Merge "Move ceph-mon's checkPGs cron job to ceph-client"

ceph-client/templates/bin/utils/_checkPGs.sh.tpl

@@ -18,6 +18,6 @@ limitations under the License.
 
 set -ex
 
-monPod=$(kubectl get pods --namespace=${DEPLOYMENT_NAMESPACE} --selector=application=ceph --selector=component=mon --output=jsonpath={.items[0].metadata.name} 2>/dev/null)
+mgrPod=$(kubectl get pods --namespace=${DEPLOYMENT_NAMESPACE} --selector=application=ceph --selector=component=mgr --output=jsonpath={.items[0].metadata.name} 2>/dev/null)
 
-kubectl exec -t ${monPod} --namespace=${DEPLOYMENT_NAMESPACE} -- /tmp/utils-checkPGs.py All 2>/dev/null
+kubectl exec -t ${mgrPod} --namespace=${DEPLOYMENT_NAMESPACE} -- /tmp/utils-checkPGs.py All 2>/dev/null

ceph-client/templates/configmap-bin.yaml

@@ -54,4 +54,10 @@ data:
 {{ tuple "bin/utils/_checkDNS.sh.tpl" . | include "helm-toolkit.utils.template" | indent 4 }}
   utils-checkDNS_start.sh: |
 {{ tuple "bin/utils/_checkDNS_start.sh.tpl" . | include "helm-toolkit.utils.template" | indent 4 }}
+
+  utils-checkPGs.py: |
+{{ tuple "bin/utils/_checkPGs.py.tpl" . | include "helm-toolkit.utils.template" | indent 4 }}
+
+  utils-checkPGs.sh: |
+{{ tuple "bin/utils/_checkPGs.sh.tpl" . | include "helm-toolkit.utils.template" | indent 4 }}
 {{- end }}

ceph-client/templates/cronjob-checkPGs.yaml

@@ -73,7 +73,9 @@ spec:
         spec:
           serviceAccountName: {{ $serviceAccountName }}
           nodeSelector:
-            {{ .Values.labels.mon.node_selector_key }}: {{ .Values.labels.mon.node_selector_value }}
+            {{ .Values.labels.mgr.node_selector_key }}: {{ .Values.labels.mgr.node_selector_value }}
+          initContainers:
+{{ tuple $envAll "pool_checkpgs" list | include "helm-toolkit.snippets.kubernetes_entrypoint_init_container" | indent 12 }}
           containers:
           - name: {{ $serviceAccountName }}
 {{ tuple $envAll "ceph_config_helper" | include "helm-toolkit.snippets.image" | indent 12 }}
@@ -85,15 +87,15 @@ spec:
             command:
               - /tmp/utils-checkPGs.sh
             volumeMounts:
-              - name: ceph-mon-bin
+              - name: ceph-client-bin
                 mountPath: /tmp/utils-checkPGs.py
                 subPath: utils-checkPGs.py
                 readOnly: true
-              - name: ceph-mon-bin
+              - name: ceph-client-bin
                 mountPath: /tmp/utils-checkPGs.sh
                 subPath: utils-checkPGs.sh
                 readOnly: true
-              - name: ceph-mon-etc
+              - name: ceph-client-etc
                 mountPath: /etc/ceph/ceph.conf
                 subPath: ceph.conf
                 readOnly: true
@@ -116,13 +118,13 @@ spec:
           restartPolicy: Never
           hostNetwork: true
           volumes:
-            - name: ceph-mon-bin
+            - name: ceph-client-bin
               configMap:
-                name: ceph-mon-bin
+                name: ceph-client-bin
                 defaultMode: 0555
-            - name: ceph-mon-etc
+            - name: ceph-client-etc
               configMap:
-                name: ceph-mon-etc
+                name: ceph-client-etc
                 defaultMode: 0444
             - name: ceph-client-admin-keyring
               secret:

ceph-client/templates/deployment-mgr.yaml

@@ -157,6 +157,14 @@ spec:
             - name: pod-run
               mountPath: /run
               readOnly: false
+            - name: ceph-client-bin
+              mountPath: /tmp/utils-checkPGs.py
+              subPath: utils-checkPGs.py
+              readOnly: true
+            - name: ceph-client-bin
+              mountPath: /tmp/utils-checkPGs.sh
+              subPath: utils-checkPGs.sh
+              readOnly: true
       volumes:
         - name: pod-etc-ceph
           emptyDir: {}

ceph-client/values.yaml

@@ -122,6 +122,20 @@ network:
   public: 192.168.0.0/16
   cluster: 192.168.0.0/16
 
