The XVM Manager GUI and the CXFS Manager GUI contain the same tasks for configuring and administering XVM volumes. This guide documents only those tasks that pertain to XVM volume elements, disks, local filesystems, and privileges. For information about cluster functions and CXFS filesystems, see the CXFS 7 Administrator Guide for SGI InfiniteStorage.

To start the GUI using the command line, do the following:

---

Note: If CXFS is installed, the xvmgr command will automatically invoke the CXFS Manager GUI, which provides access to the same set of XVM tasks as the XVM Manager GUI.

---

If you want to use a web-based version of the GUI, do the following:

Table 10-1 describes the platforms where the GUI may be started, connected to, and displayed.

To ensure that the required GUI privileges are available for performing all of the tasks, you should log in to the GUI as root. However, some or all privileges can be granted to any other user using the GUI privilege tasks; see “Privileges Tasks”.

A dialog box will prompt you to log in to an XVM host. You can choose one of the following connection types:

The XVM Manager GUI provides access to the tools that help you configure and administer XVM:

This section discusses the following:

GUI Window Layout

Figure 10-1 shows the XVM Manager window.

Figure 10-1. XVM Manager GUI

By default, the window is divided into two sections: the view area to the left and the details area to the right. The details area shows generic overview text if no item is selected in the view area. You can use the arrows in the middle of the window to shift the display.

The command buttons along the top of the window provide a method of performing tasks quickly. When you click a button, the corresponding task executes using default values, usually without displaying a task window. For more information on the command buttons, see “Shortcuts Using Command Buttons”.

Use the Find text field to view and select single or multiple items, as described in “Selecting Items to View or Modify”.

Choose what you want to appear in the view area from the View menu.

To view the details of any volume element, select it; see “Selecting Items to View or Modify”. The configuration and status details for the item will appear in the details area to the right, along with the Applicable Tasks list, which displays tasks you may wish to launch after evaluating the item's configuration details. Click a task to launch it; based on the item selected, default values will appear in the task window. For information on launching tasks, see “Performing Tasks”.

Figure 10-2 shows an XVM GUI window with a selected component and the details area showing the details for that component.

Figure 10-2. Item Selected

To see the configuration and status details about an item in the details area, select its name (which will appear in blue); details will appear in a new window.

In general, clicking on blue text yields a new window display, which could contain one of the following:

• Item details

• Term definitions

• Input instructions

• Task windows

When you are running XVM as a standalone product, the Domain is set to local by default in various tasks, indicating that the XVM objects that you create in this session are in the local domain. For more information about XVM domains, see “Local Domain and Cluster Domain” in Chapter 2.

Shortcuts Using Command Buttons

The command buttons along the top of the GUI window provide a method of performing tasks quickly. When you click a button, the corresponding task executes using default values, usually without displaying a task window. To override the defaults, launch the task from the Tasks menu. Table 10-2 summarizes the shortcuts.

Table 10-2. XVM Manager GUI Command Buttons

Button	Task
	Labels selected disks (if the selected disks include foreign and/or labeled disks, the Label Disks task will be run)
	Brings up the Slice Disk task with the selected disks as default inputs
	Creates a concat with a temporary name
	Creates a mirror with a temporary name
	Creates a stripe with a temporary name
	Creates a data subvolume with a temporary name
	Creates a volume with a temporary name
	Starts the Performance Co-Pilot XVM I/O monitor pmgxvm on the server, displaying via X Windows to your local administration station (for information on using Performance Co-Pilot, see Performance Co-Pilot for Linux User's and Administrator's Guide)
	Detaches the selected volume elements from their current parents
	Deletes the selected non-slice volume elements or unlabels the selected disks directly, or brings up the appropriate Delete task window for the selected component

Using Drag-and-Drop

The GUI lets you use drag-and-drop to do the following:

• Structure volume topologies

• Administer XVM disks

---

Caution: Always exercise care when restructuring volume elements with drag-and-drop because data that resides on the volume element can be lost. The GUI attempts to warn the user when it can predict that there is a high likelihood of data loss. However, when a volume is not associated with a mounted filesystem, neither the xvm command nor the GUI can determine whether that volume holds important data.

---

To select multiple GUI icons, select the first icon by clicking the left mouse button, then press the Ctrl button while clicking on the additional icons. To select consecutive icons, select the first icon and press shift while selecting the last icon.

