Datasets Datasets are where you create and manage the file shares that end users use to complete their everyday work. After you have created one or more pools, you can create datasets within those pools. Shares Sharing from the dataset level is where the admins configure the share protocol and, in the case of SMB, the share name for the dataset. Share Types You can configure the following share types for your BrickStor storage. SMB NFS SMB For SMB shares you have the option to enable the dataset to be shared out as a top-level SMB Share. If you enable Access Based Enumeration (ABE) the system hides the share from anyone browsing via SMB who doesn’t have read access to that share. Host Base Access control further restricts access by source IP. NFS BrickStor supports NFSv3 and NFSv4.0/4.1/4.2. NFS 4 and above supports ACLs while the NFS v3 standard only supports host based access control and POSIX permissions. NFS shares must be the same name as the dataset and share the path of the dataset starting with /storage and then the pool name. With NFS v4.2 clients BrickStor will support context security labels when the Data Security labels box is selected Clicking on the NFS Read/Write Volume will take you to performance metrics related to NFS and the dataset. Creating Datasets When creating a dataset, take note of the following caveats: * You cannot enable or disable dataset encryption after you have created the dataset and committed the changes. * You cannot disable deduplication for any dataset that has had it enabled without moving the data to a new dataset and destroying the old dataset. * Most other operations are reversible; however the changes only apply to new blocks and files as data in the dataset is modified and created. To create a dataset, complete the following steps: In the Connections pane, select either a pool or global container. In the Details pane, click the add icon next to the Children label. You can also click the add icon in the lower portion of the Details pane. The Create Dataset dialog box appears. In the Create Dataset dialog box, type a name for the dataset. Under Type - Storage Profile, choose a storage profile, based on your proposed workload. A storage profile defines a number of settings optimized for a particular kind of workload. Additionally, different storage profiles may have different settings available that are appropriate for that particular workload. This includes which methods are available to share a volume. Volume profiles (e.g. General Volume) create iSCSI volumes, while the profiles that do not contain 'Volume' create datasets that may be accessed using NFS and/or SMB (depending on the particular profile). For example, the VMware Virtual Machines storage profile can only be shared via NFS. Each storage profile also has an associated auto snapshot profile. The associated snapshot profile is the default snapshot policy for any datasets or volumes that are assigned the given dataset profile see Auto Snapshot Data Protection for more information. The available storage profiles are: If you are setting up a File System: General File System Rendering Streaming Media File System Archive File System E-Discovery File System Temp File System If you’re setting up Server Storage: MongoDB Volume MS Exchange Volume Oracle Volume If you are setting up Virtualization Storage: Hyper-V Virtual Machines Hyper-V Virtual Machines Volume VMware VDI VMware Virtual Machines VMware Virtual Machines Volume Xen Virtual Machines If you are setting up a Volume: General Volume Archive Volume Temp Volume If you are setting up a custom file system or volume: Custom File System Custom Volume Select whether to enable Dataset Encryption on this dataset. You must enable encryption during dataset creation. Optionally, enter a Data Quota. Accept the default Data Reservation or enter a new value. Select your desired share type, either: NFS SMB Click Create. In the Changes pane, click Commit Changes. Working with Datasets After you create a dataset, BrickStor SP Manager allows you to modify most settings displayed in the initial create dataset dialog as well as additional settings. Dataset Permissions After you create a dataset, you can configure access control permissions for that dataset. When joined to Active Directory or LDAP you can use AD user names and groups. You can recursively apply permissions to a dataset and its descendants and reset ownership by selecting the appropriate check boxes. Configuring Dataset Permissions To configure dataset permissions, complete the following steps: Select your dataset in the Connections pane Select the Permissions tab in the Detail pane Click the Add Permission button Adding permissions to a dataset Using the Add Permission dialog, you can select previously used users or groups, or search for a user or group. Add permissions search results In the drop-down above the user or group, you can modify the type of permission. The default is Read/Write. Additional options include recursively applying permissions or setting the new user or group as the owner. Once those choices are made, click the Commit button in the Changes pane to apply. Choose permissions options and commit or undo If the Recursively Apply box is not checked, permissions will only apply to newly created files and folders. Files created in existing folders will not be updated. When Recursively Apply is checked, all files and sub-datasets will have permissions overwritten. On datasets with a large number of files, this