgress) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[28] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use FetchProgress.ProtoReflect.Descriptor instead. func (*FetchProgress) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{28} } func (x *FetchProgress) GetSource() string { if x != nil { return x.Source } return "" } func (x *FetchProgress) GetBytesDone() int64 { if x != nil { return x.BytesDone } return 0 } func (x *FetchProgress) GetBytesTotal() int64 { if x != nil { return x.BytesTotal } return 0 } // StartShard launches one shard. It is idempotent by shard_id: the control // plane retries, and a repeat must not produce a second process. type StartShard struct { state protoimpl.MessageState `protogen:"open.v1"` ReplicaId string `protobuf:"bytes,1,opt,name=replica_id,json=replicaId,proto3" json:"replica_id,omitempty"` ShardId string `protobuf:"bytes,2,opt,name=shard_id,json=shardId,proto3" json:"shard_id,omitempty"` Rank int32 `protobuf:"varint,3,opt,name=rank,proto3" json:"rank,omitempty"` Nnodes int32 `protobuf:"varint,4,opt,name=nnodes,proto3" json:"nnodes,omitempty"` LeaderAddr string `protobuf:"bytes,5,opt,name=leader_addr,json=leaderAddr,proto3" json:"leader_addr,omitempty"` LeaderPort int32 `protobuf:"varint,6,opt,name=leader_port,json=leaderPort,proto3" json:"leader_port,omitempty"` GpuIndices []int32 `protobuf:"varint,7,rep,packed,name=gpu_indices,json=gpuIndices,proto3" json:"gpu_indices,omitempty"` // hcas are the adapters of the chosen fabric and nothing else. Handing a // shard adapters from a second fabric lets NCCL pick one whose peer is // unreachable, and the job hangs instead of failing (plan §5.3). Hcas []string `protobuf:"bytes,8,rep,name=hcas,proto3" json:"hcas,omitempty"` MgmtInterface string `protobuf:"bytes,9,opt,name=mgmt_interface,json=mgmtInterface,proto3" json:"mgmt_interface,omitempty"` HttpPort int32 `protobuf:"varint,10,opt,name=http_port,json=httpPort,proto3" json:"http_port,omitempty"` Profile *EngineProfile `protobuf:"bytes,11,opt,name=profile,proto3" json:"profile,omitempty"` StartupTimeoutMs int64 `protobuf:"varint,12,opt,name=startup_timeout_ms,json=startupTimeoutMs,proto3" json:"startup_timeout_ms,omitempty"` // pool and fabric are control plane concepts the agent cannot derive. They // are sent so the agent can echo them back in RunningShard after a restart. Pool string `protobuf:"bytes,13,opt,name=pool,proto3" json:"pool,omitempty"` Fabric string `protobuf:"bytes,14,opt,name=fabric,proto3" json:"fabric,omitempty"` // profile_fp identifies the version of the profile, so the agent can say // which one a running shard was started from. The agent never interprets it. ProfileFp string `protobuf:"bytes,15,opt,name=profile_fp,json=profileFp,proto3" json:"profile_fp,omitempty"` // api_key is what the engine demands of every request, derived by the // control plane from the replica id. The router sends it; nothing else // that can reach the node's ports has it. ApiKey string `protobuf:"bytes,16,opt,name=api_key,json=apiKey,proto3" json:"api_key,omitempty"` // cpu_threads is how many CPU threads a shard with no GPU computes on: // the cores placement reserved for it. Zero leaves it to the agent. CpuThreads int32 `protobuf:"varint,17,opt,name=cpu_threads,json=cpuThreads,proto3" json:"cpu_threads,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *StartShard) Reset() { *x = StartShard{} mi := &file_llmgw_v1_agent_proto_msgTypes[29] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *StartShard) String() string { return protoimpl.X.MessageStringOf(x) } func (*StartShard) ProtoMessage() {} func (x *StartShard) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[29] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use StartShard.ProtoReflect.Descriptor instead. func (*StartShard) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{29} } func (x *StartShard) GetReplicaId() string { if x != nil { return x.ReplicaId } return "" } func (x *StartShard) GetShardId() string { if x != nil { return x.ShardId } return "" } func (x *StartShard) GetRank() int32 { if x != nil { return x.Rank } return 0 } func (x *StartShard) GetNnodes() int32 { if x != nil { return x.Nnodes } return 0 } func (x *StartShard) GetLeaderAddr() string { if x != nil { return x.LeaderAddr } return "" } func (x *StartShard) GetLeaderPort() int32 { if x != nil { return x.LeaderPort } return 0 } func (x *StartShard) GetGpuIndices() []int32 { if x != nil { return x.GpuIndices } return nil } func (x *StartShard) GetHcas() []string { if x != nil { return x.Hcas } return nil } func (x *StartShard) GetMgmtInterface() string { if x != nil { return x.MgmtInterface } return "" } func (x *StartShard) GetHttpPort() int32 { if x != nil { return x.HttpPort } return 0 } func (x *StartShard) GetProfile() *EngineProfile { if x != nil { return x.Profile } return nil } func (x *StartShard) GetStartupTimeoutMs() int64 { if x != nil { return x.StartupTimeoutMs } return 0 } func (x *StartShard) GetPool() string { if x != nil { return x.Pool } return "" } func (x *StartShard) GetFabric() string { if x != nil { return x.Fabric } return "" } func (x *StartShard) GetProfileFp() string { if x != nil { return x.ProfileFp } return "" } func (x *StartShard) GetApiKey() string { if x != nil { return x.ApiKey } return "" } func (x *StartShard) GetCpuThreads() int32 { if x != nil { return x.CpuThreads } return 0 } type EngineProfile struct { state protoimpl.MessageState `protogen:"open.v1"` Name string `protobuf:"bytes,1,opt,name=name,proto3" json:"name,omitempty"` Engine string `protobuf:"bytes,2,opt,name=engine,proto3" json:"engine,omitempty"` GpuClass string `protobuf:"bytes,3,opt,name=gpu_class,json=gpuClass,proto3" json:"gpu_class,omitempty"` Tp int32 `protobuf:"varint,4,opt,name=tp,proto3" json:"tp,omitempty"` Pp int32 `protobuf:"varint,5,opt,name=pp,proto3" json:"pp,omitempty"` GpusPerNode int32 `protobuf:"varint,6,opt,name=gpus_per_node,json=gpusPerNode,proto3" json:"gpus_per_node,omitempty"` Nodes int32 `protobuf:"varint,7,opt,name=nodes,proto3" json:"nodes,omitempty"` Args []string `protobuf:"bytes,8,rep,name=args,proto3" json:"args,omitempty"` Env map[string]string `protobuf:"bytes,9,rep,name=env,proto3" json:"env,omitempty" protobuf_key:"bytes,1,opt,name=key" protobuf_val:"bytes,2,opt,name=value"` ModelPath string `protobuf:"bytes,10,opt,name=model_path,json=modelPath,proto3" json:"model_path,omitempty"` ServedModelName string `protobuf:"bytes,11,opt,name=served_model_name,json=servedModelName,proto3" json:"served_model_name,omitempty"` // adapters are LoRA adapters loaded beside the model, in order: the index // is the id a request applies one by. Adapters []*LoraAdapter `protobuf:"bytes,12,rep,name=adapters,proto3" json:"adapters,omitempty"` // kv_events has a vLLM engine publish its KV cache events to the agent, for // the router's precise index. KvEvents bool `protobuf:"varint,13,opt,name=kv_events,json=kvEvents,proto3" json:"kv_events,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *EngineProfile) Reset() { *x = EngineProfile{} mi := &file_llmgw_v1_agent_proto_msgTypes[30] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *EngineProfile) String() string { return protoimpl.X.MessageStringOf(x) } func (*EngineProfile) ProtoMessage() {} func (x *EngineProfile) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[30] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use EngineProfile.ProtoReflect.Descriptor instead. func (*EngineProfile) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{30} } func (x *EngineProfile) GetName() string { if x != nil { return x.Name } return "" } func (x *EngineProfile) GetEngine() string { if x != nil { return x.Engine } return "" } func (x *EngineProfile) GetGpuClass() string { if x != nil { return x.GpuClass } return "" } func (x *EngineProfile) GetTp() int32 { if x != nil { return x.Tp } return 0 } func (x *EngineProfile) GetPp() int32 { if x != nil { return x.Pp } return 0 } func (x *EngineProfile) GetGpusPerNode() int32 { if x != nil { return x.GpusPerNode } return 0 } func (x *EngineProfile) GetNodes() int32 { if x != nil { return x.Nodes } return 0 } func (x *EngineProfile) GetArgs() []string { if x != nil { return x.Args } return nil } func (x *EngineProfile) GetEnv() map[string]string { if x != nil { return x.Env } return nil } func (x *EngineProfile) GetModelPath() string { if x != nil { return x.ModelPath } return "" } func (x *EngineProfile) GetServedModelName() string { if x != nil { return x.ServedModelName } return "" } func (x *EngineProfile) GetAdapters() []*LoraAdapter { if x != nil { return x.Adapters } return nil } func (x *EngineProfile) GetKvEvents() bool { if x != nil { return x.KvEvents } return false } // RemoveWeights deletes cached model files a node no longer needs. // // The paths are computed by the control plane from the catalog, but the agent // checks them again before unlinking anything: it deletes only files inside its // own cache directory whose names this system generated. A control plane that // is wrong, compromised or simply newer than the agent must not be able to // talk it into deleting a model an operator spent a week copying there. type RemoveWeights struct { state protoimpl.MessageState `protogen:"open.v1"` Paths []string `protobuf:"bytes,1,rep,name=paths,proto3" json:"paths,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *RemoveWeights) Reset() { *x = RemoveWeights{} mi := &file_llmgw_v1_agent_proto_msgTypes[31] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *RemoveWeights) String() string { return protoimpl.X.MessageStringOf(x) } func (*RemoveWeights) ProtoMessage() {} func (x *RemoveWeights) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[31] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use RemoveWeights.ProtoReflect.Descriptor instead. func (*RemoveWeights) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{31} } func (x *RemoveWeights) GetPaths() []string { if x != nil { return x.Paths } return nil } // StopShard terminates a shard and frees its GPUs. type StopShard struct { state protoimpl.MessageState `protogen:"open.v1"` ShardId string `protobuf:"bytes,1,opt,name=shard_id,json=shardId,proto3" json:"shard_id,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *StopShard) Reset() { *x = StopShard{} mi := &file_llmgw_v1_agent_proto_msgTypes[32] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *StopShard) String() string { return protoimpl.X.MessageStringOf(x) } func (*StopShard) ProtoMessage() {} func (x *StopShard) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[32] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use StopShard.ProtoReflect.Descriptor instead. func (*StopShard) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{32} } func (x *StopShard) GetShardId() string { if x != nil { return x.ShardId } return "" } // DrainShard stops routing to a shard while letting in-flight work finish. type DrainShard struct { state protoimpl.MessageState `protogen:"open.v1"` ShardId string `protobuf:"bytes,1,opt,name=shard_id,json=shardId,proto3" json:"shard_id,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *DrainShard) Reset() { *x = DrainShard{} mi := &file_llmgw_v1_agent_proto_msgTypes[33] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *DrainShard) String() string { return protoimpl.X.MessageStringOf(x) } func (*DrainShard) ProtoMessage() {} func (x *DrainShard) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[33] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use DrainShard.ProtoReflect.Descriptor instead. func (*DrainShard) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{33} } func (x *DrainShard) GetShardId() string { if x != nil { return x.ShardId } return "" } // Ack confirms a registration or reports that a command could not be accepted. type Ack struct { state protoimpl.MessageState `protogen:"open.v1"` MessageId string `protobuf:"bytes,1,opt,name=message_id,json=messageId,proto3" json:"message_id,omitempty"` Ok bool `protobuf:"varint,2,opt,name=ok,proto3" json:"ok,omitempty"` Error string `protobuf:"bytes,3,opt,name=error,proto3" json:"error,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *Ack) Reset() { *x = Ack{} mi := &file_llmgw_v1_agent_proto_msgTypes[34] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *Ack) String() string { return protoimpl.X.MessageStringOf(x) } func (*Ack) ProtoMessage() {} func (x *Ack) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[34] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use Ack.ProtoReflect.Descriptor instead. func (*Ack) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{34} } func (x *Ack) GetMessageId() string { if x != nil { return x.MessageId } return "" } func (x *Ack) GetOk() bool { if x != nil { return x.Ok } return false } func (x *Ack) GetError() string { if x != nil { return x.Error } return "" } // LogsRequest asks a node for the recent lines of one of its logs. // // Over the stream the agent already holds, because that is the one way into a // node every deployment has: the admin page otherwise means ssh through a jump // host to a machine whose address nobody remembers, to read a journal only // root may open. type LogsRequest struct { state protoimpl.MessageState `protogen:"open.v1"` // request_id pairs the reply with its request; several may be in flight. RequestId string `protobuf:"bytes,1,opt,name=request_id,json=requestId,proto3" json:"request_id,omitempty"` // source is "agent" for the agent's own log, or a shard id for that // engine's output. Source string `protobuf:"bytes,2,opt,name=source,proto3" json:"source,omitempty"` Lines int32 `protobuf:"varint,3,opt,name=lines,proto3" json:"lines,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *LogsRequest) Reset() { *x = LogsRequest{} mi := &file_llmgw_v1_agent_proto_msgTypes[35] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *LogsRequest) String() string { return protoimpl.X.MessageStringOf(x) } func (*LogsRequest) ProtoMessage() {} func (x *LogsRequest) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[35] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use LogsRequest.ProtoReflect.Descriptor instead. func (*LogsRequest) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{35} } func (x *LogsRequest) GetRequestId() string { if x != nil { return x.RequestId } return "" } func (x *LogsRequest) GetSource() string { if x != nil { return x.Source } return "" } func (x *LogsRequest) GetLines() int32 { if x != nil { return x.Lines } return 0 } // LogsReply answers a LogsRequest. type LogsReply struct { state protoimpl.MessageState `protogen:"open.v1"` RequestId string `protobuf:"bytes,1,opt,name=request_id,json=requestId,proto3" json:"request_id,omitempty"` Lines []string `protobuf:"bytes,2,rep,name=lines,proto3" json:"lines,omitempty"` // sources are every log this node can show, so a page can offer them // without knowing what runs there. Sources []string `protobuf:"bytes,3,rep,name=sources,proto3" json:"sources,omitempty"` // error says why the lines are absent: an unknown source, say. Error string `protobuf:"bytes,4,opt,name=error,proto3" json:"error,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *LogsReply) Reset() { *x = LogsReply{} mi := &file_llmgw_v1_agent_proto_msgTypes[36] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *LogsReply) String() string { return protoimpl.X.MessageStringOf(x) } func (*LogsReply) ProtoMessage() {} func (x *LogsReply) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[36] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use LogsReply.ProtoReflect.Descriptor instead. func (*LogsReply) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{36} } func (x *LogsReply) GetRequestId() string { if x != nil { return x.RequestId } return "" } func (x *LogsReply) GetLines() []string { if x != nil { return x.Lines } return nil } func (x *LogsReply) GetSources() []string { if x != nil { return x.Sources } return nil } func (x *LogsReply) GetError() string { if x != nil { return x.Error } return "" } // FabricScanRequest asks a node to map the fabric one of its ports is on, // with ibnetdiscover: every switch, every adapter and every link between // them. One node per fabric is asked, every few minutes; a fabric looks the // same from all of its ports. type FabricScanRequest struct { state protoimpl.MessageState `protogen:"open.v1"` RequestId string `protobuf:"bytes,1,opt,name=request_id,json=requestId,proto3" json:"request_id,omitempty"` Hca string `protobuf:"bytes,2,opt,name=hca,proto3" json:"hca,omitempty"` Port int32 `protobuf:"varint,3,opt,name=port,proto3" json:"port,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *FabricScanRequest) Reset() { *x = FabricScanRequest{} mi := &file_llmgw_v1_agent_proto_msgTypes[37] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *FabricScanRequest) String() string { return protoimpl.X.MessageStringOf(x) } func (*FabricScanRequest) ProtoMessage() {} func (x *FabricScanRequest) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[37] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use