You are looking at the documentation of a prior release. To read the documentation of the latest release, please visit here.

New to KubeDB? Please start here.

MemcachedOpsRequest

What is MemcachedOpsRequest

MemcachedOpsRequest is a Kubernetes Custom Resource Definitions (CRD). It provides declarative configuration for Memcached administrative operations like database version updating, horizontal scaling, vertical scaling, etc. in a Kubernetes native way.

MemcachedOpsRequest CRD Specifications

Like any official Kubernetes resource, a MemcachedOpsRequest has TypeMeta, ObjectMeta, Spec and Status sections.

Here, some sample MemcachedOpsRequest CRs for different administrative operations is given below.

Sample MemcachedOpsRequest for updating database version:

apiVersion: ops.kubedb.com/v1alpha1
kind: MemcachedOpsRequest
metadata:
  name: update-memcd
  namespace: demo
spec:
  type: UpdateVersion
  databaseRef:
    name: memcd-quickstart
  updateVersion:
    targetVersion: 1.6.22

Sample MemcachedOpsRequest for horizontal scaling:

apiVersion: ops.kubedb.com/v1alpha1
kind: MemcachedOpsRequest
metadata:
  name: memcd-horizontal-down
  namespace: demo
spec:
  type: HorizontalScaling
  databaseRef:
    name: memcd-quickstart
  horizontalScaling:
    replicas: 2

Sample MemcachedOpsRequest for vertical scaling:

apiVersion: ops.kubedb.com/v1alpha1
kind: MemcachedOpsRequest
metadata:
  name: memcd-vertical
  namespace: demo
spec:
  type: VerticalScaling
  databaseRef:
    name: memcd-quickstart
  verticalScaling:
    memcached:
      resources:
        requests:
          memory: "200Mi"
          cpu: "200m"
        limits:
          memory: "200Mi"
          cpu: "200m"

Sample MemcachedOpsRequest for reconfiguration:

apiVersion: ops.kubedb.com/v1alpha1
kind: MemcachedOpsRequest
metadata:
  name: memcd-reconfig
  namespace: demo
spec:
  type: Reconfigure
  databaseRef:
    name: memcd-quickstart
  configuration:
    applyConfig:
      memcached.conf: |
        -m 50
        -c 50        

MemcachedOpsRequest Spec

A MemcachedOpsRequest object has the following fields in the spec section:

spec.databaseRef

spec.databaseRef is a required field that point to the Memcached object where the administrative operations will be applied. This field consists of the following sub-field:

  • spec.databaseRef.name : specifies the name of the Memcached object.

spec.type

spec.type specifies the kind of operation that will be applied to the database. Currently, the following types of operations are allowed in MemcachedOpsRequest.

  • UpdateVersion
  • HorizontalScaling
  • VerticalScaling
  • Restart
  • Reconfigure

spec.updateVersion

If you want to update your Memcacheds version, you have to

You can only update between Memcached versions. KubeDB does not support downgrade for Memcached.

spec.horizontalScaling

If you want to scale-up or scale-down your Memcached, you have to specify spec.horizontalScaling section. This field consists of the following sub-field:

  • spec.horizontalScaling.replicas indicates the desired number of replicas for your Memcahced instance after scaling. For example, if your cluster currently has 4 replicas, and you want to add additional 2 replicas then you have to specify 6 in spec.horizontalScaling.replicas field. Similarly, if you want to remove one replicas, you have to specify 3 in spec.horizontalScaling.replicas field.

spec.verticalScaling

spec.verticalScaling is a required field specifying the information of Memcached resources like cpu, memory that will be scaled. This field consists of the following sub-fields:

  • spec.verticalScaling.memcached indicates the Memcached server resources. It has the below structure:
requests:
  memory: "200Mi"
  cpu: "200m"
limits:
  memory: "200Mi"
  cpu: "200m"

Here, when you specify the resource request for Memcached container, the scheduler uses this information to decide which node to place the container of the Pod on and when you specify a resource limit for Memcached container, the kubelet enforces those limits so that the running container is not allowed to use more of that resource than the limit you set. You can found more details from here.

  • spec.verticalScaling.exporter indicates the exporter container resources. It has the same structure as spec.verticalScaling.memcached and you can scale the resource the same way as memcached container.

You can increase/decrease resources for both memcached container and exporter container on a single MemcachedOpsRequest CR.

spec.timeout

As we internally retry the ops request steps multiple times, this timeout field helps the users to specify the timeout for those steps of the ops request (in second). If a step doesn’t finish within the specified timeout, the ops request will result in failure.

spec.apply

This field controls the execution of opsRequest depending on the database state. It has two supported values: Always & IfReady. Use IfReady, if you want to process the opsRequest only when the database is Ready. And use Always, if you want to process the execution of opsReq irrespective of the Database state.

