@oneuptime/common 11.0.1 → 11.0.2

package/Types/Monitor/CephMetricCatalog.ts

@@ -0,0 +1,409 @@
+import { CephResourceScope } from "./MonitorStepCephMonitor";
+import MetricsAggregationType from "../Metrics/MetricsAggregationType";
+
+export type CephMetricCategory = "Cluster Health" | "OSD" | "Pool" | "PG";
+
+export interface CephMetricDefinition {
+  id: string;
+  friendlyName: string;
+  description: string;
+  metricName: string;
+  category: CephMetricCategory;
+  defaultAggregation: MetricsAggregationType;
+  defaultResourceScope: CephResourceScope;
+  unit?: string;
+}
+
+/*
+ * Metric names follow the ceph-mgr Prometheus module naming scheme.
+ * Per-daemon series carry a `ceph_daemon` datapoint label (e.g. "osd.3",
+ * "mon.a"); per-pool series carry a `pool_id` label only — the pool name
+ * exists solely on `ceph_pool_metadata`, so filter/group pool data series
+ * by `pool_id` and join the metadata series when a display name is needed.
+ */
+const cephMetricCatalog: Array<CephMetricDefinition> = [
+  // Cluster Health Metrics
+  {
+    id: "ceph-health-status",
+    friendlyName: "Cluster Health Status",
+    description:
+      "Overall Ceph cluster health: 0 = HEALTH_OK, 1 = HEALTH_WARN, 2 = HEALTH_ERR. The single most important Ceph metric to alert on.",
+    metricName: "ceph_health_status",
+    category: "Cluster Health",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Cluster,
+  },
+  {
+    id: "ceph-health-detail",
+    friendlyName: "Health Check Detail",
+    description:
+      "One series per active health check (e.g. OSD_DOWN, PG_DEGRADED) with name and severity labels. Exposed by the mgr prometheus module on Quincy and later — absent on older releases.",
+    metricName: "ceph_health_detail",
+    category: "Cluster Health",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Cluster,
+    unit: "count",
+  },
+  {
+    id: "ceph-healthcheck-slow-ops",
+    friendlyName: "Slow Operations",
+    description:
+      "Number of slow OSD/monitor operations reported by the SLOW_OPS health check. Non-zero values surface as client I/O latency or hangs.",
+    metricName: "ceph_healthcheck_slow_ops",
+    category: "Cluster Health",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Cluster,
+    unit: "count",
+  },
+  {
+    id: "ceph-daemon-health-metrics",
+    friendlyName: "Daemon Health Metrics",
+    description:
+      "Per-daemon health metrics from the mgr, keyed by a type label (e.g. SLOW_OPS, PENDING_CREATING_PGS) and a ceph_daemon label. Filter type = SLOW_OPS and group by ceph_daemon to pinpoint the exact OSD or monitor reporting slow operations.",
+    metricName: "ceph_daemon_health_metrics",
+    category: "Cluster Health",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Osd,
+    unit: "count",
+  },
+  {
+    id: "ceph-mon-quorum-status",
+    friendlyName: "Monitor Quorum Status",
+    description:
+      "Whether each Ceph monitor daemon is in quorum (1) or not (0). One series per monitor — sum to count members in quorum, or take the minimum to detect any monitor out of quorum.",
+    metricName: "ceph_mon_quorum_status",
+    category: "Cluster Health",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Mon,
+    unit: "count",
+  },
+  {
+    id: "ceph-mon-metadata",
+    friendlyName: "Monitor Metadata",
+    description:
+      "Metadata series for each monitor daemon (value is always 1) carrying hostname and ceph_version labels. Sum it to count monitors in the cluster.",
+    metricName: "ceph_mon_metadata",
+    category: "Cluster Health",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Mon,
+    unit: "count",
+  },
+  {
+    id: "ceph-cluster-total-bytes",
+    friendlyName: "Cluster Total Capacity",
+    description:
+      "Total raw storage capacity of the cluster across all OSDs, before replication or erasure-coding overhead.",
+    metricName: "ceph_cluster_total_bytes",
+    category: "Cluster Health",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Cluster,
+    unit: "bytes",
+  },
+  {
+    id: "ceph-cluster-total-used-bytes",
+    friendlyName: "Cluster Used Capacity",
+    description:
+      "Raw storage currently used across all OSDs. Compare against Cluster Total Capacity to track how full the cluster is.",
+    metricName: "ceph_cluster_total_used_bytes",
+    category: "Cluster Health",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Cluster,
+    unit: "bytes",
+  },
+
+  // OSD Metrics
+  {
+    id: "ceph-osd-up",
+    friendlyName: "OSD Up",
+    description:
+      "Whether each OSD daemon is up (1) or down (0). One series per OSD — take the minimum to detect any OSD down, or sum to count OSDs that are up.",
+    metricName: "ceph_osd_up",
+    category: "OSD",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Osd,