FabricScanRequest.ProtoReflect.Descriptor instead. func (*FabricScanRequest) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{37} } func (x *FabricScanRequest) GetRequestId() string { if x != nil { return x.RequestId } return "" } func (x *FabricScanRequest) GetHca() string { if x != nil { return x.Hca } return "" } func (x *FabricScanRequest) GetPort() int32 { if x != nil { return x.Port } return 0 } // FabricScanReply is ibnetdiscover's output as it printed it. Parsed by the // control plane, so a format the parser does not yet know is a control plane // upgrade rather than one on every node. type FabricScanReply struct { state protoimpl.MessageState `protogen:"open.v1"` RequestId string `protobuf:"bytes,1,opt,name=request_id,json=requestId,proto3" json:"request_id,omitempty"` Output string `protobuf:"bytes,2,opt,name=output,proto3" json:"output,omitempty"` Error string `protobuf:"bytes,3,opt,name=error,proto3" json:"error,omitempty"` TookMs int64 `protobuf:"varint,4,opt,name=took_ms,json=tookMs,proto3" json:"took_ms,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *FabricScanReply) Reset() { *x = FabricScanReply{} mi := &file_llmgw_v1_agent_proto_msgTypes[38] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *FabricScanReply) String() string { return protoimpl.X.MessageStringOf(x) } func (*FabricScanReply) ProtoMessage() {} func (x *FabricScanReply) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[38] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use FabricScanReply.ProtoReflect.Descriptor instead. func (*FabricScanReply) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{38} } func (x *FabricScanReply) GetRequestId() string { if x != nil { return x.RequestId } return "" } func (x *FabricScanReply) GetOutput() string { if x != nil { return x.Output } return "" } func (x *FabricScanReply) GetError() string { if x != nil { return x.Error } return "" } func (x *FabricScanReply) GetTookMs() int64 { if x != nil { return x.TookMs } return 0 } // FabricTestRequest asks a node for one side of an RDMA bandwidth test // between two nodes: ib_write_bw, from perftest, over one adapter. The // server side is asked first and answers when the test is over; the client // dials it. What it proves is what a replica spanning nodes needs and nothing // on one node can show: that these two adapters reach each other through the // fabric, and how fast. type FabricTestRequest struct { state protoimpl.MessageState `protogen:"open.v1"` RequestId string `protobuf:"bytes,1,opt,name=request_id,json=requestId,proto3" json:"request_id,omitempty"` // role is "server" or "client". Role string `protobuf:"bytes,2,opt,name=role,proto3" json:"role,omitempty"` Hca string `protobuf:"bytes,3,opt,name=hca,proto3" json:"hca,omitempty"` Port int32 `protobuf:"varint,4,opt,name=port,proto3" json:"port,omitempty"` // peer is the server's management address, for the client. perftest // exchanges its connection details over TCP before any RDMA. Peer string `protobuf:"bytes,5,opt,name=peer,proto3" json:"peer,omitempty"` // tcp_port is that exchange's port, the same on both sides. TcpPort int32 `protobuf:"varint,6,opt,name=tcp_port,json=tcpPort,proto3" json:"tcp_port,omitempty"` Seconds int32 `protobuf:"varint,7,opt,name=seconds,proto3" json:"seconds,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *FabricTestRequest) Reset() { *x = FabricTestRequest{} mi := &file_llmgw_v1_agent_proto_msgTypes[39] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *FabricTestRequest) String() string { return protoimpl.X.MessageStringOf(x) } func (*FabricTestRequest) ProtoMessage() {} func (x *FabricTestRequest) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[39] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use FabricTestRequest.ProtoReflect.Descriptor instead. func (*FabricTestRequest) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{39} } func (x *FabricTestRequest) GetRequestId() string { if x != nil { return x.RequestId } return "" } func (x *FabricTestRequest) GetRole() string { if x != nil { return x.Role } return "" } func (x *FabricTestRequest) GetHca() string { if x != nil { return x.Hca } return "" } func (x *FabricTestRequest) GetPort() int32 { if x != nil { return x.Port } return 0 } func (x *FabricTestRequest) GetPeer() string { if x != nil { return x.Peer } return "" } func (x *FabricTestRequest) GetTcpPort() int32 { if x != nil { return x.TcpPort } return 0 } func (x *FabricTestRequest) GetSeconds() int32 { if x != nil { return x.Seconds } return 0 } // FabricTestReply is what one side measured. type FabricTestReply struct { state protoimpl.MessageState `protogen:"open.v1"` RequestId string `protobuf:"bytes,1,opt,name=request_id,json=requestId,proto3" json:"request_id,omitempty"` // gbps is the average bandwidth, in gigabits a second. Gbps float64 `protobuf:"fixed64,2,opt,name=gbps,proto3" json:"gbps,omitempty"` // output is the end of the tool's output, for when the number is not // enough. Output string `protobuf:"bytes,3,opt,name=output,proto3" json:"output,omitempty"` Error string `protobuf:"bytes,4,opt,name=error,proto3" json:"error,omitempty"` TookMs int64 `protobuf:"varint,5,opt,name=took_ms,json=tookMs,proto3" json:"took_ms,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *FabricTestReply) Reset() { *x = FabricTestReply{} mi := &file_llmgw_v1_agent_proto_msgTypes[40] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *FabricTestReply) String() string { return protoimpl.X.MessageStringOf(x) } func (*FabricTestReply) ProtoMessage() {} func (x *FabricTestReply) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[40] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use FabricTestReply.ProtoReflect.Descriptor instead. func (*FabricTestReply) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{40} } func (x *FabricTestReply) GetRequestId() string { if x != nil { return x.RequestId } return "" } func (x *FabricTestReply) GetGbps() float64 { if x != nil { return x.Gbps } return 0 } func (x *FabricTestReply) GetOutput() string { if x != nil { return x.Output } return "" } func (x *FabricTestReply) GetError() string { if x != nil { return x.Error } return "" } func (x *FabricTestReply) GetTookMs() int64 { if x != nil { return x.TookMs } return 0 } // UpgradeAgent tells an agent to replace its own binary with a release and // restart. // // The agent fetches the release for its own platform and checks it against // the release's checksums before replacing anything, keeps the old binary // beside the new, then exits so systemd starts the new one. Engines keep // running throughout, as for any agent restart, and are adopted on the way // back. Over the stream the agent already holds, so upgrading a fleet needs // no ssh into it. type UpgradeAgent struct { state protoimpl.MessageState `protogen:"open.v1"` RequestId string `protobuf:"bytes,1,opt,name=request_id,json=requestId,proto3" json:"request_id,omitempty"` // version must be a release: v0.10.4 or v0.10.4-rc.2. Version string `protobuf:"bytes,2,opt,name=version,proto3" json:"version,omitempty"` // releases is the base URL release files are downloaded from. Releases string `protobuf:"bytes,3,opt,name=releases,proto3" json:"releases,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *UpgradeAgent) Reset() { *x = UpgradeAgent{} mi := &file_llmgw_v1_agent_proto_msgTypes[41] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *UpgradeAgent) String() string { return protoimpl.X.MessageStringOf(x) } func (*UpgradeAgent) ProtoMessage() {} func (x *UpgradeAgent) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[41] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use UpgradeAgent.ProtoReflect.Descriptor instead. func (*UpgradeAgent) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{41} } func (x *UpgradeAgent) GetRequestId() string { if x != nil { return x.RequestId } return "" } func (x *UpgradeAgent) GetVersion() string { if x != nil { return x.Version } return "" } func (x *UpgradeAgent) GetReleases() string { if x != nil { return x.Releases } return "" } // UpgradeReply says whether the new binary is in place. Sent before the // agent exits; an empty error means it will be back on the new version. type UpgradeReply struct { state protoimpl.MessageState `protogen:"open.v1"` RequestId string `protobuf:"bytes,1,opt,name=request_id,json=requestId,proto3" json:"request_id,omitempty"` Version string `protobuf:"bytes,2,opt,name=version,proto3" json:"version,omitempty"` Error string `protobuf:"bytes,3,opt,name=error,proto3" json:"error,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *UpgradeReply) Reset() { *x = UpgradeReply{} mi := &file_llmgw_v1_agent_proto_msgTypes[42] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *UpgradeReply) String() string { return protoimpl.X.MessageStringOf(x) } func (*UpgradeReply) ProtoMessage() {} func (x *UpgradeReply) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[42] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use UpgradeReply.ProtoReflect.Descriptor instead. func (*UpgradeReply) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{42} } func (x *UpgradeReply) GetRequestId() string { if x != nil { return x.RequestId } return "" } func (x *UpgradeReply) GetVersion() string { if x != nil { return x.Version } return "" } func (x *UpgradeReply) GetError() string { if x != nil { return x.Error } return "" } // LoraAdapter is one adapter: the name clients ask for it by, and its source, // which the agent fetches as it does the model's weights. type LoraAdapter struct { state protoimpl.MessageState `protogen:"open.v1"` Name string `protobuf:"bytes,1,opt,name=name,proto3" json:"name,omitempty"` Source string `protobuf:"bytes,2,opt,name=source,proto3" json:"source,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *LoraAdapter) Reset() { *x = LoraAdapter{} mi := &file_llmgw_v1_agent_proto_msgTypes[43] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *LoraAdapter) String() string { return protoimpl.X.MessageStringOf(x) } func (*LoraAdapter) ProtoMessage() {} func (x *LoraAdapter) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[43] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use LoraAdapter.ProtoReflect.Descriptor instead. func (*LoraAdapter) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{43} } func (x *LoraAdapter) GetName() string { if x != nil { return x.Name } return "" } func (x *LoraAdapter) GetSource() string { if x != nil { return x.Source } return "" } // TrainStart has a node train a LoRA adapter (BACKLOG F4): on a base model's // weights from its registry, on question/answer pairs, with a Python trainer // the agent carries, in an environment it makes over an installed engine's. // One at a time per node. type TrainStart struct { state protoimpl.MessageState `protogen:"open.v1"` JobId string `protobuf:"bytes,1,opt,name=job_id,json=jobId,proto3" json:"job_id,omitempty"` // base_repo and base_revision name the base model's weights on its // registry (Hugging Face): safetensors, not the GGUF a catalog may serve. BaseRepo string `protobuf:"bytes,2,opt,name=base_repo,json=baseRepo,proto3" json:"base_repo,omitempty"` BaseRevision string `protobuf:"bytes,3,opt,name=base_revision,json=baseRevision,proto3" json:"base_revision,omitempty"` // data is the training pairs, NDJSON of {"q", "a"}. Data []byte `protobuf:"bytes,4,opt,name=data,proto3" json:"data,omitempty"` // params is the trainer's settings as JSON: rank, alpha, epochs, lr, // batch, max_len. Params string `protobuf:"bytes,5,opt,name=params,proto3" json:"params,omitempty"` // gpu_memory_mib is what the job needs of a card; the agent starts it // only when one has that much free, which lets it run beside an engine. GpuMemoryMib int32 `protobuf:"varint,6,opt,name=gpu_memory_mib,json=gpuMemoryMib,proto3" json:"gpu_memory_mib,omitempty"` // gguf asks for the adapter in llama.cpp's format too. Gguf bool `protobuf:"varint,7,opt,name=gguf,proto3" json:"gguf,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *TrainStart) Reset() { *x = TrainStart{} mi := &file_llmgw_v1_agent_proto_msgTypes[44] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *TrainStart) String() string { return protoimpl.X.MessageStringOf(x) } func (*TrainStart) ProtoMessage() {} func (x *TrainStart) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[44] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use TrainStart.ProtoReflect.Descriptor instead. func (*TrainStart) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{44} } func (x *TrainStart) GetJobId() string { if x != nil { return x.JobId } return "" } func (x *TrainStart) GetBaseRepo() string { if x != nil { return x.BaseRepo } return "" } func (x *TrainStart) GetBaseRevision() string { if x != nil { return x.BaseRevision } return "" } func (x *TrainStart) GetData() []byte { if x != nil { return x.Data } return nil } func (x *TrainStart) GetParams() string { if x != nil { return x.Params } return "" } func (x *TrainStart) GetGpuMemoryMib() int32 { if x != nil { return x.GpuMemoryMib } return 0 } func (x *TrainStart) GetGguf() bool { if x != nil { return x.Gguf } return false } // TrainCancel stops a training job. type TrainCancel struct { state protoimpl.MessageState `protogen:"open.v1"` JobId string `protobuf:"bytes,1,opt,name=job_id,json=jobId,proto3" json:"job_id,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *TrainCancel) Reset() { *x = TrainCancel{} mi := &file_llmgw_v1_agent_proto_msgTypes[45] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *TrainCancel) String() string { return protoimpl.X.MessageStringOf(x) } func (*TrainCancel) ProtoMessage() {} func (x *TrainCancel) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[45] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use TrainCancel.ProtoReflect.Descriptor instead. func (*TrainCancel) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{45} } func (x *TrainCancel) GetJobId() string { if x != nil { return x.JobId } return "" } // TrainStatus is how a training job is going, sent as it changes. type TrainStatus struct { state protoimpl.MessageState `protogen:"open.v1"` JobId string `protobuf:"bytes,1,opt,name=job_id,json=jobId,proto3" json:"job_id,omitempty"` // state is preparing, running, done or failed. State string `protobuf:"bytes,2,opt,name=state,proto3" json:"state,omitempty"` Step int32 `protobuf:"varint,3,opt,name=step,proto3" json:"step,omitempty"` Steps int32 `protobuf:"varint,4,opt,name=steps,proto3" json:"steps,omitempty"` Loss float64 `protobuf:"fixed64,5,opt,name=loss,proto3" json:"loss,omitempty"` // message is what is being done, or why it failed. Message string `protobuf:"bytes,6,opt,name=message,proto3" json:"message,omitempty"` // peak_mib is the most the job held of the card. PeakMib int32 `protobuf:"varint,7,opt,name=peak_mib,json=peakMib,proto3" json:"peak_mib,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *TrainStatus) Reset() { *x = TrainStatus{} mi := &file_llmgw_v1_agent_proto_msgTypes[46] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *TrainStatus) String() string { return protoimpl.X.MessageStringOf(x) } func (*TrainStatus) ProtoMessage() {} func (x *TrainStatus) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[46] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use TrainStatus.ProtoReflect.Descriptor instead. func (*TrainStatus) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{46} } func (x *TrainStatus) GetJobId() string { if x != nil { return x.JobId } return "" } func (x *TrainStatus) GetState() string { if x != nil { return x.State } return "" } func (x *TrainStatus) GetStep() int32 { if x != nil { return x.Step } return 0 } func (x *TrainStatus) GetSteps() int32 { if x != nil { return x.Steps } return 0 } func (x *TrainStatus) GetLoss() float64 { if x != nil { return x.Loss } return 0 } func (x *TrainStatus) GetMessage() string { if x != nil { return x.Message } return "" } func (x *TrainStatus) GetPeakMib() int32 { if x != nil { return x.PeakMib } return 0 } // TrainArtifact is a piece of what a training job made, sent before its // final status: a gzipped tar of the PEFT adapter (kind peft), or the GGUF // adapter (kind gguf). type TrainArtifact struct { state protoimpl.MessageState `protogen:"open.v1"` JobId string `protobuf:"bytes,1,opt,name=job_id,json=jobId,proto3" json:"job_id,omitempty"` Kind string `protobuf:"bytes,2,opt,name=kind,proto3" json:"kind,omitempty"` Seq int32 `protobuf:"varint,3,opt,name=seq,proto3" json:"seq,omitempty"` Data []byte `protobuf:"bytes,4,opt,name=data,proto3" json:"data,omitempty"` Last bool `protobuf:"varint,5,opt,name=last,proto3" json:"last,omitempty"` unknownFields protoimpl.UnknownFields sizeCache protoimpl.SizeCache } func (x *TrainArtifact) Reset() { *x = TrainArtifact{} mi := &file_llmgw_v1_agent_proto_msgTypes[47] ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) ms.StoreMessageInfo(mi) } func (x *TrainArtifact) String() string { return protoimpl.X.MessageStringOf(x) } func (*TrainArtifact) ProtoMessage() {} func (x *TrainArtifact) ProtoReflect() protoreflect.Message { mi := &file_llmgw_v1_agent_proto_msgTypes[47] if x != nil { ms := protoimpl.X.MessageStateOf(protoimpl.Pointer(x)) if ms.LoadMessageInfo() == nil { ms.StoreMessageInfo(mi) } return ms } return mi.MessageOf(x) } // Deprecated: Use TrainArtifact.ProtoReflect.Descriptor instead. func (*TrainArtifact) Descriptor() ([]byte, []int) { return file_llmgw_v1_agent_proto_rawDescGZIP(), []int{47} } func (x *TrainArtifact) GetJobId() string { if x != nil { return x.JobId } return "" } func (x *TrainArtifact) GetKind() string { if x != nil { return x.Kind } return "" } func (x *TrainArtifact) GetSeq() int32 { if x != nil { return x.Seq } return 0 } func (x *TrainArtifact) GetData() []byte { if x != nil { return x.Data } return nil } func (x *TrainArtifact) GetLast() bool { if x != nil { return x.Last } return false } var File_llmgw_v1_agent_proto protoreflect.FileDescriptor const file_llmgw_v1_agent_proto_rawDesc = "" + "\n" + "\x14llmgw/v1/agent.proto\x12\bllmgw.v1\"\xb7\x04\n" + "\fAgentMessage\x120\n" + "\bregister\x18\x01 \x01(\v2\x12.llmgw.v1.RegisterH\x00R\bregister\x123\n" + "\theartbeat\x18\x02 \x01(\v2\x13.llmgw.v1.HeartbeatH\x00R\theartbeat\x12-\n" + "\x05shard\x18\x03 \x01(\v2\x15.llmgw.v1.ShardStatusH\x00R\x05shard\x12)\n" + "\x04logs\x18\x04 \x01(\v2\x13.llmgw.v1.LogsReplyH\x00R\x04logs\x12<\n" + "\vfabric_scan\x18\x05 \x01(\v2\x19.llmgw.v1.FabricScanReplyH\x00R\n" + "fabricScan\x12<\n" + "\vfabric_test\x18\x06 \x01(\v2\x19.llmgw.v1.FabricTestReplyH\x00R\n" + "fabricTest\x122\n" + "\aupgrade\x18\a \x01(\v2\x16.llmgw.v1.UpgradeReplyH\x00R\aupgrade\x121\n" + "\tkv_events\x18\b \x01(\v2\x12.llmgw.v1.KVEventsH\x00R\bkvEvents\x12:\n" + "\ftrain_status\x18\t \x01(\v2\x15.llmgw.v1.TrainStatusH\x00R\vtrainStatus\x12@\n" + "\x0etrain_artifact\x18\n" + " \x01(\v2\x17.llmgw.v1.TrainArtifactH\x00R\rtrainArtifactB\x05\n" + "\x03msg\"\xb5\x01\n" + "\bKVEvents\x12\x1d\n" + "\n" + "replica_id\x18\x01 \x01(\tR\treplicaId\x12\x1d\n" + "\n" + "block_size\x18\x02 \x01(\x05R\tblockSize\x12\x1f\n" + "\vclear_first\x18\x03 \x01(\bR\n" + "clearFirst\x12\x16\n" + "\x06stored\x18\x04 \x03(\x06R\x06stored\x12\x18\n" + "\aremoved\x18\x05 \x03(\x06R\aremoved\x12\x18\n" + "\aproblem\x18\x06 \x01(\tR\aproblem\"\xf8\a\n" + "\x0eControlMessage\x12,\n" + "\x05start\x18\x01 \x01(\v2\x14.llmgw.v1.StartShardH\x00R\x05start\x12)\n" + "\x04stop\x18\x02 \x01(\v2\x13.llmgw.v1.StopShardH\x00R\x04stop\x12,\n" + "\x05drain\x18\x03 \x01(\v2\x14.llmgw.v1.DrainShardH\x00R\x05drain\x12!\n" + "\x03ack\x18\x04 \x01(\v2\r.llmgw.v1.AckH\x00R\x03ack\x12@\n" + "\x0eremove_weights\x18\x05 \x01(\v2\x17.llmgw.v1.RemoveWeightsH\x00R\rremoveWeights\x123\n" + "\tprovision\x18\x06 \x01(\v2\x13.llmgw.v1.ProvisionH\x00R\tprovision\x12+\n" + "\x04logs\x18\a \x01(\v2\x15.llmgw.v1.LogsRequestH\x00R\x04logs\x12>\n" + "\vfabric_scan\x18\b \x01(\v2\x1b.llmgw.v1.FabricScanRequestH\x00R\n" + "fabricScan\x12>\n" + "\vfabric_test\x18\t \x01(\v2\x1b.llmgw.v1.FabricTestRequestH\x00R\n" + "fabricTest\x122\n" + "\aupgrade\x18\n" + " \x01(\v2\x16.llmgw.v1.UpgradeAgentH\x00R\aupgrade\x12@\n" + "\x0econtrol_planes\x18\v \x01(\v2\x17.llmgw.v1.ControlPlanesH\x00R\rcontrolPlanes\x120\n" + "\bfirewall\x18\r \x01(\v2\x12.llmgw.v1.FirewallH\x00R\bfirewall\x12=\n" + "\rrestart_agent\x18\x0e \x01(\v2\x16.llmgw.v1.RestartAgentH\x00R\frestartAgent\x12@\n" + "\x0eagent_settings\x18\x0f \x01(\v2\x17.llmgw.v1.AgentSettingsH\x00R\ragentSettings\x127\n" + "\vtrain_start\x18\x10 \x01(\v2\x14.llmgw.v1.TrainStartH\x00R\n" + "trainStart\x12:\n" + "\ftrain_cancel\x18\x11 \x01(\v2\x15.llmgw.v1.TrainCancelH\x00R\vtrainCancel\x129\n" + "\x05trace\x18\f \x03(\v2#.llmgw.v1.ControlMessage.TraceEntryR\x05trace\x1a8\n" + "\n" + "TraceEntry\x12\x10\n" + "\x03key\x18\x01 \x01(\tR\x03key\x12\x14\n" + "\x05value\x18\x02 \x01(\tR\x05value:\x028\x01B\x05\n" + "\x03msg\"&\n" + "\fRestartAgent\x12\x16\n" + "\x06reason\x18\x01 \x01(\tR\x06reason\"\x94\x01\n" + "\bFirewall\x12\x18\n" + "\aenabled\x18\x01 \x01(\bR\aenabled\x12\x1a\n" + "\bnetworks\x18\x02 \x03(\tR\bnetworks\x12*\n" + "\x11engine_ports_from\x18\x03 \x01(\x05R\x0fenginePortsFrom\x12&\n" + "\x0fengine_ports_to\x18\x04 \x01(\x05R\renginePortsTo\"#\n" + "\rAgentSettings\x12\x12\n" + "\x04yaml\x18\x01 \x01(\tR\x04yaml\")\n" + "\rControlPlanes\x12\x18\n" + "\aservers\x18\x01 \x03(\tR\aservers\"\x80\a\n" + "\bRegister\x12\x17\n" + "\anode_id\x18\x01 \x01(\tR\x06nodeId\x12\x1a\n" + "\bhostname\x18\x02 \x01(\tR\bhostname\x12\x1b\n" + "\tmgmt_addr\x18\x03 \x01(\tR\bmgmtAddr\x126\n" + "\x06labels\x18\x04 \x03(\v2\x1e.llmgw.v1.Register.LabelsEntryR\x06labels\x12!\n" + "\x04gpus\x18\x05 \x03(\v2\r.llmgw.v1.GPUR\x04gpus\x124\n" + "\afabrics\x18\x06 \x03(\v2\x1a.llmgw.v1.FabricAttachmentR\afabrics\x12#\n" + "\ragent_version\x18\a \x01(\tR\fagentVersion\x12O\n" + "\x0fengine_versions\x18\b \x03(\v2&.llmgw.v1.Register.EngineVersionsEntryR\x0eengineVersions\x12=\n" + "\x0erunning_shards\x18\t \x03(\v2\x16.llmgw.v1.RunningShardR\rrunningShards\x12&\n" + "\x0fmemory_total_mb\x18\n" + " \x01(\x03R\rmemoryTotalMb\x120\n" + "\aweights\x18\v \x03(\v2\x16.llmgw.v1.LocalWeightsR\aweights\x12+\n" + "\astorage\x18\f \x03(\v2\x11.llmgw.v1.StorageR\astorage\x12\"\n" + "\x04host\x18\r \x01(\v2\x0e.llmgw.v1.HostR\x04host\x12%\n" + "\x0eengines_probed\x18\x0e \x01(\bR\renginesProbed\x12\x1b\n" + "\tcpu_count\x18\x12 \x01(\x05R\bcpuCount\x12\"\n" + "\x04imex\x18\x0f \x01(\v2\x0e.llmgw.v1.IMEXR\x04imex\x12/\n" + "\x13installable_engines\x18\x10 \x03(\tR\x12installableEngines\x12\x1a\n" + "\bfeatures\x18\x11 \x03(\tR\bfeatures\x1a9\n" + "\vLabelsEntry\x12\x10\n" + "\x03key\x18\x01 \x01(\tR\x03key\x12\x14\n" + "\x05value\x18\x02 \x01(\tR\x05value:\x028\x01\x1aA\n" + "\x13EngineVersionsEntry\x12\x10\n" + "\x03key\x18\x01 \x01(\tR\x03key\x12\x14\n" + "\x05value\x18\x02 \x01(\tR\x05value:\x028\x01\"\xb1\x01\n" + "\x04Host\x12\x13\n" + "\x05os_id\x18\x01 \x01(\tR\x04osId\x12\x1d\n" + "\n" + "os_version\x18\x02 \x01(\tR\tosVersion\x12\x1f\n" + "\vos_codename\x18\x03 \x01(\tR\n" + "osCodename\x12\x12\n" + "\x04arch\x18\x04 \x01(\tR\x04arch\x12\x1b\n" + "\tcpu_flags\x18\x05 \x03(\tR\bcpuFlags\x12#\n" + "\rnvidia_driver\x18\x06 \x01(\tR\fnvidiaDriver\"u\n" + "\aStorage\x12\x12\n" + "\x04path\x18\x01 \x01(\tR\x04path\x12\x1d\n" + "\n" + "free_bytes\x18\x02 \x01(\x03R\tfreeBytes\x12\x1f\n" + "\vtotal_bytes\x18\x03 \x01(\x03R\n" + "totalBytes\x12\x16\n" + "\x06shared\x18\x04 \x01(\bR\x06shared\"\xe3\x03\n" + "\fRunningShard\x12\x1d\n" + "\n" + "replica_id\x18\x01 \x01(\tR\treplicaId\x12\x19\n" + "\bshard_id\x18\x02 \x01(\tR\ashardId\x12\x12\n" + "\x04rank\x18\x03 \x01(\x05R\x04rank\x12\x16\n" + "\x06nnodes\x18\x04 \x01(\x05R\x06nnodes\x12\x1f\n" + "\vgpu_indices\x18\x05 \x03(\x05R\n" + "gpuIndices\x12\x12\n" + "\x04hcas\x18\x06 \x03(\tR\x04hcas\x12\x1b\n" + "\thttp_port\x18\a \x01(\x05R\bhttpPort\x121\n" + "\x05state\x18\b \x01(\x0e2\x1b.llmgw.v1.ShardStatus.StateR\x05state\x12\x14\n" + "\x05model\x18\t \x01(\tR\x05model\x12\x18\n" + "\aprofile\x18\n" + " \x01(\tR\aprofile\x12\x12\n" + "\x04pool\x18\v \x01(\tR\x04pool\x12\x16\n" + "\x06fabric\x18\f \x01(\tR\x06fabric\x12\x1f\n" + "\vleader_addr\x18\r \x01(\tR\n" + "leaderAddr\x12\x1f\n" + "\vleader_port\x18\x0e \x01(\x05R\n" + "leaderPort\x12\x1d\n" + "\n" + "profile_fp\x18\x0f \x01(\tR\tprofileFp\x12+\n" + "\acompute\x18\x10 \x01(\v2\x11.llmgw.v1.ComputeR\acompute\"\xca\x06\n" + "\x03GPU\x12\x14\n" + "\x05index\x18\x01 \x01(\x05R\x05index\x12\x12\n" + "\x04uuid\x18\x02 \x01(\tR\x04uuid\x12\x12\n" + "\x04name\x18\x03 \x01(\tR\x04name\x12&\n" + "\x0fmemory_total_mb\x18\x04 \x01(\x05R\rmemoryTotalMb\x12$\n" + "\x0ememory_used_mb\x18\x05 \x01(\x05R\fmemoryUsedMb\x12-\n" + "\x12compute_capability\x18\x06 \x01(\tR\x11computeCapability\x12\x1b\n" + "\tnuma_node\x18\a \x01(\x05R\bnumaNode\x12\x1c\n" + "\n" + "pci_bus_id\x18\b \x01(\tR\bpciBusId\x12%\n" + "\x0edriver_version\x18\t \x01(\tR\rdriverVersion\x12!\n" + "\fcuda_version\x18\n" + " \x01(\tR\vcudaVersion\x12!\n" + "\fnvlink_peers\x18\v \x03(\x05R\vnvlinkPeers\x12\x1a\n" + "\bproblems\x18\f \x03(\tR\bproblems\x12#\n" + "\rtemperature_c\x18\r \x01(\x05R\ftemperatureC\x12\x1d\n" + "\n" + "ecc_errors\x18\x0e \x01(\x05R\teccErrors\x12/\n" + "\x13utilization_percent\x18\x0f \x01(\x05R\x12utilizationPercent\x128\n" + "\tnic_paths\x18\x10 \x03(\v2\x1b.llmgw.v1.GPU.NicPathsEntryR\bnicPaths\x12!\n" + "\ffabric_state\x18\x11 \x01(\tR\vfabricState\x12#\n" + "\rfabric_status\x18\x12 \x01(\tR\ffabricStatus\x12!