design(hld): high level proposal for moveEngine CR

keps/0002-kep-moveengine-hld.md

@@ -0,0 +1,152 @@
+---
+kep-number: 1
+title: MoveEngine - High Level Design
+authors:
+  - "@vitta"
+owners:
+  - Uma Mukkara
+  - Mayank
+editor: 
+creation-date: 2019-05-16
+last-updated: 2019-05-19
+status: provisional
+see-also:
+  - KEP-None
+replaces:
+  - KEP-None
+superseded-by:
+  - KEP-None
+---
+
+# KubeMove - High Level Design
+
+
+
+## Table of Contents
+
+* [Summary](#summary)
+* [Motivation](#motivation)
+    * [Goals](#goals)
+    * [Non-Goals](#non-goals)
+* [Proposal](#proposal)
+    * [Implementation Details/Notes/Constraints](#implementation-detailsnotesconstraints)
+    * [Risks and Mitigations](#risks-and-mitigations)
+* [Design acceptance criteria](#design-acceptance-criteria)
+* [Implementation History](#implementation-history)
+
+## Summary
+
+From the KubeMove HLD KEP, KubeMoveEngine CRD defines the placeholders for the workflow for a given application. The CR is created by the user by invoking the spec that has the following details
+   - Application to which this CR applies
+   - Target location including the namespace. User can specify multiple target locations, in which case multiple KMovePair CRs get created.
+   - The Dynamic Data Mobilizer or DDM details
+   - schedule time to transfer data at regular interval
+   - Telemetry management
+
+`MoveEngine` controller that is watching this CR will deploy the KubeMove framework(`MoveEngine`, `DataSync` controller) at remote cluster to
+setup the application and transfer the data to remote cluster. This controller creates the datasync CR periodically and invokes the `DataSync`
+controller to initiate the data transfer through DDM. Once DDM completes the data transfer, `MoveEngine` controller updates the remote `DataSync` controller by creating dataSync CR at remote cluster to validate/verify the data transfer, and to perform the restore of data.
+
+## Motivation
+
+This KEP is to get into details of MoveEngine CR
+
+### Goals
+
+This KEP is to get into details of MoveEngine CR
+
+### Non-Goals
+
+## Proposal
+
+`spec` of MoveEngine CR consists of following fields:
+```
+    # remoteCluster details
+    movePair: cluster-2
+
+    #namespaces which needs to be migrated, empty for default, all for all namespace
+    namespace: ns1
+
+    # targeted namespace, empty for source namespace
+    remoteNamespace: ns2
+
+    #label selectors
+    selectors:
+    - app.example.com=test
+
+    # sync Period
+    syncPeriod: */5 * * * *
+
+    # moveEngine mode, value can be either Backup or Restore
+    # Restore: if current engine is meant for restore purpose
+    #         In case of restore, there should be only one moveEngine
+    #         CR, with Restore mode, for given namespace and selectors.
+    # Backup : if current engine is meant for backup purpose
+    mode: Restore
+
+    # name of the plugin to be invoked for data transfer of volume
+    # this plugin will setup the necessary infrastructure ,
+    # at source and/or remote, to transfer the app and/or data. 
+    pluginProvider: PLUGIN_NAME
+
+    # includeResources defines if volumes or cluster
+    # resource needs to be synced or not
+    includeResources: true
+status:
+    #syncing stage
+    stage: Final
+
+    #status of last sync
+    lastStatus: Completed
+
+    # last sync time
+    lastSync: dd-mm-yyyy hh:mm:ss
+    volumes:
+    - namespace: ns1
+      persistentVolumeClaim: claim1
+      lastStatus: Completed
+      lastSynced: dd-mm-yyyy hh:mm:ss
+      reason: None
+      volume: xyz
+      remoteVolume: xyz1
+      resources:
+        kind: secret
+        name: app-key
+        phase: Created
+        status: Completed
+        reason: reason for failure
+        lastSyncedTime: last synced time
+```
+Controller does the following:
+1. Check the network connectivity to cluster in movePair CR
+
+2. Discover the resources based on label selector
+TODO: Fill the details regarding discovery of the resources
+
+3. Deploy the resources on destination cluster
+   - Create clientset of destination cluster
+   - Transform the filtered resources. TODO: Fill more details
+   - From source cluster, deploy transformed resources using above clientset in destination cluster
+
+Another approach is to
+ - Install object store in destination cluster
+ - Send the filtered resources to object storage
+ - Transform the resources in object store at destination
+ - From destination cluster, deploy transformed resources
+
+### Implementation Details/Notes/Constraints
+
+TODO
+
+### Risks and Mitigation
+
+
+## Design acceptance criteria
+
+This KEP is in provisional state. To move the design into `accepted` state, at least two sponsors are needed as design approvers from two different companies.
+
+## Implementation History
+
+- None
+
+