Edge Node Clusters: nodes, volumes, replicas, upgrades

Node Info: role, age, ip
Pod Info: status, restarts, ip
Volume and Replica Info: rebuild progress, health
Upgrade Status: cluster wide and node components

Signed-off-by: Andrew Durbin <[email protected]>

proto/info/edge_node_cluster.proto

@@ -0,0 +1,267 @@
+// Copyright(c) 2024 Zededa, Inc.
+// All rights reserved.
+
+syntax = "proto3";
+
+package org.lfedge.eve.info;
+
+import "info/error.proto";
+
+option go_package = "github.com/lf-edge/eve-api/go/info";
+option java_package = "org.lfedge.eve.info";
+
+import "google/protobuf/timestamp.proto";
+
+// KubeNodeStatus follows Kubernetes Node Conditions 
+//   Refer here: https://pkg.go.dev/k8s.io/api/core/v1#NodeConditionType
+enum KubeNodeConditionType {
+  KUBE_NODE_CONDITION_TYPE_READY = 0;
+  KUBE_NODE_CONDITION_TYPE_MEMORY_PRESSURE = 1;
+  KUBE_NODE_CONDITION_TYPE_DISK_PRESSURE = 2;
+  KUBE_NODE_CONDITION_TYPE_PID_PRESSURE = 3;
+  KUBE_NODE_CONDITION_TYPE_NETWORK_UNAVAIL = 4;
+}
+
+message KubeNodeCondition {
+  // Type of the condition
+  KubeNodeConditionType type = 1;
+
+  // Condition state
+  bool set = 2;
+}
+
+message KubeNodeInfo {
+  // Name of the node, will match device name
+  string name = 1;
+
+  // Status of the node
+  repeated KubeNodeCondition conditions = 2;
+
+  // Role of the node is server or not
+  bool role_server = 3;
+
+  // Creation Time of the node in the cluster
+  google.protobuf.Timestamp creation_timestamp = 4;
+
+  // Version of the API Server running on the node
+  string api_server_sersion = 5;
+
+  // Internal IP address of the node
+  string internal_ip = 6;
+
+  bool schedulable = 7;
+}
+
+// StorageHealthStatus is a higher level tracking status to show redundancy/failure-zone level
+// and rebuild progress.
+enum StorageHealthStatus {
+  STORAGE_HEALTH_STATUS_UNKNOWN = 0;
+  STORAGE_HEALTH_STATUS_HEALTHY = 1;
+  STORAGE_HEALTH_STATUS_DEGRADED_2_REPLICA_AVAILABLE_REPLICATING = 2;
+  STORAGE_HEALTH_STATUS_DEGRADED_2_REPLICA_AVAILABLE_NOT_REPLICATING = 3;
+  STORAGE_HEALTH_STATUS_DEGRADED_1_REPLICA_AVAILABLE_REPLICATING = 4;
+  STORAGE_HEALTH_STATUS_DEGRADED_1_REPLICA_AVAILABLE_NOT_REPLICATING = 5;
+  STORAGE_HEALTH_STATUS_FAILED = 6;
+}
+
+// StorageVolumeState is the kubernetes 'state' field of a replicated csi-driver volume.
+//  Refer to: https://github.com/longhorn/longhorn-manager/blob/v1.6.2/k8s/pkg/apis/longhorn/v1beta1/volume.go#L14
+enum StorageVolumeState {
+  STORAGE_VOLUME_STATE_UNKNOWN = 0;
+  STORAGE_VOLUME_STATE_CREATING = 1;
+  STORAGE_VOLUME_STATE_ATTACHED = 2;
+  STORAGE_VOLUME_STATE_DETACHED = 3;
+  STORAGE_VOLUME_STATE_ATTACHING = 4;
+  STORAGE_VOLUME_STATE_DETACHING = 5;
+  STORAGE_VOLUME_STATE_DELETING = 6;
+}
+
+// StorageVolumeRobustness is the 'robustness' of a replicated csi-driver volume.
+//  Refer to: https://github.com/longhorn/longhorn-manager/blob/v1.6.2/k8s/pkg/apis/longhorn/v1beta1/volume.go#L25
+enum StorageVolumeRobustness {
+  STORAGE_VOLUME_ROBUSTNESS_UNKNOWN = 0;
+  STORAGE_VOLUME_ROBUSTNESS_HEALTHY = 1;
+  STORAGE_VOLUME_ROBUSTNESS_DEGRADED = 2;
+  STORAGE_VOLUME_ROBUSTNESS_FAULTED = 3;
+}
+
+// StorageVolumePVCStatus is the kubernetes 'phase' of a PVC. Listed as status in cli.
+//  Refer to: https://kubernetes.io/docs/concepts/storage/persistent-volumes#phase
+enum StorageVolumePVCStatus {
+  STORAGE_VOLUME_PVC_STATUS_UNKNOWN = 0;
+  STORAGE_VOLUME_PVC_STATUS_BOUND = 1;
+  STORAGE_VOLUME_PVC_STATUS_PENDING = 2; // Accepted but not yet scheduled
+  STORAGE_VOLUME_PVC_STATUS_AVAILABLE = 3;
+  STORAGE_VOLUME_PVC_STATUS_RELEASED = 4;
+  STORAGE_VOLUME_PVC_STATUS_FAILED = 5;
+}
+
+// StorageVolumeReplicaStatus is a higher level status which combines replica and engine 
+//    status to show a simplified view of a replica rebuild state.
+enum StorageVolumeReplicaStatus {
+  STORAGE_VOLUME_REPLICA_STATUS_UNKNOWN = 0;
+  STORAGE_VOLUME_REPLICA_STATUS_REBUILDING = 1;
+  STORAGE_VOLUME_REPLICA_STATUS_ONLINE = 2;
+  STORAGE_VOLUME_REPLICA_STATUS_FAILED = 3; // Replacement/Rebuilt replica not yet scheduled.
+}
+
+message KubePodNameSpaceInfo {
+  // Name of the namespace
+  string name = 1;
+
+  // Number of pods in the namespace
+  uint32 pod_count = 2;
+
+  // Number of pods in the namespace that are running
+  uint32 pod_running_count = 3;