\n" + "\fcluster_uuid\x18\x13 \x01(\tR\vclusterUuid\x12\x1b\n" + "\tclique_id\x18\x14 \x01(\tR\bcliqueId\x12)\n" + "\x10fabric_bandwidth\x18\x15 \x01(\tR\x0ffabricBandwidth\x12%\n" + "\x03mig\x18\x16 \x01(\v2\x13.llmgw.v1.MIGDeviceR\x03mig\x1a;\n" + "\rNicPathsEntry\x12\x10\n" + "\x03key\x18\x01 \x01(\tR\x03key\x12\x14\n" + "\x05value\x18\x02 \x01(\tR\x05value:\x028\x01\"\x84\x02\n" + "\tMIGDevice\x12!\n" + "\fparent_index\x18\x01 \x01(\x05R\vparentIndex\x12\x1f\n" + "\vparent_uuid\x18\x02 \x01(\tR\n" + "parentUuid\x123\n" + "\x16parent_memory_total_mb\x18\x03 \x01(\x05R\x13parentMemoryTotalMb\x12\x16\n" + "\x06device\x18\x04 \x01(\x05R\x06device\x12\x18\n" + "\aprofile\x18\x05 \x01(\tR\aprofile\x12!\n" + "\fgpu_instance\x18\x06 \x01(\x05R\vgpuInstance\x12)\n" + "\x10compute_instance\x18\a \x01(\x05R\x0fcomputeInstance\"[\n" + "\x10FabricAttachment\x12\x16\n" + "\x06fabric\x18\x01 \x01(\tR\x06fabric\x12\x12\n" + "\x04hcas\x18\x02 \x03(\tR\x04hcas\x12\x1b\n" + "\trate_gbps\x18\x03 \x01(\x05R\brateGbps\"Y\n" + "\fLocalWeights\x12\x12\n" + "\x04path\x18\x01 \x01(\tR\x04path\x12\x1d\n" + "\n" + "size_bytes\x18\x02 \x01(\x03R\tsizeBytes\x12\x16\n" + "\x06shared\x18\x03 \x01(\bR\x06shared\"\xa4\x03\n" + "\bNodeLoad\x12\x1f\n" + "\vcpu_percent\x18\x01 \x01(\x01R\n" + "cpuPercent\x12\x1b\n" + "\tcpu_count\x18\x02 \x01(\x05R\bcpuCount\x12\x14\n" + "\x05load1\x18\x03 \x01(\x01R\x05load1\x12\x14\n" + "\x05load5\x18\x04 \x01(\x01R\x05load5\x12\x16\n" + "\x06load15\x18\x05 \x01(\x01R\x06load15\x12*\n" + "\x11memory_used_bytes\x18\x06 \x01(\x03R\x0fmemoryUsedBytes\x124\n" + "\x16memory_available_bytes\x18\a \x01(\x03R\x14memoryAvailableBytes\x12+\n" + "\x06shards\x18\b \x03(\v2\x13.llmgw.v1.ShardLoadR\x06shards\x12\x1c\n" + "\n" + "cpu_temp_c\x18\t \x01(\x01R\bcpuTempC\x12\x1e\n" + "\vcpu_power_w\x18\n" + " \x01(\x01R\tcpuPowerW\x12 \n" + "\finlet_temp_c\x18\v \x01(\x01R\n" + "inletTempC\x12'\n" + "\x04gpus\x18\f \x03(\v2\x13.llmgw.v1.GPUSensorR\x04gpus\"\xde\b\n" + "\tGPUSensor\x12\x14\n" + "\x05index\x18\x01 \x01(\x05R\x05index\x12\x15\n" + "\x06temp_c\x18\x02 \x01(\x01R\x05tempC\x12\"\n" + "\rmemory_temp_c\x18\x03 \x01(\x01R\vmemoryTempC\x12\x17\n" + "\apower_w\x18\x04 \x01(\x01R\x06powerW\x12\"\n" + "\rpower_limit_w\x18\x05 \x01(\x01R\vpowerLimitW\x12\x1a\n" + "\bthrottle\x18\x06 \x03(\tR\bthrottle\x12/\n" + "\x13utilization_percent\x18\a \x01(\x01R\x12utilizationPercent\x12<\n" + "\x1amemory_utilization_percent\x18\b \x01(\x01R\x18memoryUtilizationPercent\x12$\n" + "\x0ememory_used_mb\x18\t \x01(\x01R\fmemoryUsedMb\x12&\n" + "\x0fmemory_total_mb\x18\n" + " \x01(\x01R\rmemoryTotalMb\x12 \n" + "\fsm_clock_mhz\x18\v \x01(\x01R\n" + "smClockMhz\x12'\n" + "\x0fecc_uncorrected\x18\f \x01(\x01R\x0eeccUncorrected\x12\x19\n" + "\bhas_wear\x18\r \x01(\bR\ahasWear\x12!\n" + "\fpcie_replays\x18\x0e \x01(\x03R\vpcieReplays\x12*\n" + "\x11retired_pages_sbe\x18\x0f \x01(\x03R\x0fretiredPagesSbe\x12*\n" + "\x11retired_pages_dbe\x18\x10 \x01(\x03R\x0fretiredPagesDbe\x12-\n" + "\x12retirement_pending\x18\x11 \x01(\bR\x11retirementPending\x12:\n" + "\x19remapped_rows_correctable\x18\x12 \x01(\x03R\x17remappedRowsCorrectable\x12>\n" + "\x1bremapped_rows_uncorrectable\x18\x13 \x01(\x03R\x19remappedRowsUncorrectable\x12#\n" + "\rremap_pending\x18\x14 \x01(\bR\fremapPending\x12#\n" + "\rremap_failure\x18\x15 \x01(\bR\fremapFailure\x12*\n" + "\x11nvlink_crc_errors\x18\x16 \x01(\x03R\x0fnvlinkCrcErrors\x120\n" + "\x14nvlink_replay_errors\x18\x17 \x01(\x03R\x12nvlinkReplayErrors\x124\n" + "\x16nvlink_recovery_errors\x18\x18 \x01(\x03R\x14nvlinkRecoveryErrors\x12\x1d\n" + "\n" + "xid_errors\x18\x19 \x01(\x03R\txidErrors\x12\x19\n" + "\blast_xid\x18\x1a \x01(\x05R\alastXid\x12\"\n" + "\rlast_xid_text\x18\x1b \x01(\tR\vlastXidText\x12\"\n" + "\rlast_xid_unix\x18\x1c \x01(\x03R\vlastXidUnix\"m\n" + "\tShardLoad\x12\x19\n" + "\bshard_id\x18\x01 \x01(\tR\ashardId\x12\x1b\n" + "\tcpu_cores\x18\x02 \x01(\x01R\bcpuCores\x12(\n" + "\x10memory_rss_bytes\x18\x03 \x01(\x03R\x0ememoryRssBytes\"\x88\x05\n" + "\x06IBPort\x12\x10\n" + "\x03hca\x18\x01 \x01(\tR\x03hca\x12\x12\n" + "\x04port\x18\x02 \x01(\x05R\x04port\x12\x14\n" + "\x05state\x18\x03 \x01(\tR\x05state\x12\x1d\n" + "\n" + "phys_state\x18\x04 \x01(\tR\tphysState\x12\x1b\n" + "\trate_gbps\x18\x05 \x01(\x05R\brateGbps\x12\x1d\n" + "\n" + "link_layer\x18\x06 \x01(\tR\tlinkLayer\x12:\n" + "\bcounters\x18\a \x03(\v2\x1e.llmgw.v1.IBPort.CountersEntryR\bcounters\x12\x1b\n" + "\tnode_guid\x18\b \x01(\tR\bnodeGuid\x12\x1b\n" + "\tport_guid\x18\t \x01(\tR\bportGuid\x12\x17\n" + "\asm_guid\x18\n" + " \x01(\tR\x06smGuid\x12\x15\n" + "\x06sm_lid\x18\v \x01(\x05R\x05smLid\x12\x1f\n" + "\vsm_priority\x18\f \x01(\x05R\n" + "smPriority\x12\x19\n" + "\bsm_state\x18\r \x01(\tR\asmState\x12\x19\n" + "\bsm_error\x18\x0e \x01(\tR\asmError\x12\x18\n" + "\anetwork\x18\x0f \x01(\tR\anetwork\x12\x1d\n" + "\n" + "gid_prefix\x18\x10 \x01(\tR\tgidPrefix\x12@\n" + "\x0fsubnet_managers\x18\x11 \x03(\v2\x17.llmgw.v1.SubnetManagerR\x0esubnetManagers\x122\n" + "\x15subnet_managers_error\x18\x12 \x01(\tR\x13subnetManagersError\x1a;\n" + "\rCountersEntry\x12\x10\n" + "\x03key\x18\x01 \x01(\tR\x03key\x12\x14\n" + "\x05value\x18\x02 \x01(\x04R\x05value:\x028\x01\"g\n" + "\rSubnetManager\x12\x12\n" + "\x04guid\x18\x01 \x01(\tR\x04guid\x12\x10\n" + "\x03lid\x18\x02 \x01(\x05R\x03lid\x12\x1a\n" + "\bpriority\x18\x03 \x01(\x05R\bpriority\x12\x14\n" + "\x05state\x18\x04 \x01(\tR\x05state\"8\n" + "\x04IMEX\x12\x14\n" + "\x05state\x18\x01 \x01(\tR\x05state\x12\x1a\n" + "\bchannels\x18\x02 \x01(\x05R\bchannels\"\xb2\x01\n" + "\x0eGPUFabricState\x12\x14\n" + "\x05index\x18\x01 \x01(\x05R\x05index\x12\x14\n" + "\x05state\x18\x02 \x01(\tR\x05state\x12\x16\n" + "\x06status\x18\x03 \x01(\tR\x06status\x12!\n" + "\fcluster_uuid\x18\x04 \x01(\tR\vclusterUuid\x12\x1b\n" + "\tclique_id\x18\x05 \x01(\tR\bcliqueId\x12\x1c\n" + "\tbandwidth\x18\x06 \x01(\tR\tbandwidth\"\xa9\a\n" + "\tHeartbeat\x12\x17\n" + "\anode_id\x18\x01 \x01(\tR\x06nodeId\x12+\n" + "\x12gpu_memory_used_mb\x18\x02 \x03(\x05R\x0fgpuMemoryUsedMb\x12!\n" + "\fnew_problems\x18\x03 \x03(\tR\vnewProblems\x12+\n" + "\astorage\x18\x04 \x03(\v2\x11.llmgw.v1.StorageR\astorage\x12'\n" + "\x0fweights_changed\x18\x05 \x01(\bR\x0eweightsChanged\x120\n" + "\aweights\x18\x06 \x03(\v2\x16.llmgw.v1.LocalWeightsR\aweights\x12&\n" + "\x0fmemory_total_mb\x18\a \x01(\x03R\rmemoryTotalMb\x12'\n" + "\x0fengines_changed\x18\b \x01(\bR\x0eenginesChanged\x12P\n" + "\x0fengine_versions\x18\t \x03(\v2'.llmgw.v1.Heartbeat.EngineVersionsEntryR\x0eengineVersions\x12+\n" + "\x11provision_changed\x18\n" + " \x01(\bR\x10provisionChanged\x127\n" + "\tprovision\x18\v \x03(\v2\x19.llmgw.v1.ProvisionStatusR\tprovision\x12&\n" + "\x04load\x18\f \x01(\v2\x12.llmgw.v1.NodeLoadR\x04load\x12+\n" + "\bib_ports\x18\r \x03(\v2\x10.llmgw.v1.IBPortR\aibPorts\x12'\n" + "\x0ffabrics_changed\x18\x0e \x01(\bR\x0efabricsChanged\x124\n" + "\afabrics\x18\x0f \x03(\v2\x1a.llmgw.v1.FabricAttachmentR\afabrics\x127\n" + "\n" + "gpu_fabric\x18\x10 \x03(\v2\x18.llmgw.v1.GPUFabricStateR\tgpuFabric\x12\"\n" + "\x04imex\x18\x11 \x01(\v2\x0e.llmgw.v1.IMEXR\x04imex\x12\x1b\n" + "\tcpu_count\x18\x12 \x01(\x05R\bcpuCount\x121\n" + "\x15gpu_foreign_memory_mb\x18\x13 \x03(\x05R\x12gpuForeignMemoryMb\x1aA\n" + "\x13EngineVersionsEntry\x12\x10\n" + "\x03key\x18\x01 \x01(\tR\x03key\x12\x14\n" + "\x05value\x18\x02 \x01(\tR\x05value:\x028\x01\";\n" + "\tProvision\x12.\n" + "\aengines\x18\x01 \x03(\v2\x14.llmgw.v1.EngineWantR\aengines\"\x8d\x02\n" + "\n" + "EngineWant\x12\x16\n" + "\x06engine\x18\x01 \x01(\tR\x06engine\x12\x18\n" + "\aversion\x18\x02 \x01(\tR\aversion\x12\x18\n" + "\avariant\x18\x03 \x01(\tR\avariant\x12!\n" + "\foffline_base\x18\x04 \x01(\tR\vofflineBase\x12N\n" + "\x0eoffline_sha256\x18\x05 \x03(\v2'.llmgw.v1.EngineWant.OfflineSha256EntryR\rofflineSha256\x1a@\n" + "\x12OfflineSha256Entry\x12\x10\n" + "\x03key\x18\x01 \x01(\tR\x03key\x12\x14\n" + "\x05value\x18\x02 \x01(\tR\x05value:\x028\x01\"\x8b\x01\n" + "\x0fProvisionStatus\x12\x16\n" + "\x06engine\x18\x01 \x01(\tR\x06engine\x12\x18\n" + "\aversion\x18\x02 \x01(\tR\aversion\x12\x18\n" + "\avariant\x18\x03 \x01(\tR\avariant\x12\x14\n" + "\x05state\x18\x04 \x01(\tR\x05state\x12\x16\n" + "\x06detail\x18\x05 \x01(\tR\x06detail\"\x8b\x01\n" + "\aCompute\x12\x18\n" + "\abackend\x18\x01 \x01(\tR\abackend\x12\x18\n" + "\adevices\x18\x02 \x03(\tR\adevices\x12)\n" + "\x10layers_offloaded\x18\x03 \x01(\x05R\x0flayersOffloaded\x12!