Deploying in Internal Mode
Internal mode (Standard cluster type in the UI) is the most common deployment method for Ceph storage. It creates a local Ceph cluster in the current workload cluster and distributes data replicas across hard drives on different hosts to maintain availability during single-host failures.
This page uses host devices from the Create Cluster wizard. To back OSDs with Local Storage Block PersistentVolumes instead, see Deploying in Internal Mode with Local Storage.
For a cluster whose nodes run Alauda OS, create the provider MachineTemplate, ConfigPool, and MachineDeployment resources in the global cluster, normally in cpaas-system, and create the StorageCluster/CephCluster in the target workload cluster. Use provider-managed persistent disks and stable /dev/disk/by-id/... identities for OSD devices. Do not follow Replacing or Removing Storage Nodes for an in-place node replacement: do not mark the OSDs out, remove the node entry, or purge the OSDs.
TOC
PrerequisitesPrepare PackagePrepare InfrastructurePrecautionsProcedureRelated OperationsCreate Stretch Type ClusterCleanup Distributed StoragePrerequisites
Prepare Package
-
Download the Alauda Container Platform Storage Essentials installation package corresponding to your platform architecture.
-
Upload the Alauda Container Platform Storage Essentials installation package using the Upload Packages mechanism.
-
Download the Alauda Storage installation package corresponding to your platform architecture.
-
Upload the Alauda Storage installation package using the Upload Packages mechanism.
Prepare Infrastructure
-
At least 3 nodes are required in the storage cluster.
-
Each node must have at least 1 blank hard disk or 1 unformatted hard disk partition available.
-
The available hard disk capacity is recommended to be greater than 50 G.
-
If you are using an attached Kubernetes cluster with Containerd as the runtime component, please ensure that the
LimitNOFILEparameter value in the/etc/systemd/system/containerd.servicefile is configured to1048576on all nodes of the cluster, to ensure successful deployment of distributed storage. For configuration instructions, please refer to Modifying Containerd Configuration Information.Note: When upgrading from versions earlier than v3.10.2 to the current version, if you need to deploy Ceph distributed storage on your custom Kubernetes cluster with Containerd as the runtime component, you must also set the
LimitNOFILEparameter value in the/etc/systemd/system/containerd.servicefile to1048576on all nodes of the cluster.
Precautions
Internal mode deployment and external mode deployment are mutually exclusive. Choose only one deployment method for a cluster.
Procedure
Deploy Alauda Container Platform Storage Essentials
-
Login, go to the Administrator page.
-
Click Marketplace > OperatorHub to enter the OperatorHub page.
-
Find the Alauda Container Platform Storage Essentials, click Install, and navigate to the Install Alauda Container Platform Storage Essentials page.
Configuration Parameters:
(Optional) Install Alauda Container Platform Local Disk Provisioner
To use Selection Device when adding host devices, install Alauda Container Platform Local Disk Provisioner first. It discovers unused disks on cluster nodes and writes LocalVolumeDiscoveryResult objects. Input Device does not require this operator.
See Install Local Storage. To provision Block PersistentVolumes and create OSDs from those volumes, do not continue this wizard. Use Deploying in Internal Mode with Local Storage after Local Storage Guides.
Deploy Operator
-
Navigate to Administrator.
-
In the left sidebar, click Storage Management > Distributed Storage.
-
Click Configure Now.
-
In the Deploy Operator wizard page, click the Deploy Operator button at the bottom right.
-
When the page automatically advances to the next step, it indicates that the Operator has been deployed successfully.
-
If the deployment fails, please refer to the prompt on the interface Clean Up Deployed Information and Retry, and redeploy the Operator; if you wish to return to the distributed storage selection page, click Application Store, first uninstall the resources in the already deployed rook-operator, and then uninstall rook-operator.
-
To upgrade an existing Operator later, set the
Subscriptionupgrade strategy to Manual before you upload a new package. See Upgrade the Operator.
-
Create Cluster
-
In the Create Cluster wizard page, configure the relevant parameters and click the Create Cluster button at the bottom right.
-
Click Advanced Configuration for advanced component configuration.
-
When the page automatically advances to the next step, it indicates that the Ceph cluster has been deployed successfully.
-
If the creation fails, you may click to clean up Created Information or Retry to automatically clean up the resources and recreate the cluster, or manually clean up resources according to the documentation Distributed Storage Service Resource Cleanup.
-
Create Storage Pool
-
In the Create Storage Pool wizard page, configure the relevant parameters and click the Create Storage Pool button at the bottom right.
If it is object storage, you also need to configure the following parameters:
-
When the page automatically advances to the next step, it indicates that the storage pool has been deployed successfully.
-
If the deployment fails, please refer to the interface prompts to check the core components, and then click Clean Up Created Information and Retry to recreate the storage pool.
-
-
Click Create Storage Pool. In the Details tab, you can view information about the created storage pool.
Related Operations
Create Stretch Type Cluster
For details, please refer to Create Stretch Type Cluster.
Cleanup Distributed Storage
For details, please refer to Cleanup Distributed Storage.
Ceph is a trademark or registered trademark of Red Hat, Inc. or its subsidiaries in the United States and other countries. Rook® is a registered trademark of The Linux Foundation. Alauda is an independent vendor. This product is not affiliated with, endorsed by, or sponsored by LF Projects, LLC or The Linux Foundation. All trademarks are the property of their respective owners and are used here for identification purposes only.