+    unit: "count",
+  },
+  {
+    id: "ceph-osd-in",
+    friendlyName: "OSD In",
+    description:
+      "Whether each OSD is in the cluster (1) or marked out (0). An OSD that is up but out no longer stores data and triggers rebalancing.",
+    metricName: "ceph_osd_in",
+    category: "OSD",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Osd,
+    unit: "count",
+  },
+  {
+    id: "ceph-osd-apply-latency",
+    friendlyName: "OSD Apply Latency",
+    description:
+      "Time for each OSD to apply an operation to its backing store, in milliseconds. Sustained high values point to a failing or saturated disk.",
+    metricName: "ceph_osd_apply_latency_ms",
+    category: "OSD",
+    defaultAggregation: MetricsAggregationType.Avg,
+    defaultResourceScope: CephResourceScope.Osd,
+    unit: "ms",
+  },
+  {
+    id: "ceph-osd-commit-latency",
+    friendlyName: "OSD Commit Latency",
+    description:
+      "Time for each OSD to commit an operation to its journal/WAL, in milliseconds. Compare against apply latency to distinguish journal from backing-store slowness.",
+    metricName: "ceph_osd_commit_latency_ms",
+    category: "OSD",
+    defaultAggregation: MetricsAggregationType.Avg,
+    defaultResourceScope: CephResourceScope.Osd,
+    unit: "ms",
+  },
+  {
+    id: "ceph-osd-stat-bytes",
+    friendlyName: "OSD Total Capacity",
+    description: "Total raw capacity of each OSD's backing device.",
+    metricName: "ceph_osd_stat_bytes",
+    category: "OSD",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Osd,
+    unit: "bytes",
+  },
+  {
+    id: "ceph-osd-stat-bytes-used",
+    friendlyName: "OSD Used Capacity",
+    description:
+      "Raw bytes used on each OSD. Compare against OSD Total Capacity to spot imbalanced or nearfull OSDs before they block writes.",
+    metricName: "ceph_osd_stat_bytes_used",
+    category: "OSD",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Osd,
+    unit: "bytes",
+  },
+  {
+    id: "ceph-osd-numpg",
+    friendlyName: "OSD Placement Groups",
+    description:
+      "Number of placement groups hosted by each OSD. Outliers indicate CRUSH imbalance — overloaded OSDs become latency hotspots.",
+    metricName: "ceph_osd_numpg",
+    category: "OSD",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Osd,
+    unit: "count",
+  },
+  {
+    id: "ceph-osd-metadata",
+    friendlyName: "OSD Metadata",
+    description:
+      "Metadata series for each OSD daemon (value is always 1) carrying hostname, device_class, and ceph_version labels. Sum it to count OSDs in the cluster.",
+    metricName: "ceph_osd_metadata",
+    category: "OSD",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Osd,
+    unit: "count",
+  },
+
+  // Pool Metrics
+  {
+    id: "ceph-pool-stored",
+    friendlyName: "Pool Stored Data",
+    description:
+      "Logical data stored in each pool, before replication or erasure-coding overhead.",
+    metricName: "ceph_pool_stored",
+    category: "Pool",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "bytes",
+  },
+  {
+    id: "ceph-pool-max-avail",
+    friendlyName: "Pool Max Available",
+    description:
+      "Estimated space still writable to each pool given current cluster capacity and the pool's replication factor. Alert when this gets low to avoid a full pool.",
+    metricName: "ceph_pool_max_avail",
+    category: "Pool",
+    defaultAggregation: MetricsAggregationType.Min,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "bytes",
+  },
+  {
+    id: "ceph-pool-objects",
+    friendlyName: "Pool Object Count",
+    description: "Number of objects stored in each pool.",
+    metricName: "ceph_pool_objects",
+    category: "Pool",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "count",
+  },
+  {
+    id: "ceph-pool-rd",
+    friendlyName: "Pool Read Operations",
+    description:
+      "Cumulative read operations served by each pool. Chart as a rate to see read IOPS.",
+    metricName: "ceph_pool_rd",
+    category: "Pool",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "ops",
+  },
+  {
+    id: "ceph-pool-wr",
+    friendlyName: "Pool Write Operations",
+    description:
+      "Cumulative write operations served by each pool. Chart as a rate to see write IOPS.",
+    metricName: "ceph_pool_wr",
+    category: "Pool",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "ops",
+  },
+  {
+    id: "ceph-pool-rd-bytes",
+    friendlyName: "Pool Read Throughput",
+    description:
+      "Cumulative bytes read from each pool. Chart as a rate to see read throughput.",
+    metricName: "ceph_pool_rd_bytes",
+    category: "Pool",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "bytes",
+  },
+  {
+    id: "ceph-pool-wr-bytes",
