feat: Add health check for CephCluster CRD

Signed-off-by: Arnaud Farbos <[email protected]>

Author: afarbos

resource_customizations/ceph.rook.io/CephCluster/health.lua

@@ -0,0 +1,66 @@
+-- CRD documentation: https://rook.github.io/docs/rook/latest-release/CRDs/Cluster/ceph-cluster-crd/
+local hs = {
+    status = "Progressing",
+    message = ""
+}
+
+function append_to_message(message)
+    if message ~= "" then
+        if hs.message ~= "" then
+            hs.message = hs.message .. " - " .. message
+        else
+            hs.message = message
+        end
+    end
+end
+
+if obj.status == nil then
+    append_to_message("Waiting for status to be reported")
+    return hs
+end
+
+-- Check the main Ceph health status first - https://github.com/ceph/ceph/blob/v20.3.0/src/include/health.h#L12
+if obj.status.ceph ~= nil and obj.status.ceph.health ~= nil then
+    local ceph_health = obj.status.ceph.health
+    local details_message = ""
+
+    -- Build details message from status.ceph.details if available
+    if obj.status.ceph.details ~= nil then
+        local detail_parts = {}
+        for detail_type, detail_info in pairs(obj.status.ceph.details) do
+            if detail_info.message ~= nil then
+                table.insert(detail_parts, detail_info.message)
+            end
+        end
+        if #detail_parts > 0 then
+            details_message =  " (" .. table.concat(detail_parts, "; ") .. ")"
+        end
+    end
+
+    if ceph_health == "HEALTH_ERR" or ceph_health == "HEALTH_WARN" then
+        hs.status = "Degraded"
+    elseif ceph_health == "HEALTH_OK" then
+        hs.status = "Healthy"
+    end
+    append_to_message("Ceph health is " .. ceph_health .. details_message)
+end
+
+-- Check state - https://github.com/rook/rook/blob/v1.17.7/pkg/apis/ceph.rook.io/v1/types.go#L621
+if obj.status.state ~= nil then
+    if hs.status == "Healthy" or hs.status == "Connected" then
+        append_to_message("Ceph cluster state is " .. obj.status.state)
+        if obj.status.state == "Created" then
+            hs.status = "Healthy"
+        elseif obj.status.state == "Error" then
+            hs.status = "Degraded"
+        else
+            hs.status = "Progressing"
+        end
+    end
+end
+
+if obj.status.message ~= nil then
+    append_to_message(obj.status.message)
+end
+
+return hs

resource_customizations/ceph.rook.io/CephCluster/health_test.yaml

@@ -0,0 +1,21 @@
+tests:
+- healthStatus:
+    status: Healthy
+    message: 'Ceph health is HEALTH_OK - Ceph cluster state is Created - Cluster created successfully'
+  inputPath: testdata/healthy.yaml
+- healthStatus:
+    status: Degraded
+    message: 'Ceph health is HEALTH_WARN (4 osds down; 2 pools degraded) - Cluster has warnings'
+  inputPath: testdata/degraded_warn.yaml
+- healthStatus:
+    status: Degraded
+    message: 'Ceph health is HEALTH_ERR (8 osds down) - Cluster has critical errors'
+  inputPath: testdata/degraded_error.yaml
+- healthStatus:
+    status: Progressing
+    message: 'Cluster is being created'
+  inputPath: testdata/state_creating.yaml
+- healthStatus:
+    status: Progressing
+    message: 'Waiting for status to be reported'
+  inputPath: testdata/no_status.yaml

resource_customizations/ceph.rook.io/CephCluster/testdata/degraded_error.yaml

@@ -0,0 +1,17 @@
+apiVersion: ceph.rook.io/v1
+kind: CephCluster
+metadata:
+  name: test-ceph-cluster
+  namespace: rook-ceph
+spec:
+  cephVersion:
+    image: quay.io/ceph/ceph:v19.2.0
+status:
+  ceph:
+    health: HEALTH_ERR
+    details:
+      OSD_DOWN:
+        message: 8 osds down
+        severity: HEALTH_ERR
+  state: Error
+  message: Cluster has critical errors

resource_customizations/ceph.rook.io/CephCluster/testdata/degraded_warn.yaml

@@ -0,0 +1,20 @@
+apiVersion: ceph.rook.io/v1
+kind: CephCluster
+metadata:
+  name: test-ceph-cluster
+  namespace: rook-ceph
+spec:
+  cephVersion:
+    image: quay.io/ceph/ceph:v19.2.0
+status:
+  ceph:
+    health: HEALTH_WARN
+    details:
+      OSD_DOWN:
+        message: 4 osds down
+        severity: HEALTH_WARN
+      POOL_DEGRADED:
+        message: 2 pools degraded
+        severity: HEALTH_WARN
+  state: Created
+  message: Cluster has warnings

resource_customizations/ceph.rook.io/CephCluster/testdata/healthy.yaml

@@ -0,0 +1,44 @@
+apiVersion: ceph.rook.io/v1
+kind: CephCluster
+metadata:
+  name: test-ceph-cluster
+  namespace: rook-ceph
+spec:
+  cephVersion:
+    image: quay.io/ceph/ceph:v19.2.0
+  mon:
+    count: 3
+  mgr:
+    count: 2
+  storage:
+    useAllNodes: true
+    useAllDevices: false
+status:
+  ceph:
+    health: HEALTH_OK
+    lastChecked: "2025-08-11T16:03:08Z"
+    fsid: c121226d-cac9-492f-8b0b-c05693243380
+    capacity:
+      bytesAvailable: 35183103942656
+      bytesTotal: 35184372088832
+      bytesUsed: 1268146176
+      lastUpdated: "2025-08-11T16:03:08Z"
+  conditions:
+  - lastHeartbeatTime: "2025-08-11T16:03:08Z"
+    lastTransitionTime: "2025-08-11T16:03:08Z"
+    message: Cluster created successfully
+    reason: ClusterCreated
+    status: "True"
+    type: Ready
+  - lastHeartbeatTime: "2025-08-11T16:03:08Z"
+    lastTransitionTime: "2025-08-11T16:03:08Z"
+    message: All OSDs are running
+    reason: OSDsRunning
+    status: "True"
+    type: Progressing
+  message: Cluster created successfully
+  phase: Ready
+  state: Created
+  version:
+    image: quay.io/ceph/ceph:v19.2.0
+    version: 19.2.0-0

resource_customizations/ceph.rook.io/CephCluster/testdata/no_status.yaml

@@ -0,0 +1,8 @@
+apiVersion: ceph.rook.io/v1
+kind: CephCluster
+metadata:
+  name: test-ceph-cluster
+  namespace: rook-ceph
+spec:
+  cephVersion:
+    image: quay.io/ceph/ceph:v19.2.0

resource_customizations/ceph.rook.io/CephCluster/testdata/state_creating.yaml

@@ -0,0 +1,11 @@
+apiVersion: ceph.rook.io/v1
+kind: CephCluster
+metadata:
+  name: test-ceph-cluster
+  namespace: rook-ceph
+spec:
+  cephVersion:
+    image: quay.io/ceph/ceph:v19.2.0
+status:
+  state: Creating
+  message: Cluster is being created