operation could take some time as each file and folder is updated. Copy Permissions from Another Dataset Admins can copy the permissions of another dataset to the selected data set with the Copy From button. This feature will allow you to copy the permissions of any dataset on any appliance you are currently logged into. Quotas and Reservations After creating a dataset, you can configure quotas and reservations. You can quota only the data or you can quota the data with snapshots and descendants. You can also set reservations on the dataset for both instead of thinly provisioning the dataset. You can type a number and scale such as MB, GB, TB or you can use the slider above the text box to set the quota or reservation. Dataset Bars Throughout the Brickstor SP Manager, dataset bars are used to provide a color-coded quick view of the utilization of a dataset. The fraction of the bar that is filled in represents the amount of space being utilized. Since there are different types of utilization, different colors are used to indicate which category of utilization is shown. There are currently three categories of utilization using the following colors: Purple. The purple bar displays the ability to store data. The purple data bar is displayed if a data reservation has been set, a data quota has been set, the dataset has no children, or there is 25% or less free space for data (5% or less for archive storage profiles). Teal. The teal bar displays the ability to provision sub datasets. This is displayed if the dataset has children. Red/Orange The orange bar is displayed when the data set is low on storage. Depending on quotas, refquotas, reservations and refreservations you could have different free space for each. Instead of showing two bars for each dataset, the UI attempts to show the relevant ones based on each datasets configuration and status. For example, container datasets generally show the teal color because they do not directly store data or snapshots. If the sub datasets do not have any children, the sub dataset will have a purple bar. Dataset Storage Utilization Storage Utilization allows you to view detailed information about the physical storage consumed by a dataset. Viewing Dataset Storage Utilization Statistics In the Connections pane, select a dataset. In the Details pane, select Storage Utilization. iSCSI BrickStor allows you to configure iSCSI targets. iSCSI targets are used by iSCSI initiators to establish a network connection. The target includes LUNs, which are collections of disk blocks accessible via the iSCSI protocol over the network. A target can offer one or more LUNs to the iSCSI clients that initiate a connection with the iSCSI server. The system creates iSCSI volumes under the Global/VBD dataset. In an HA cluster, iSCSI volumes fail over gracefully as part of the pool and resource group to which it was assigned. HA only supports iSCSI for boot devices. Configuring iSCSI Volumes and Sharing as a Target To configure a volume and share as an iSCSI target, complete the following steps: SSH into the BrickStorOS as root. At the BrickStor CLI, enter the following command to enable the target service. svcadm enable -r svc:/network/iscsi/target:default Enter the following command to create the default target. # itadm create-target Now, check the status of your targets to make sure they were properly configured, by running the following command: # itadm list-target –v TARGET NAME STATE SESSIONS iqn.2010-03.com.racktopsystems:02:c434c8d7-5643-6364-af5d-cb0bae33d531 online 0 alias: - auth: none (defaults) targetchapuser: - targetchapsecret: unset tpg-tags: default Open BrickStor SP Manager and log into the BrickStor appliance to complete the iSCSI configuration. In the Connections pane, select a Pool and then select the General tab in the Details pane. In the lower portion of the screen, click the Add icon. In the Create Dataset dialog box, type a name for the dataset. Under Type-Storage Profile, select one of the following options: General Volume Archive Volume Temp Volume Select a Size, either using the slider or by entering a number. Select a Block Size. The dataset block size must match the block on the initiator’s OS when you format the volume. Check Thin Provision if you want to allocate disk storage space in a flexible manner, based on the minimum space required at any given time. Under Enter initiator(s) to share with, type the name of the initiator. You can add multiple initiators in this field. The initiator must be entered in one of the following formats: iqn: iqn.yyyy-mm.reverse-domain-name:unique-name wwn: wwn.01234567ABCDEF eui: eui.01234567ABCDEF Under LUN, leave the field blank if you want the system to auto select the LUN that it will allocate. To manually select a LUN, enter a value. Click Create. In the Changes pane, click Commit Changes. Managing iSCSI Volumes After you create an iSCSI volume, you can manage the volume on the Pool level Sharing tab in BrickStor SP Manager. To manage iSCSI volumes, complete the following steps: In BrickStor SP Manager, select the Pool level in the Connections pane. In the Details pane, select an iSCSI volume under Descendent iSCSI volumes. On the iSCSI page, you can complete any of the following actions: To… Do this… Enable or disable an iSCSI volume Click the toggle switch to either Online or Offline. Delete an initiator Click the adjacent trash icon. Add an initiator Click Add Initiator. Remove initiators Click Remove All. Restore initiators Click Restore All.