+    friendlyName: "Pool Write Throughput",
+    description:
+      "Cumulative bytes written to each pool. Chart as a rate to see write throughput.",
+    metricName: "ceph_pool_wr_bytes",
+    category: "Pool",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "bytes",
+  },
+  {
+    id: "ceph-pool-metadata",
+    friendlyName: "Pool Metadata",
+    description:
+      "Metadata series for each pool (value is always 1) carrying the pool's name label — the ONLY series that maps pool_id to a name. Sum it to count pools in the cluster.",
+    metricName: "ceph_pool_metadata",
+    category: "Pool",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "count",
+  },
+
+  /*
+   * PG (Placement Group) Metrics
+   *
+   * All ceph_pg_* state metrics are exported PER POOL with a `pool_id`
+   * datapoint label on every supported release (Nautilus and later) —
+   * they are NOT single cluster-wide gauges. Sum across pools for a
+   * cluster-wide count; Max yields only the largest pool's value.
+   */
+  {
+    id: "ceph-pg-total",
+    friendlyName: "Total PGs",
+    description:
+      "Number of placement groups per pool (pool_id label). Sum across pools for the cluster-wide total; subtract Active PGs (also summed) to count PGs unable to serve I/O.",
+    metricName: "ceph_pg_total",
+    category: "PG",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "count",
+  },
+  {
+    id: "ceph-pg-active",
+    friendlyName: "Active PGs",
+    description:
+      "Number of placement groups in the active state (able to serve I/O) per pool (pool_id label). In a healthy cluster the sum across pools equals the summed total PG count.",
+    metricName: "ceph_pg_active",
+    category: "PG",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "count",
+  },
+  {
+    id: "ceph-pg-clean",
+    friendlyName: "Clean PGs",
+    description:
+      "Number of placement groups in the clean state (fully replicated, no recovery pending) per pool (pool_id label). In a healthy cluster the sum across pools equals the summed total PG count.",
+    metricName: "ceph_pg_clean",
+    category: "PG",
+    defaultAggregation: MetricsAggregationType.Sum,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "count",
+  },
+  {
+    id: "ceph-pg-degraded",
+    friendlyName: "Degraded PGs",
+    description:
+      "Number of placement groups with fewer data replicas than configured, per pool (pool_id label). Non-zero values mean redundancy is reduced — typically after an OSD failure.",
+    metricName: "ceph_pg_degraded",
+    category: "PG",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "count",
+  },
+  {
+    id: "ceph-pg-undersized",
+    friendlyName: "Undersized PGs",
+    description:
+      "Number of placement groups mapped to fewer OSDs than their configured replica count, per pool (pool_id label). Sustained non-zero values mean the cluster cannot restore full redundancy.",
+    metricName: "ceph_pg_undersized",
+    category: "PG",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Pool,
+    unit: "count",
+  },
+  {
+    id: "ceph-num-objects-degraded",
+    friendlyName: "Degraded Objects",
+    description:
+      "Number of objects with fewer replicas than configured. Should trend to zero as recovery completes after an OSD failure.",
+    metricName: "ceph_num_objects_degraded",
+    category: "PG",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Cluster,
+    unit: "count",
+  },
+  {
+    id: "ceph-num-objects-misplaced",
+    friendlyName: "Misplaced Objects",
+    description:
+      "Number of objects not stored on their CRUSH-intended OSDs (data is safe, placement is wrong). Non-zero during rebalancing — should trend to zero.",
+    metricName: "ceph_num_objects_misplaced",
+    category: "PG",
+    defaultAggregation: MetricsAggregationType.Max,
+    defaultResourceScope: CephResourceScope.Cluster,
+    unit: "count",
+  },
+];
+
+export function getAllCephMetrics(): Array<CephMetricDefinition> {
+  return cephMetricCatalog;
+}
+
+export function getCephMetricsByCategory(
+  category: CephMetricCategory,
+): Array<CephMetricDefinition> {
+  return cephMetricCatalog.filter((m: CephMetricDefinition) => {
+    return m.category === category;
+  });
+}
+
+export function getCephMetricById(
+  id: string,
+): CephMetricDefinition | undefined {
+  return cephMetricCatalog.find((m: CephMetricDefinition) => {
+    return m.id === id;
+  });
+}
+
+export function getCephMetricByMetricName(
+  metricName: string,
+): CephMetricDefinition | undefined {
+  return cephMetricCatalog.find((m: CephMetricDefinition) => {
+    return m.metricName === metricName;
+  });
+}
+
+export function getAllCephMetricCategories(): Array<CephMetricCategory> {
+  return ["Cluster Health", "OSD", "Pool", "PG"];
+}