+
+  // Number of pods in the namespace that are pending
+  uint32 pod_pending_count = 4;
+
+  // Number of pods in the namespace that are failed
+  uint32 pod_failed_count = 5;
+
+  // Number of pods in the namespace that are succeeded
+  uint32 pod_succeeded_count = 6;
+}
+
+enum KubePodStatus {
+  KUBE_POD_STATUS_UNKNOWN = 0;
+  KUBE_POD_STATUS_PENDING = 1;
+  KUBE_POD_STATUS_RUNNING = 2;
+  KUBE_POD_STATUS_SUCCEEDED = 3;
+  KUBE_POD_STATUS_CONTAINER_CREATING = 4;
+  KUBE_POD_STATUS_CRASHLOOP_BACKOFF = 5;
+  KUBE_POD_STATUS_ERROR = 6;
+  KUBE_POD_STATUS_EVICTED = 7;
+  KUBE_POD_STATUS_FAILED = 8;
+}
+
+message KubeEVEAppPodInfo {
+  // Name of the EVE application
+  string name = 1;
+
+  // Application Status
+  KubePodStatus status = 2;
+
+  // Restart count of the application
+  uint32 restart_count = 3;
+
+  // Restart time of the application, seconds ago
+  google.protobuf.Timestamp restart_time = 4;
+
+  // Creation Time of the application
+  google.protobuf.Timestamp creation_timestamp = 5;
+
+  // IP address of the application, on cni0 interface
+  string ip_address = 6;
+
+  // Node name on which the application is running
+  string node_name = 7;
+}
+
+message KubeVolumeReplicaInfo {
+  // Name of the volume replica
+  string name = 1;
+
+  // Node replica resides on, will match node name
+  string owner_node = 2;
+
+  // Rebuild progress of the volume replica
+  uint32 rebuild_progress_percentage = 3;
+
+  // Replica status
+  StorageVolumeReplicaStatus status = 4;
+}
+
+message KubeVolumeInfo {
+  // Name of the volume
+  string name = 1;
+
+  // Status of the volume
+  StorageVolumeState state = 2;
+
+  // Robustness of the volume
+  StorageVolumeRobustness robustness = 3;
+
+  // Creation Time of the volume in the cluster
+  google.protobuf.Timestamp creation_timestamp = 4;
+
+  // Provisioned size of the volume in bytes
+  uint64 provisioned_bytes = 5;
+
+  // Allocated size of the volume in bytes
+  uint64 allocated_bytes = 6;
+
+  // PV/PVC status of the volume
+  StorageVolumePVCStatus pvc_status = 7;
+
+  // Replicas of the volume
+  repeated KubeVolumeReplicaInfo replica = 8;
+}
+
+message KubeStorageInfo {
+  // Overall status of Longhorn
+  StorageHealthStatus health = 1;
+
+  // Time of the most recent health status transition
+  google.protobuf.Timestamp transition_time = 2;
+
+  // Status of all the volumes in Longhorn
+  repeated KubeVolumeInfo volumes = 3;
+}
+
+// KubeCompUpgradeStatus will track status of each
+//  KubeComp which will upgrade serially in a cluster.
+enum KubeCompUpgradeStatus {
+  KUBE_COMP_UPGRADE_STATUS_UNKNOWN = 0;
+  KUBE_COMP_UPGRADE_STATUS_DOWNLOAD = 1;
+  KUBE_COMP_UPGRADE_STATUS_DOWNLOAD_FAILED = 2;
+  KUBE_COMP_UPGRADE_STATUS_IN_PROGRESS = 3;
+  KUBE_COMP_UPGRADE_STATUS_FAILED = 4;
+  KUBE_COMP_UPGRADE_STATUS_COMPLETED = 5;
+}
+
+// KubeComp is a component installed in eve after usb install.
+//  These are provided to show more detail on cluster upgrade progress.
+enum KubeComp {
+  KUBE_COMP_UNKNOWN = 0;
+  KUBE_COMP_CONTAINERD = 1; //every node will publish
+  KUBE_COMP_K3S = 2; // every node will publish
+  KUBE_COMP_MULTUS = 3; // Only the first node to upgrade eve-os will publish the remaining here and below
+  KUBE_COMP_KUBEVIRT = 4;
+  KUBE_COMP_CDI = 5;
+  KUBE_COMP_LONGHORN = 6;
+}
+
+message KubeClusterUpgradeStatus {
+  // current_node will have an empty value when no node is in an upgrade
+  string current_node = 1;
+
+  // component currently under upgrade, COMP_UNKNOWN when no upgrades in progress
+  KubeComp component = 2;
+
+  // status of the current component under upgrade, KUBE_COMP_UPGRADE_STATUS_UNKNOWN when
+  //   no upgrades in progress
+  KubeCompUpgradeStatus status = 3;
+
+  // Error info in case of failure
+  org.lfedge.eve.info.ErrorInfo error = 4;
+}
+
+message KubeClusterInfo {
+  // Info message on a list of cluster nodes
+  repeated KubeNodeInfo nodes = 1;
+
+  // Info message on a list of namespaces's pods summary
+  repeated KubePodNameSpaceInfo pod_name_spaces = 2;
+
+  // Info message on a list of EVE applications
+  repeated KubeEVEAppPodInfo eve_apps = 3;
+
+  // Info message on cluster storage
+  KubeStorageInfo storage = 4;
+
+  // The current status of the cluster upgrade
+  KubeClusterUpgradeStatus upgrade = 5;
+}