You cannot use drag-and-drop between two GUI windows or with shortcut command buttons.

Configuring Disks

To label or unlabel disks using drag-and-drop, select the following:

View -> Disks

Select an unlabeled disk then drag and drop it on the Labeled Disks heading, or select a labeled disk then drag and drop it on the Unlabeled Disks heading.

You can give away a disk using the task menu or drag-and-drop. In the Disks view, select a disk and then drag and drop it on the Cluster Disks heading.

---

Note: Giving away a disk presents less risk of data loss than stealing a disk.

---

You can label a disk by clicking a selected (highlighted) disk and typing a name into the resulting name text field.

Table 10-3 shows keys to the icons used in the XVM Manager GUI.

Table 10-3. Key to XVM Manager GUI Icons

Icon	Entity
	XVM disk
	Unlabeled disk
	Foreign disk
	Slice
	Volume
	Subvolume
	Concat
	Mirror
	Stripe
	Slot
	Local filesystem
	Expanded tree
	Collapsed tree
	Copy on write
	Repository

Table 10-4 shows keys to the states used in the XVM Manager GUI.

Table 10-4. Key to XVM Manager GUI States

Icon	State
	(Grey icon) Defined, offline, inactive or unknown
	Enabled for mount
	(Blue icon) Online, ready for use, up, or mounted without error
	(Green swatch) Open, in use
	(Orange arrow) Mirror reviving
	Disabled
	(Red icon) Error detected, down or mounted with error

A concat is a volume element that concatenates all of its children into one address space.

When you create a concat, the system automatically creates a parent volume and data subvolume for the concat. If you explicitly name this generated volume, the volume name is stored in label space and persists across reboots.

---

Note: To achieve good I/O performance, use the /etc/failover2.conf file. See Chapter 6, “XVM Path Failover”.

---

To create a concat, do the following:

A mirror is a volume element that maintains identical data images on its underlying volume elements. A mirror cannot have more than eight legs. You can specify whether a particular leg of a mirror is to be preferred for reading by marking it as a primary leg.

---

Note: The components of a mirror do not have to be identical in size, but if they are not there will be unused space in the larger components.

---

When you create a mirror, the system automatically creates a parent volume and data subvolume for the mirror. If you explicitly name this generated volume, the name is stored in label space and persists across reboots.

---

Note: To achieve good I/O performance, use the /etc/failover2.conf file. See Chapter 6, “XVM Path Failover”.

---

To create a mirror, do the following:

A stripe is a volume element that distributes a set of volume elements across an address space.

---

Note: It is legal to create a stripe that consists of volume elements of unequal size, although this may leave some space unused.

---

The actual size of the stripe depends on the stripe unit size and the size of the volume elements that make up the stripe. In the simplest case, the volume elements are all the same size and are an even multiple of the stripe unit size. For example, if the stripe unit is 128 512-byte blocks (the default) and you create a stripe consisting of two slices that are each 256,000 blocks, all of the space on each slice is used. The stripe size is the full 512,000 blocks of the two slices.

On the other hand, if two slices that make up a stripe are each 250,000 blocks and the stripe unit is 128 blocks, then only 249,984 of the blocks on each slice can be used for the stripe and the size of the stripe will be 499,968 blocks. This situation may arise when you create the slices on a disk by dividing the disk equally, or use the entire disk as a slice, and do not coordinate the resulting stripe size with the stripe unit size.

Even if one of the two slices that make up the two-slice stripe in the second example is 256,000 blocks (while the other is 250,000 blocks), the stripe size will be 499,968 blocks because the same amount of space in each volume element that makes up the slice is used.

Following is the general formula for determining the stripe size, where stripe_width is the number of volume elements that make up the stripe (using integer arithmetic):

stripe_size = (smallest_stripe_leg / stripe_unit) * stripe_unit * stripe_width

When you create a stripe, the system automatically creates a parent volume and data subvolume for the stripe. If you explicitly name this generated volume, the name is stored in label space and persists across reboots.

---

Note: To achieve good I/O performance, use the /etc/failover2.conf file. See Chapter 6, “XVM Path Failover”.

---

To create a stripe, do the following:

When you delete a volume element, its parent remains and has an open slot in the XVM topology tree. An element of an open subvolume can only be deleted if its deletion will not cause the subvolume to go offline. The only element that can be deleted without putting the subvolume offline is a mirror leg that is not the last leg of that mirror.