package/Types/Monitor/MetricMonitor/MetricMonitorResponse.ts

@@ -24,6 +24,48 @@ export interface KubernetesResourceBreakdown {
   attributes: Dictionary<string>;
 }
 
+export interface ProxmoxAffectedResource {
+  /** Raw pve-exporter `id` datapoint label, e.g. "node/pve1", "qemu/100". */
+  resourceId?: string | undefined;
+  /** `name` label — present only on the pve_*_info metadata series. */
+  resourceName?: string | undefined;
+  /** Agent-stamped `pve.type` attribute (node | qemu | lxc | storage). */
+  resourceType?: string | undefined;
+  /** Agent-stamped `pve.scope` attribute (cluster | node | guest | storage). */
+  scope?: string | undefined;
+  /** `node` label — present only on the pve_*_info metadata series. */
+  nodeName?: string | undefined;
+  metricValue: number;
+}
+
+export interface ProxmoxResourceBreakdown {
+  clusterName: string;
+  metricName: string;
+  metricFriendlyName: string;
+  affectedResources: Array<ProxmoxAffectedResource>;
+  attributes: Dictionary<string>;
+}
+
+export interface CephAffectedResource {
+  /** `ceph_daemon` datapoint label, e.g. "osd.3", "mon.a". */
+  daemon?: string | undefined;
+  /** `pool_id` label — the only pool identity on pool data series. */
+  poolId?: string | undefined;
+  /** Pool `name` label — present only on ceph_pool_metadata. */
+  poolName?: string | undefined;
+  /** `hostname` label — present only on the *_metadata series. */
+  hostname?: string | undefined;
+  metricValue: number;
+}
+
+export interface CephResourceBreakdown {
+  clusterName: string;
+  metricName: string;
+  metricFriendlyName: string;
+  affectedResources: Array<CephAffectedResource>;
+  attributes: Dictionary<string>;
+}
+
 export default interface MetricMonitorResponse {
   projectId: ObjectID;
   startAndEndDate?: InBetween<Date>;
@@ -32,6 +74,8 @@ export default interface MetricMonitorResponse {
   monitorId: ObjectID;
   evaluationSummary?: MonitorEvaluationSummary | undefined;
   kubernetesResourceBreakdown?: KubernetesResourceBreakdown | undefined;
+  proxmoxResourceBreakdown?: ProxmoxResourceBreakdown | undefined;
+  cephResourceBreakdown?: CephResourceBreakdown | undefined;
   /**
    * Per-series breakdown when any queryConfig sets groupByAttributeKeys.
    * Each entry carries a fingerprint, the label values identifying that