proto/info/error.proto

@@ -0,0 +1,62 @@
+// Copyright(c) 2024 Zededa, Inc.
+// All rights reserved.
+
+syntax = "proto3";
+
+package org.lfedge.eve.info;
+option go_package = "github.com/lf-edge/eve-api/go/info";
+option java_package = "org.lfedge.eve.info";
+
+import "google/protobuf/timestamp.proto";
+
+// Entity contains the entity type
+enum Entity {
+  // Invalid Device Entity
+  ENTITY_UNSPECIFIED = 0;
+  // Base OS entity
+  ENTITY_BASE_OS = 1;
+  // System Adapter Entity
+  ENTITY_SYSTEM_ADAPTER = 2;
+  // Vault Entity
+  ENTITY_VAULT = 3;
+  // Attestation Entity
+  ENTITY_ATTESTATION = 4;
+  // App Instance Entity
+  ENTITY_APP_INSTANCE = 5;
+  // Port Entity
+  ENTITY_PORT = 6;
+  // Network Entity
+  ENTITY_NETWORK = 7;
+  // Network Instance Entity
+  ENTITY_NETWORK_INSTANCE = 8;
+  // ContentTree Entity
+  ENTITY_CONTENT_TREE = 9;
+  // Blob Entity
+  ENTITY_CONTENT_BLOB = 10;
+  // VOLUME Entity
+  ENTITY_VOLUME = 11;
+}
+
+// Severity tells the severity type
+enum Severity {
+  SEVERITY_UNSPECIFIED = 0; // severity unspecified
+  SEVERITY_NOTICE = 1;      // severity notice
+  SEVERITY_WARNING = 2;     // severity warning
+  SEVERITY_ERROR = 3;       // severity error
+}
+
+// DeviceEntity contains the device entity details
+message DeviceEntity {
+  Entity entity = 1;           // entity type
+  string entity_id = 2;        // entity uuid
+  string entity_name = 3;      // entity name
+}
+
+// Errors in response to the application of configuration
+message ErrorInfo {
+  string description = 1;
+  google.protobuf.Timestamp timestamp = 2;  // Timestamp at which error had occurred
+  Severity severity = 3;                    // Severity of the error
+  repeated DeviceEntity entities = 4;       // objects referenced by the description or retry_condition
+  string retry_condition = 5;               // condition to retry
+}