If a volume element contains any attached children, it cannot be deleted unless you choose to delete all of its descendents, or all of its descendants except slices (detaching and keeping the slices).

To delete one or more volume elements, do the following:

You can rename a volume, subvolume, concat, stripe, or mirror. (You cannot change the name of a slice.) The name you give an object when you explicitly rename it remains persistent across reboots.

To rename a volume element, do the following:

You can insert a mirror or a concat above another volume element other than a volume or a subvolume.

You can grow a volume element on a running system by inserting a concat and you can add mirroring on a running system by inserting a mirror. The volume element you are growing or mirroring can be part of an open subvolume and can have active I/O occurring.

To insert a new mirror/concat above an existing volume element, do the following:

You can remove an unneeded mirror or concat from an XVM topology tree if the mirror or concat has only one child. This task links the child to the parent of the mirror or concat that you remove.

You can remove a mirror or concat in an open subvolume. Generally, you perform this task during configuration to reverse a previous insert operation.

To remove mirrors and concats with only one child, do the following:

When a volume element is disabled, no I/O will be issued to it until you explicitly enable it.

To enable one or more volume elements that have been disabled, do the following:

When a volume element is disabled, no I/O will be issued to it until you explicitly enable the volume element. An object remains disabled across reboots.

To disable one or more volume elements, do the following:

A volume element is online when it is properly configured. It is able to be opened, or it is already open.

The system kernel may disable a volume element and take that element offline. This could happen, for example, when a mirror leg shows an I/O error.

To bring a set of volume elements online, do the following:

When you remake a volume element, you collapse holes in the XVM topology tree beneath volume element. The children remain in their current order.

To remake a volume element, do the following:

You can detach volume elements from their parents.

An element of an open subvolume can only be deleted if its deletion will not cause the subvolume to go offline. The only element that can be deleted without putting the subvolume offline is a mirror leg that is not the last leg of that mirror. You cannot detach the last valid leg of an open mirror from that mirror, because this will cause the mirror to go offline.

To detach volume elements, do the following:

When you create a volume, you can specify subvolumes to attach to the volume.

A volume cannot have more than one subvolume of a given type. The system-defined subvolumes are data, log, and rt. See “Subvolume” in Chapter 2.

To create a volume, do the following:

When you create a subvolume you can optionally attach a concat, mirror, stripe, or slice to the subvolume. The volume element attached to the subvolume cannot be a volume or another subvolume.

There cannot be more than one subvolume with the same type under a volume. For example, there can be only one data subvolume under a volume. A user-defined subvolume type is in the range 16 through 255. You can have multiple user-defined subvolumes under a volume if each has a unique type number. Subvolume types in the user-defined range are not interpreted in any way by the system. See “Subvolume” in Chapter 2.

When you create a subvolume, you specify the user ID of the subvolume owner (default 0), the group ID of the subvolume owner (default is 0), and the mode of the subvolume (default is 0644).

If you do not specify a child, an empty subvolume is created that can be attached to later.

When you create a subvolume, the system automatically creates a parent volume for it. If you explicitly name this generated volume, the name is stored in label space and persists across reboots.

To create a subvolume, do the following:

You can change user ID, group ID, and the permissions mode that apply to one or more subvolumes.

Do the following:

Statistics collection for a subvolume, stripe, concat, mirror, or slice may be set to on or off and you can reset the current statistics to 0. Statistics are enabled/disabled only for the volume elements that you specify. Statistics for volume elements show the number of read and write operations as well as the number of 512-byte blocks read and written.

In a clustered environment, statistics are maintained for the local node only.

To modify statistics collection on one or more volume elements, do the following:

To create XVM logical volumes on a physical disk, you must label the disk as an XVM disk. Labeling a disk writes an XVM physical volume (physvol) label on a disk and allows XVM to control the partitioning on the disk. In a CXFS cluster, any physvols that will be shared must be physically connected to all nodes in the cluster.

You cannot label a disk as an XVM disk if the disk contains any partitions that are currently in use as mounted filesystems. On systems with many disks, these checks can be time-consuming. You can specify whether you want to override this restriction and not check for in-use partitions.

---

Caution: Data corruption or system panics can result from labeling disks with partitions that are in use.