package/Types/Monitor/MonitorStep.ts

@@ -50,6 +50,12 @@ import MonitorStepKubernetesMonitor, {
 import MonitorStepDockerMonitor, {
   MonitorStepDockerMonitorUtil,
 } from "./MonitorStepDockerMonitor";
+import MonitorStepProxmoxMonitor, {
+  MonitorStepProxmoxMonitorUtil,
+} from "./MonitorStepProxmoxMonitor";
+import MonitorStepCephMonitor, {
+  MonitorStepCephMonitorUtil,
+} from "./MonitorStepCephMonitor";
 import Zod, { ZodSchema } from "../../Utils/Schema/Zod";
 
 /*
@@ -169,6 +175,12 @@ export interface MonitorStepType {
 
   // Docker monitor
   dockerMonitor?: MonitorStepDockerMonitor | undefined;
+
+  // Proxmox monitor
+  proxmoxMonitor?: MonitorStepProxmoxMonitor | undefined;
+
+  // Ceph monitor
+  cephMonitor?: MonitorStepCephMonitor | undefined;
 }
 
 export default class MonitorStep extends DatabaseProperty {
@@ -208,6 +220,8 @@ export default class MonitorStep extends DatabaseProperty {
       externalStatusPageMonitor: undefined,
       kubernetesMonitor: undefined,
       dockerMonitor: undefined,
+      proxmoxMonitor: undefined,
+      cephMonitor: undefined,
     };
   }
 
@@ -252,6 +266,8 @@ export default class MonitorStep extends DatabaseProperty {
       externalStatusPageMonitor: undefined,
       kubernetesMonitor: undefined,
       dockerMonitor: undefined,
+      proxmoxMonitor: undefined,
+      cephMonitor: undefined,
     };
 
     return monitorStep;
@@ -441,6 +457,18 @@ export default class MonitorStep extends DatabaseProperty {
     return this;
   }
 
+  public setProxmoxMonitor(
+    proxmoxMonitor: MonitorStepProxmoxMonitor,
+  ): MonitorStep {
+    this.data!.proxmoxMonitor = proxmoxMonitor;
+    return this;
+  }
+
+  public setCephMonitor(cephMonitor: MonitorStepCephMonitor): MonitorStep {
+    this.data!.cephMonitor = cephMonitor;
+    return this;
+  }
+
   public setCustomCode(customCode: string): MonitorStep {
     this.data!.customCode = customCode;
     return this;
@@ -477,6 +505,8 @@ export default class MonitorStep extends DatabaseProperty {
         exceptionMonitor: undefined,
         kubernetesMonitor: undefined,
         dockerMonitor: undefined,
+        proxmoxMonitor: undefined,
+        cephMonitor: undefined,
       },
     };
   }
@@ -643,6 +673,26 @@ export default class MonitorStep extends DatabaseProperty {
       }
     }
 
+    if (monitorType === MonitorType.Proxmox) {
+      if (!value.data.proxmoxMonitor) {
+        return "Proxmox monitor configuration is required";
+      }
+
+      if (!value.data.proxmoxMonitor.clusterIdentifier) {
+        return "Proxmox cluster is required";
+      }
+    }
+
+    if (monitorType === MonitorType.Ceph) {
+      if (!value.data.cephMonitor) {
+        return "Ceph monitor configuration is required";
+      }
+
+      if (!value.data.cephMonitor.clusterIdentifier) {
+        return "Ceph cluster is required";
+      }
+    }
+
     return null;
   }
 