\n" + "\flayers_total\x18\x04 \x01(\x05R\vlayersTotal\"\xe8\x03\n" + "\vShardStatus\x12\x17\n" + "\anode_id\x18\x01 \x01(\tR\x06nodeId\x12\x1d\n" + "\n" + "replica_id\x18\x02 \x01(\tR\treplicaId\x12\x19\n" + "\bshard_id\x18\x03 \x01(\tR\ashardId\x121\n" + "\x05state\x18\x04 \x01(\x0e2\x1b.llmgw.v1.ShardStatus.StateR\x05state\x12\x14\n" + "\x05error\x18\x05 \x01(\tR\x05error\x12\x1b\n" + "\texit_code\x18\x06 \x01(\x05R\bexitCode\x12\x1f\n" + "\vlast_output\x18\a \x03(\tR\n" + "lastOutput\x12-\n" + "\x05fetch\x18\b \x01(\v2\x17.llmgw.v1.FetchProgressR\x05fetch\x12+\n" + "\acompute\x18\t \x01(\v2\x11.llmgw.v1.ComputeR\acompute\"\xa2\x01\n" + "\x05State\x12\x15\n" + "\x11STATE_UNSPECIFIED\x10\x00\x12\x11\n" + "\rSTATE_PENDING\x10\x01\x12\x12\n" + "\x0eSTATE_STARTING\x10\x02\x12\x11\n" + "\rSTATE_LOADING\x10\x03\x12\x0f\n" + "\vSTATE_READY\x10\x04\x12\x12\n" + "\x0eSTATE_DRAINING\x10\x05\x12\x11\n" + "\rSTATE_STOPPED\x10\x06\x12\x10\n" + "\fSTATE_FAILED\x10\a\"g\n" + "\rFetchProgress\x12\x16\n" + "\x06source\x18\x01 \x01(\tR\x06source\x12\x1d\n" + "\n" + "bytes_done\x18\x02 \x01(\x03R\tbytesDone\x12\x1f\n" + "\vbytes_total\x18\x03 \x01(\x03R\n" + "bytesTotal\"\x93\x04\n" + "\n" + "StartShard\x12\x1d\n" + "\n" + "replica_id\x18\x01 \x01(\tR\treplicaId\x12\x19\n" + "\bshard_id\x18\x02 \x01(\tR\ashardId\x12\x12\n" + "\x04rank\x18\x03 \x01(\x05R\x04rank\x12\x16\n" + "\x06nnodes\x18\x04 \x01(\x05R\x06nnodes\x12\x1f\n" + "\vleader_addr\x18\x05 \x01(\tR\n" + "leaderAddr\x12\x1f\n" + "\vleader_port\x18\x06 \x01(\x05R\n" + "leaderPort\x12\x1f\n" + "\vgpu_indices\x18\a \x03(\x05R\n" + "gpuIndices\x12\x12\n" + "\x04hcas\x18\b \x03(\tR\x04hcas\x12%\n" + "\x0emgmt_interface\x18\t \x01(\tR\rmgmtInterface\x12\x1b\n" + "\thttp_port\x18\n" + " \x01(\x05R\bhttpPort\x121\n" + "\aprofile\x18\v \x01(\v2\x17.llmgw.v1.EngineProfileR\aprofile\x12,\n" + "\x12startup_timeout_ms\x18\f \x01(\x03R\x10startupTimeoutMs\x12\x12\n" + "\x04pool\x18\r \x01(\tR\x04pool\x12\x16\n" + "\x06fabric\x18\x0e \x01(\tR\x06fabric\x12\x1d\n" + "\n" + "profile_fp\x18\x0f \x01(\tR\tprofileFp\x12\x17\n" + "\aapi_key\x18\x10 \x01(\tR\x06apiKey\x12\x1f\n" + "\vcpu_threads\x18\x11 \x01(\x05R\n" + "cpuThreads\"\xcd\x03\n" + "\rEngineProfile\x12\x12\n" + "\x04name\x18\x01 \x01(\tR\x04name\x12\x16\n" + "\x06engine\x18\x02 \x01(\tR\x06engine\x12\x1b\n" + "\tgpu_class\x18\x03 \x01(\tR\bgpuClass\x12\x0e\n" + "\x02tp\x18\x04 \x01(\x05R\x02tp\x12\x0e\n" + "\x02pp\x18\x05 \x01(\x05R\x02pp\x12\"\n" + "\rgpus_per_node\x18\x06 \x01(\x05R\vgpusPerNode\x12\x14\n" + "\x05nodes\x18\a \x01(\x05R\x05nodes\x12\x12\n" + "\x04args\x18\b \x03(\tR\x04args\x122\n" + "\x03env\x18\t \x03(\v2 .llmgw.v1.EngineProfile.EnvEntryR\x03env\x12\x1d\n" + "\n" + "model_path\x18\n" + " \x01(\tR\tmodelPath\x12*\n" + "\x11served_model_name\x18\v \x01(\tR\x0fservedModelName\x121\n" + "\badapters\x18\f \x03(\v2\x15.llmgw.v1.LoraAdapterR\badapters\x12\x1b\n" + "\tkv_events\x18\r \x01(\bR\bkvEvents\x1a6\n" + "\bEnvEntry\x12\x10\n" + "\x03key\x18\x01 \x01(\tR\x03key\x12\x14\n" + "\x05value\x18\x02 \x01(\tR\x05value:\x028\x01\"%\n" + "\rRemoveWeights\x12\x14\n" + "\x05paths\x18\x01 \x03(\tR\x05paths\"&\n" + "\tStopShard\x12\x19\n" + "\bshard_id\x18\x01 \x01(\tR\ashardId\"'\n" + "\n" + "DrainShard\x12\x19\n" + "\bshard_id\x18\x01 \x01(\tR\ashardId\"J\n" + "\x03Ack\x12\x1d\n" + "\n" + "message_id\x18\x01 \x01(\tR\tmessageId\x12\x0e\n" + "\x02ok\x18\x02 \x01(\bR\x02ok\x12\x14\n" + "\x05error\x18\x03 \x01(\tR\x05error\"Z\n" + "\vLogsRequest\x12\x1d\n" + "\n" + "request_id\x18\x01 \x01(\tR\trequestId\x12\x16\n" + "\x06source\x18\x02 \x01(\tR\x06source\x12\x14\n" + "\x05lines\x18\x03 \x01(\x05R\x05lines\"p\n" + "\tLogsReply\x12\x1d\n" + "\n" + "request_id\x18\x01 \x01(\tR\trequestId\x12\x14\n" + "\x05lines\x18\x02 \x03(\tR\x05lines\x12\x18\n" + "\asources\x18\x03 \x03(\tR\asources\x12\x14\n" + "\x05error\x18\x04 \x01(\tR\x05error\"X\n" + "\x11FabricScanRequest\x12\x1d\n" + "\n" + "request_id\x18\x01 \x01(\tR\trequestId\x12\x10\n" + "\x03hca\x18\x02 \x01(\tR\x03hca\x12\x12\n" + "\x04port\x18\x03 \x01(\x05R\x04port\"w\n" + "\x0fFabricScanReply\x12\x1d\n" + "\n" + "request_id\x18\x01 \x01(\tR\trequestId\x12\x16\n" + "\x06output\x18\x02 \x01(\tR\x06output\x12\x14\n" + "\x05error\x18\x03 \x01(\tR\x05error\x12\x17\n" + "\atook_ms\x18\x04 \x01(\x03R\x06tookMs\"\xb5\x01\n" + "\x11FabricTestRequest\x12\x1d\n" + "\n" + "request_id\x18\x01 \x01(\tR\trequestId\x12\x12\n" + "\x04role\x18\x02 \x01(\tR\x04role\x12\x10\n" + "\x03hca\x18\x03 \x01(\tR\x03hca\x12\x12\n" + "\x04port\x18\x04 \x01(\x05R\x04port\x12\x12\n" + "\x04peer\x18\x05 \x01(\tR\x04peer\x12\x19\n" + "\btcp_port\x18\x06 \x01(\x05R\atcpPort\x12\x18\n" + "\aseconds\x18\a \x01(\x05R\aseconds\"\x8b\x01\n" + "\x0fFabricTestReply\x12\x1d\n" + "\n" + "request_id\x18\x01 \x01(\tR\trequestId\x12\x12\n" + "\x04gbps\x18\x02 \x01(\x01R\x04gbps\x12\x16\n" + "\x06output\x18\x03 \x01(\tR\x06output\x12\x14\n" + "\x05error\x18\x04 \x01(\tR\x05error\x12\x17\n" + "\atook_ms\x18\x05 \x01(\x03R\x06tookMs\"c\n" + "\fUpgradeAgent\x12\x1d\n" + "\n" + "request_id\x18\x01 \x01(\tR\trequestId\x12\x18\n" + "\aversion\x18\x02 \x01(\tR\aversion\x12\x1a\n" + "\breleases\x18\x03 \x01(\tR\breleases\"]\n" + "\fUpgradeReply\x12\x1d\n" + "\n" + "request_id\x18\x01 \x01(\tR\trequestId\x12\x18\n" + "\aversion\x18\x02 \x01(\tR\aversion\x12\x14\n" + "\x05error\x18\x03 \x01(\tR\x05error\"9\n" + "\vLoraAdapter\x12\x12\n" + "\x04name\x18\x01 \x01(\tR\x04name\x12\x16\n" + "\x06source\x18\x02 \x01(\tR\x06source\"\xcb\x01\n" + "\n" + "TrainStart\x12\x15\n" + "\x06job_id\x18\x01 \x01(\tR\x05jobId\x12\x1b\n" + "\tbase_repo\x18\x02 \x01(\tR\bbaseRepo\x12#\n" + "\rbase_revision\x18\x03 \x01(\tR\fbaseRevision\x12\x12\n" + "\x04data\x18\x04 \x01(\fR\x04data\x12\x16\n" + "\x06params\x18\x05 \x01(\tR\x06params\x12$\n" + "\x0egpu_memory_mib\x18\x06 \x01(\x05R\fgpuMemoryMib\x12\x12\n" + "\x04gguf\x18\a \x01(\bR\x04gguf\"$\n" + "\vTrainCancel\x12\x15\n" + "\x06job_id\x18\x01 \x01(\tR\x05jobId\"\xad\x01\n" + "\vTrainStatus\x12\x15\n" + "\x06job_id\x18\x01 \x01(\tR\x05jobId\x12\x14\n" + "\x05state\x18\x02 \x01(\tR\x05state\x12\x12\n" + "\x04step\x18\x03 \x01(\x05R\x04step\x12\x14\n" + "\x05steps\x18\x04 \x01(\x05R\x05steps\x12\x12\n" + "\x04loss\x18\x05 \x01(\x01R\x04loss\x12\x18\n" + "\amessage\x18\x06 \x01(\tR\amessage\x12\x19\n" + "\bpeak_mib\x18\a \x01(\x05R\apeakMib\"t\n" + "\rTrainArtifact\x12\x15\n" + "\x06job_id\x18\x01 \x01(\tR\x05jobId\x12\x12\n" + "\x04kind\x18\x02 \x01(\tR\x04kind\x12\x10\n" + "\x03seq\x18\x03 \x01(\x05R\x03seq\x12\x12\n" + "\x04data\x18\x04 \x01(\fR\x04data\x12\x12\n" + "\x04last\x18\x05 \x01(\bR\x04last2O\n" + "\fAgentService\x12?\n" + "\aConnect\x12\x16.llmgw.v1.AgentMessage\x1a\x18.llmgw.v1.ControlMessage(\x010\x01B\x93\x01\n" + "\fcom.llmgw.v1B\n" + "AgentProtoP\x01Z6gitea.mynodes.xyz/ilker/llmgw/api/gen/llmgw/v1;llmgwv1\xa2\x02\x03LXX\xaa\x02\bLlmgw.V1\xca\x02\bLlmgw\\V1\xe2\x02\x14Llmgw\\V1\\GPBMetadata\xea\x02\tLlmgw::V1b\x06proto3" var ( file_llmgw_v1_agent_proto_rawDescOnce sync.Once file_llmgw_v1_agent_proto_rawDescData []byte ) func file_llmgw_v1_agent_proto_rawDescGZIP() []byte { file_llmgw_v1_agent_proto_rawDescOnce.Do(func() { file_llmgw_v1_agent_proto_rawDescData = protoimpl.X.CompressGZIP(unsafe.Slice(unsafe.StringData(file_llmgw_v1_agent_proto_rawDesc), len(file_llmgw_v1_agent_proto_rawDesc))) }) return file_llmgw_v1_agent_proto_rawDescData } var file_llmgw_v1_agent_proto_enumTypes = make([]protoimpl.EnumInfo, 1) var file_llmgw_v1_agent_proto_msgTypes = make([]protoimpl.MessageInfo, 56) var file_llmgw_v1_agent_proto_goTypes = []any{ (ShardStatus_State)(0), // 0: llmgw.v1.ShardStatus.State (*AgentMessage)(nil), // 1: llmgw.v1.AgentMessage (*KVEvents)(nil), // 2: llmgw.v1.KVEvents (*ControlMessage)(nil), // 3: llmgw.v1.ControlMessage (*RestartAgent)(nil), // 4: llmgw.v1.RestartAgent (*Firewall)(nil), // 5: llmgw.v1.Firewall (*AgentSettings)(nil), // 6: llmgw.v1.AgentSettings (*ControlPlanes)(nil), // 7: llmgw.v1.ControlPlanes (*Register)(nil), // 8: llmgw.v1.Register (*Host)(nil), // 9: llmgw.v1.Host (*Storage)(nil), // 10: llmgw.v1.Storage (*RunningShard)(nil), // 11: llmgw.v1.RunningShard (*GPU)(nil), // 12: llmgw.v1.GPU (*MIGDevice)(nil), // 13: llmgw.v1.MIGDevice (*FabricAttachment)(nil), // 14: llmgw.v1.FabricAttachment (*LocalWeights)(nil), // 15: llmgw.v1.LocalWeights (*NodeLoad)(nil), // 16: llmgw.v1.NodeLoad (*GPUSensor)(nil), // 17: llmgw.v1.GPUSensor (*ShardLoad)(nil), // 18: llmgw.v1.ShardLoad (*IBPort)(nil), // 19: llmgw.v1.IBPort (*SubnetManager)(nil), // 20: llmgw.v1.SubnetManager (*IMEX)(nil), // 21: llmgw.v1.IMEX (*GPUFabricState)(nil), // 