MemcachedOpsRequest Status

After creating the Ops request status section is added in MemcachedOpsRequest CR. The yaml looks like following:

  apiVersion: ops.kubedb.com/v1alpha1
  kind: MemcachedOpsRequest
  metadata:
    annotations:
      kubectl.kubernetes.io/last-applied-configuration: |
        {"apiVersion":"ops.kubedb.com/v1alpha1","kind":"MemcachedOpsRequest","metadata":{"annotations":{},"name":"memcached-mc","namespace":"demo"},"spec":{"databaseRef":{"name":"mc1"},"type":"VerticalScaling","verticalScaling":{"memcached":{"resources":{"limits":{"cpu":"200m","memory":"200Mi"},"requests":{"cpu":"200m","memory":"200Mi"}}}}}}        
    creationTimestamp: "2024-08-26T10:41:05Z"
    generation: 1
    name: memcached-mc
    namespace: demo
    resourceVersion: "4572839"
    uid: f230978f-fc7b-4f8e-afad-6023910aba0e
  spec:
    apply: IfReady
    databaseRef:
      name: mc1
    type: VerticalScaling
    verticalScaling:
      memcached:
        resources:
          limits:
            cpu: 200m
            memory: 200Mi
          requests:
            cpu: 200m
            memory: 200Mi
  status:
    conditions:
    - lastTransitionTime: "2024-08-26T10:41:05Z"
      message: Memcached ops request is vertically scaling database
      observedGeneration: 1
      reason: VerticalScale
      status: "True"
      type: VerticalScale
    - lastTransitionTime: "2024-08-26T10:41:08Z"
      message: Successfully updated PetSets Resources
      observedGeneration: 1
      reason: UpdatePetSets
      status: "True"
      type: UpdatePetSets
    - lastTransitionTime: "2024-08-26T10:41:13Z"
      message: evict pod; ConditionStatus:True; PodName:mc1-0
      observedGeneration: 1
      status: "True"
      type: EvictPod--mc1-0
    - lastTransitionTime: "2024-08-26T10:41:13Z"
      message: is pod ready; ConditionStatus:False
      observedGeneration: 1
      status: "False"
      type: IsPodReady
    - lastTransitionTime: "2024-08-26T10:41:18Z"
      message: is pod ready; ConditionStatus:True; PodName:mc1-0
      observedGeneration: 1
      status: "True"
      type: IsPodReady--mc1-0
    - lastTransitionTime: "2024-08-26T10:41:18Z"
      message: is pod resources updated; ConditionStatus:True; PodName:mc1-0
      observedGeneration: 1
      status: "True"
      type: IsPodResourcesUpdated--mc1-0
    - lastTransitionTime: "2024-08-26T10:41:18Z"
      message: Successfully Restarted Pods With Resources
      observedGeneration: 1
      reason: RestartPods
      status: "True"
      type: RestartPods
    - lastTransitionTime: "2024-08-26T10:41:18Z"
      message: Successfully Vertically Scaled Database
      observedGeneration: 1
      reason: Successful
      status: "True"
      type: Successful
    observedGeneration: 1
    phase: Successful

.status describes the current state of the MemcachedOpsRequest operation. It has the following fields:

status.phase

status.phase indicates the overall phase of the operation for this MemcachedOpsRequest. It can have the following three values:

PhaseMeaning
SuccessfulKubeDB has successfully performed the operation requested in the MemcachedOpsRequest
FailedKubeDB has failed the operation requested in the MemcachedOpsRequest
DeniedKubeDB has denied the operation requested in the MemcachedOpsRequest

status.observedGeneration

status.observedGeneration shows the most recent generation observed by the MemcachedOpsRequest controller.

status.conditions

status.conditions is an array that specifies the conditions of different steps of MemcachedOpsRequest processing. Each condition entry has the following fields:

  • types specifies the type of the condition. MemcachedOpsRequest has the following types of conditions:
TypeMeaning
ProgressingSpecifies that the operation is now progressing
SuccessfulSpecifies such a state that the operation on the database has been successful.
HaltDatabaseSpecifies such a state that the database is halted by the operator
ResumeDatabaseSpecifies such a state that the database is resumed by the operator
FailureSpecifies such a state that the operation on the database has been failed.
ScalingSpecifies such a state that the scaling operation on the database has stared
VerticalScalingSpecifies such a state that vertical scaling has performed successfully on database
HorizontalScalingSpecifies such a state that horizontal scaling has performed successfully on database
UpdateVersionSpecifies such a state that version updating on the database have performed successfully
  • The status field is a string, with possible values "True", "False", and "Unknown".
    • status will be "True" if the current transition is succeeded.
    • status will be "False" if the current transition is failed.
    • status will be "Unknown" if the current transition is denied.
  • The message field is a human-readable message indicating details about the condition.
  • The reason field is a unique, one-word, CamelCase reason for the condition’s last transition. It has the following possible values:
ReasonMeaning
OpsRequestProgressingStartedOperator has started the OpsRequest processing
OpsRequestFailedToProgressingOperator has failed to start the OpsRequest processing
SuccessfullyHaltedDatabaseDatabase is successfully halted by the operator
FailedToHaltDatabaseDatabase is failed to halt by the operator
SuccessfullyResumedDatabaseDatabase is successfully resumed to perform its usual operation
FailedToResumedDatabaseDatabase is failed to resume
DatabaseVersionupdatingStartedOperator has started updating the database version
SuccessfullyUpdatedDatabaseVersionOperator has successfully updated the database version
FailedToUpdateDatabaseVersionOperator has failed to update the database version
HorizontalScalingStartedOperator has started the horizontal scaling
SuccessfullyPerformedHorizontalScalingOperator has successfully performed on horizontal scaling
FailedToPerformHorizontalScalingOperator has failed to perform on horizontal scaling
VerticalScalingStartedOperator has started the vertical scaling
SuccessfullyPerformedVerticalScalingOperator has successfully performed on vertical scaling
FailedToPerformVerticalScalingOperator has failed to perform on vertical scaling
OpsRequestProcessedSuccessfullyOperator has completed the operation successfully requested by the OpeRequest cr
  • The lastTransitionTime field provides a timestamp for when the operation last transitioned from one state to another.
  • The observedGeneration shows the most recent condition transition generation observed by the controller.

Next Steps

  • Learn about Memcached crd here.
  • Deploy your first Memcached database with Memcached by following the guide here.