@@ -726,6 +776,12 @@ export default class MonitorStep extends DatabaseProperty {
           dockerMonitor: this.data.dockerMonitor
             ? MonitorStepDockerMonitorUtil.toJSON(this.data.dockerMonitor)
             : undefined,
+          proxmoxMonitor: this.data.proxmoxMonitor
+            ? MonitorStepProxmoxMonitorUtil.toJSON(this.data.proxmoxMonitor)
+            : undefined,
+          cephMonitor: this.data.cephMonitor
+            ? MonitorStepCephMonitorUtil.toJSON(this.data.cephMonitor)
+            : undefined,
         },
       });
     }
@@ -864,6 +920,12 @@ export default class MonitorStep extends DatabaseProperty {
       dockerMonitor: json["dockerMonitor"]
         ? (json["dockerMonitor"] as JSONObject)
         : undefined,
+      proxmoxMonitor: json["proxmoxMonitor"]
+        ? (json["proxmoxMonitor"] as JSONObject)
+        : undefined,
+      cephMonitor: json["cephMonitor"]
+        ? (json["cephMonitor"] as JSONObject)
+        : undefined,
     }) as any;
 
     return monitorStep;
@@ -902,6 +964,8 @@ export default class MonitorStep extends DatabaseProperty {
         externalStatusPageMonitor: Zod.any().optional(),
         kubernetesMonitor: Zod.any().optional(),
         dockerMonitor: Zod.any().optional(),
+        proxmoxMonitor: Zod.any().optional(),
+        cephMonitor: Zod.any().optional(),
       }).openapi({
         type: "object",
         example: {

package/Types/Monitor/MonitorStepCephMonitor.ts

@@ -0,0 +1,57 @@
+import { JSONObject } from "../JSON";
+import MetricsViewConfig from "../Metrics/MetricsViewConfig";
+import RollingTime from "../RollingTime/RollingTime";
+
+/**
+ * The kind of Ceph resource a metric series belongs to. Unlike Proxmox
+ * there is no agent-stamped scope attribute — ceph-mgr series are already
+ * equality-filterable by `ceph_daemon` / `pool_id` — so this enum is a
+ * catalog/UI hint only, never a query filter.
+ */
+export enum CephResourceScope {
+  Cluster = "Cluster",
+  Mon = "Mon",
+  Osd = "OSD",
+  Pool = "Pool",
+}
+
+export interface CephResourceFilters {
+  osdId?: string | undefined; // datapoint label `ceph_daemon` (e.g. "osd.3")
+  /**
+   * → equality on the `pool_id` datapoint label (e.g. "2"). Pool data
+   * series (ceph_pool_stored, ceph_pool_max_avail, ceph_pool_rd/wr…)
+   * carry ONLY `pool_id` — there is no pool-name label on data series;
+   * the name exists solely on ceph_pool_metadata. Filter by id and join
+   * ceph_pool_metadata when a display name is needed.
+   */
+  poolId?: string | undefined;
+}
+
+export default interface MonitorStepCephMonitor {
+  clusterIdentifier: string;
+  resourceFilters: CephResourceFilters;
+  metricViewConfig: MetricsViewConfig;
+  rollingTime: RollingTime;
+}
+
+export class MonitorStepCephMonitorUtil {
+  public static getDefault(): MonitorStepCephMonitor {
+    return {
+      clusterIdentifier: "",
+      resourceFilters: {},
+      metricViewConfig: {
+        queryConfigs: [],
+        formulaConfigs: [],
+      },
+      rollingTime: RollingTime.Past1Minute,
+    };
+  }
+
+  public static fromJSON(json: JSONObject): MonitorStepCephMonitor {
+    return json as any as MonitorStepCephMonitor;
+  }
+
+  public static toJSON(monitor: MonitorStepCephMonitor): JSONObject {
+    return monitor as any as JSONObject;
+  }
+}