22: llmgw.v1.GPUFabricState (*Heartbeat)(nil), // 23: llmgw.v1.Heartbeat (*Provision)(nil), // 24: llmgw.v1.Provision (*EngineWant)(nil), // 25: llmgw.v1.EngineWant (*ProvisionStatus)(nil), // 26: llmgw.v1.ProvisionStatus (*Compute)(nil), // 27: llmgw.v1.Compute (*ShardStatus)(nil), // 28: llmgw.v1.ShardStatus (*FetchProgress)(nil), // 29: llmgw.v1.FetchProgress (*StartShard)(nil), // 30: llmgw.v1.StartShard (*EngineProfile)(nil), // 31: llmgw.v1.EngineProfile (*RemoveWeights)(nil), // 32: llmgw.v1.RemoveWeights (*StopShard)(nil), // 33: llmgw.v1.StopShard (*DrainShard)(nil), // 34: llmgw.v1.DrainShard (*Ack)(nil), // 35: llmgw.v1.Ack (*LogsRequest)(nil), // 36: llmgw.v1.LogsRequest (*LogsReply)(nil), // 37: llmgw.v1.LogsReply (*FabricScanRequest)(nil), // 38: llmgw.v1.FabricScanRequest (*FabricScanReply)(nil), // 39: llmgw.v1.FabricScanReply (*FabricTestRequest)(nil), // 40: llmgw.v1.FabricTestRequest (*FabricTestReply)(nil), // 41: llmgw.v1.FabricTestReply (*UpgradeAgent)(nil), // 42: llmgw.v1.UpgradeAgent (*UpgradeReply)(nil), // 43: llmgw.v1.UpgradeReply (*LoraAdapter)(nil), // 44: llmgw.v1.LoraAdapter (*TrainStart)(nil), // 45: llmgw.v1.TrainStart (*TrainCancel)(nil), // 46: llmgw.v1.TrainCancel (*TrainStatus)(nil), // 47: llmgw.v1.TrainStatus (*TrainArtifact)(nil), // 48: llmgw.v1.TrainArtifact nil, // 49: llmgw.v1.ControlMessage.TraceEntry nil, // 50: llmgw.v1.Register.LabelsEntry nil, // 51: llmgw.v1.Register.EngineVersionsEntry nil, // 52: llmgw.v1.GPU.NicPathsEntry nil, // 53: llmgw.v1.IBPort.CountersEntry nil, // 54: llmgw.v1.Heartbeat.EngineVersionsEntry nil, // 55: llmgw.v1.EngineWant.OfflineSha256Entry nil, // 56: llmgw.v1.EngineProfile.EnvEntry } var file_llmgw_v1_agent_proto_depIdxs = []int32{ 8, // 0: llmgw.v1.AgentMessage.register:type_name -> llmgw.v1.Register 23, // 1: llmgw.v1.AgentMessage.heartbeat:type_name -> llmgw.v1.Heartbeat 28, // 2: llmgw.v1.AgentMessage.shard:type_name -> llmgw.v1.ShardStatus 37, // 3: llmgw.v1.AgentMessage.logs:type_name -> llmgw.v1.LogsReply 39, // 4: llmgw.v1.AgentMessage.fabric_scan:type_name -> llmgw.v1.FabricScanReply 41, // 5: llmgw.v1.AgentMessage.fabric_test:type_name -> llmgw.v1.FabricTestReply 43, // 6: llmgw.v1.AgentMessage.upgrade:type_name -> llmgw.v1.UpgradeReply 2, // 7: llmgw.v1.AgentMessage.kv_events:type_name -> llmgw.v1.KVEvents 47, // 8: llmgw.v1.AgentMessage.train_status:type_name -> llmgw.v1.TrainStatus 48, // 9: llmgw.v1.AgentMessage.train_artifact:type_name -> llmgw.v1.TrainArtifact 30, // 10: llmgw.v1.ControlMessage.start:type_name -> llmgw.v1.StartShard 33, // 11: llmgw.v1.ControlMessage.stop:type_name -> llmgw.v1.StopShard 34, // 12: llmgw.v1.ControlMessage.drain:type_name -> llmgw.v1.DrainShard 35, // 13: llmgw.v1.ControlMessage.ack:type_name -> llmgw.v1.Ack 32, // 14: llmgw.v1.ControlMessage.remove_weights:type_name -> llmgw.v1.RemoveWeights 24, // 15: llmgw.v1.ControlMessage.provision:type_name -> llmgw.v1.Provision 36, // 16: llmgw.v1.ControlMessage.logs:type_name -> llmgw.v1.LogsRequest 38, // 17: llmgw.v1.ControlMessage.fabric_scan:type_name -> llmgw.v1.FabricScanRequest 40, // 18: llmgw.v1.ControlMessage.fabric_test:type_name -> llmgw.v1.FabricTestRequest 42, // 19: llmgw.v1.ControlMessage.upgrade:type_name -> llmgw.v1.UpgradeAgent 7, // 20: llmgw.v1.ControlMessage.control_planes:type_name -> llmgw.v1.ControlPlanes 5, // 21: llmgw.v1.ControlMessage.firewall:type_name -> llmgw.v1.Firewall 4, // 22: llmgw.v1.ControlMessage.restart_agent:type_name -> llmgw.v1.RestartAgent 6, // 23: llmgw.v1.ControlMessage.agent_settings:type_name -> llmgw.v1.AgentSettings 45, // 24: llmgw.v1.ControlMessage.train_start:type_name -> llmgw.v1.TrainStart 46, // 25: llmgw.v1.ControlMessage.train_cancel:type_name -> llmgw.v1.TrainCancel 49, // 26: llmgw.v1.ControlMessage.trace:type_name -> llmgw.v1.ControlMessage.TraceEntry 50, // 27: llmgw.v1.Register.labels:type_name -> llmgw.v1.Register.LabelsEntry 12, // 28: llmgw.v1.Register.gpus:type_name -> llmgw.v1.GPU 14, // 29: llmgw.v1.Register.fabrics:type_name -> llmgw.v1.FabricAttachment 51, // 30: llmgw.v1.Register.engine_versions:type_name -> llmgw.v1.Register.EngineVersionsEntry 11, // 31: llmgw.v1.Register.running_shards:type_name -> llmgw.v1.RunningShard 15, // 32: llmgw.v1.Register.weights:type_name -> llmgw.v1.LocalWeights 10, // 33: llmgw.v1.Register.storage:type_name -> llmgw.v1.Storage 9, // 34: llmgw.v1.Register.host:type_name -> llmgw.v1.Host 21, // 35: llmgw.v1.Register.imex:type_name -> llmgw.v1.IMEX 0, // 36: llmgw.v1.RunningShard.state:type_name -> llmgw.v1.ShardStatus.State 27, // 37: llmgw.v1.RunningShard.compute:type_name -> llmgw.v1.Compute 52, // 38: llmgw.v1.GPU.nic_paths:type_name -> llmgw.v1.GPU.NicPathsEntry 13, // 39: llmgw.v1.GPU.mig:type_name -> llmgw.v1.MIGDevice 18, // 40: llmgw.v1.NodeLoad.shards:type_name -> llmgw.v1.ShardLoad 17, // 41: llmgw.v1.NodeLoad.gpus:type_name -> llmgw.v1.GPUSensor 53, // 42: llmgw.v1.IBPort.counters:type_name -> llmgw.v1.IBPort.CountersEntry 20, // 43: llmgw.v1.IBPort.subnet_managers:type_name -> llmgw.v1.SubnetManager 10, // 44: llmgw.v1.Heartbeat.storage:type_name -> llmgw.v1.Storage 15, // 45: llmgw.v1.Heartbeat.weights:type_name -> llmgw.v1.LocalWeights 54, // 46: llmgw.v1.Heartbeat.engine_versions:type_name -> llmgw.v1.Heartbeat.EngineVersionsEntry 26, // 47: llmgw.v1.Heartbeat.provision:type_name -> llmgw.v1.ProvisionStatus 16, // 48: llmgw.v1.Heartbeat.load:type_name -> llmgw.v1.NodeLoad 19, // 49: llmgw.v1.Heartbeat.ib_ports:type_name -> llmgw.v1.IBPort 14, // 50: llmgw.v1.Heartbeat.fabrics:type_name -> llmgw.v1.FabricAttachment 22, // 51: llmgw.v1.Heartbeat.gpu_fabric:type_name -> llmgw.v1.GPUFabricState 21, // 52: llmgw.v1.Heartbeat.imex:type_name -> llmgw.v1.IMEX 25, // 53: llmgw.v1.Provision.engines:type_name -> llmgw.v1.EngineWant 55, // 54: llmgw.v1.EngineWant.offline_sha256:type_name -> llmgw.v1.EngineWant.OfflineSha256Entry 0, // 55: llmgw.v1.ShardStatus.state:type_name -> llmgw.v1.ShardStatus.State 29, // 56: llmgw.v1.ShardStatus.fetch:type_name -> llmgw.v1.FetchProgress 27, // 57: llmgw.v1.ShardStatus.compute:type_name -> llmgw.v1.Compute 31, // 58: llmgw.v1.StartShard.profile:type_name -> llmgw.v1.EngineProfile 56, // 59: llmgw.v1.EngineProfile.env:type_name -> llmgw.v1.EngineProfile.EnvEntry 44, // 60: llmgw.v1.EngineProfile.adapters:type_name -> llmgw.v1.LoraAdapter 1, // 61: llmgw.v1.AgentService.Connect:input_type -> llmgw.v1.AgentMessage 3, // 62: llmgw.v1.AgentService.Connect:output_type -> llmgw.v1.ControlMessage 62, // [62:63] is the sub-list for method output_type 61, // [61:62] is the sub-list for method input_type 61, // [61:61] is the sub-list for extension type_name 61, // [61:61] is the sub-list for extension extendee 0, // [0:61] is the sub-list for field type_name } func init() { file_llmgw_v1_agent_proto_init() } func file_llmgw_v1_agent_proto_init() { if File_llmgw_v1_agent_proto != nil { return } file_llmgw_v1_agent_proto_msgTypes[0].OneofWrappers = []any{ (*AgentMessage_Register)(nil), (*AgentMessage_Heartbeat)(nil), (*AgentMessage_Shard)(nil), (*AgentMessage_Logs)(nil), (*AgentMessage_FabricScan)(nil), (*AgentMessage_FabricTest)(nil), (*AgentMessage_Upgrade)(nil), (*AgentMessage_KvEvents)(nil), (*AgentMessage_TrainStatus)(nil), (*AgentMessage_TrainArtifact)(nil), } file_llmgw_v1_agent_proto_msgTypes[2].OneofWrappers = []any{ (*ControlMessage_Start)(nil), (*ControlMessage_Stop)(nil), (*ControlMessage_Drain)(nil), (*ControlMessage_Ack)(nil), (*ControlMessage_RemoveWeights)(nil), (*ControlMessage_Provision)(nil), (*ControlMessage_Logs)(nil), (*ControlMessage_FabricScan)(nil), (*ControlMessage_FabricTest)(nil), (*ControlMessage_Upgrade)(nil), (*ControlMessage_ControlPlanes)(nil), (*ControlMessage_Firewall)(nil), (*ControlMessage_RestartAgent)(nil), (*ControlMessage_AgentSettings)(nil), (*ControlMessage_TrainStart)(nil), (*ControlMessage_TrainCancel)(nil), } type x struct{} out := protoimpl.TypeBuilder{ File: protoimpl.DescBuilder{ GoPackagePath: reflect.TypeOf(x{}).PkgPath(), RawDescriptor: unsafe.Slice(unsafe.StringData(file_llmgw_v1_agent_proto_rawDesc), len(file_llmgw_v1_agent_proto_rawDesc)), NumEnums: 1, NumMessages: 56, NumExtensions: 0, NumServices: 1, }, GoTypes: file_llmgw_v1_agent_proto_goTypes, DependencyIndexes: file_llmgw_v1_agent_proto_depIdxs, EnumInfos: file_llmgw_v1_agent_proto_enumTypes, MessageInfos: file_llmgw_v1_agent_proto_msgTypes, }.Build() File_llmgw_v1_agent_proto = out.File file_llmgw_v1_agent_proto_goTypes = nil file_llmgw_v1_agent_proto_depIdxs = nil }