+jobs:
+  pool_checkPGs:
+    # Execute every 15 minutes
+    cron: "*/15 * * * *"
+    history:
+      # Number of successful job to keep
+      successJob: 1
+      # Number of failed job to keep
+      failJob: 1
+    concurrency:
+      # Skip new job if previous job still active
+      execPolicy: Forbid
+    startingDeadlineSecs: 60
+
 conf:
   features:
     mds: true
@@ -319,6 +333,12 @@ dependencies:
       services:
         - endpoint: internal
           service: ceph_mon
+    pool_checkpgs:
+      jobs:
+        - ceph-rbd-pool
+      services:
+        - endpoint: internal
+          service: ceph_mgr
     checkdns:
       services:
         - endpoint: internal
@@ -449,3 +469,4 @@ manifests:
   job_rbd_pool: true
   service_mgr: true
   pod_test: false
+  cronjob_checkPGs: true

ceph-mon/templates/configmap-bin.yaml

@@ -54,10 +54,6 @@ data:
   moncheck-reap-zombies.py: |
 {{ tuple "bin/moncheck/_reap-zombies.py.tpl" . | include "helm-toolkit.utils.template" | indent 4 }}
 
-  utils-checkPGs.py: |
-{{ tuple "bin/utils/_checkPGs.py.tpl" . | include "helm-toolkit.utils.template" | indent 4 }}
-  utils-checkPGs.sh: |
-{{ tuple "bin/utils/_checkPGs.sh.tpl" . | include "helm-toolkit.utils.template" | indent 4 }}
   utils-checkObjectReplication.py: |
 {{ tuple "bin/utils/_checkObjectReplication.py.tpl" . | include "helm-toolkit.utils.template" | indent 4 }}
   utils-checkDNS.sh: |

ceph-mon/templates/daemonset-mon.yaml

@@ -175,14 +175,6 @@ spec:
               mountPath: /tmp/mon-check.sh
               subPath: mon-check.sh
               readOnly: true
-            - name: ceph-mon-bin
-              mountPath: /tmp/utils-checkPGs.py
-              subPath: utils-checkPGs.py
-              readOnly: true
-            - name: ceph-mon-bin
-              mountPath: /tmp/utils-checkPGs.sh
-              subPath: utils-checkPGs.sh
-              readOnly: true
             - name: ceph-mon-bin
               mountPath: /tmp/checkObjectReplication.py
               subPath: utils-checkObjectReplication.py

ceph-mon/values.yaml

@@ -105,20 +105,6 @@ network:
   public: 192.168.0.0/16
   cluster: 192.168.0.0/16
 
-jobs:
-  pool_checkPGs:
-    # Execute every 15 minutes
-    cron: "*/15 * * * *"
-    history:
-      # Number of successful job to keep
-      successJob: 1
-      # Number of failed job to keep
-      failJob: 1
-    concurrency:
-      # Skip new job if previous job still active
-      execPolicy: Forbid
-    startingDeadlineSecs: 60
-
 conf:
   templates:
     keyring:
@@ -298,4 +284,3 @@ manifests:
   service_mon: true
   service_mon_discovery: true
   job_storage_admin_keys: true
-  cronjob_checkPGs: false

tools/deployment/multinode/030-ceph.sh

@@ -90,7 +90,6 @@ jobs:
       execPolicy: Forbid
     startingDeadlineSecs: 60
 manifests:
-  cronjob_checkPGs: true
   cronjob_defragosds: true
 EOF
 

tools/deployment/osh-infra-logging/020-ceph.sh

@@ -185,7 +185,6 @@ jobs:
       execPolicy: Forbid
     startingDeadlineSecs: 60
 manifests:
-  cronjob_checkPGs: true
   cronjob_defragosds: true
   job_bootstrap: false
 EOF