---

When you label an XVM disk, you can specify how much space to assign to the volume header; the default value is the number of blocks currently in the volume header of the disk being labeled. You can also specify how much space to assign to the XVM label area in the volume header; the default is 1024 blocks (usually leaving 3072 blocks in the volume header that are not part of the XVM label area).

The default values for the volume header size and label area size support approximately 5000 XVM objects; this should be sufficient for most volume configurations. If you will have more than that many objects on the physvol that the label area must maintain, you may need to increase the XVM label area size. As a rule of thumb, one block is required for every seven objects.

---

Note: A volume element and a name for a volume element count as two objects.

---

If you add a new disk that has already been labeled as a physvol to a running system, you must manually probe the disk with XVM in order for the system to recognize the disk as an XVM disk. You do not have to do this when you are labeling a new XVM disk on your system, however, because XVM probes the disk as part of the label process. All disks are probed when the system is booted to determine which disks are XVM disks.

To label a disk as an XVM disk, do the following:

Slicing a disk creates a slice from a block range of a physvol. You can specify the starting block and the length of a slice. You can choose one of the following methods

Slices are named automatically and are persistent across reboots. You cannot rename slices.

If you do not supply a slice length, the address range will be from the indicated start block to the end of the free area containing the start block.

When volume elements other than volumes are created, the system automatically creates a volume and a subvolume that are associated with the volume element. When you create a volume element, you can explicitly name this generated volume, in which case the volume name is stored in label space and persists across reboots.

To create slices on an XVM disk, do the following:

You can rename a disk (physvol). The name you give is persistent across reboots. Do the following:

To remove a physvol from a system, you must remove the XVM label from the physvol. After unlabeling, the original partitioning information is restored from a file saved in the volume header directory under the name backvh.

In a clustered environment, you cannot unlabel a physvol that is not attached to the system you are working from.

You cannot unlabel physvols containing slices unless you also delete the slices on those physvols. When you indicate that you are deleting all slices on the disk, the slices are deleted even if the slice is part of an open subvolume and its deletion will cause the subvolume state to go offline. If any of the attempts to delete a slice fails when you are unlabeling a physvol, the unlabel will fail. If all deletes succeed, the physvol will be unlabeled.

To remove XVM labels from one or more physvols with the GUI, do the following:

Statistics for physvols show the number of read and write operations as well as the number of 512-byte blocks read and written. You can turn statistics collection for a physvol to on or off and you can reset the current statistics to 0.

In a clustered environment, statistics are maintained for the local node only.

To modify statistics collection on one or more disks with the GUI, do the following:

---

Note: This task is valid in a clustered environment only.

---

You can gracefully change the ownership of a physvol by giving it to the local domain of another machine (by default) or to another cluster (if you select Another Cluster).

---

Note: However, you cannot give away ownership of a physvol that has slices that are part of open subvolumes. For this reason, an attempt to give away a disk will fail during a mirror revive. In general, you must unmount filesystems on XVM volumes that contain the physvol and wait for mirror revives to complete before giving away a physvol.

---

Giving a disk away will result in all slices on the disk (and any empty parents that result) being deleted. The configuration information will be retained on the disk. Subvolumes that span disks might go offline if giving a disk away will cause slices belonging to that physvol to be removed.

When you give a disk away, the new owning node or cluster must read the disk before the configuration is visible to the new owner. This happens either automatically on reboot or when the new owner probes the disk.

You can specify a physvol to give away by either the physvol name or the disk name.

To give ownership of one or more disks to another node or cluster, do the following:

---

Note: This task is valid in a clustered environment only.

---

In some circumstances, the node or cluster that currently owns the physvol may be unable to give a disk away. In these cases, you can steal a disk to change the ownership of a physvol. Only physvols that are foreign to the current node or cluster can be the targets of a steal.

---

Caution: Stealing a disk unconditionally resets the owner of a physvol to the current node or cluster. No attempt is made to inform the previous owner of the ownership change. If another host or cluster has the physvol instantiated, configuration corruption or data corruption could occur. You should forcibly steal a disk only when ownership cannot be changed by giving the disk away. In a situation where you must steal a disk, you can determine the ownship by using the xvm(8) command show -v foreign; see “Foreign Disks” in Chapter 5.

---

You cannot steal a physvol that has slices that are part of open subvolumes. In general, you must unmount filesystems on XVM volumes that contain the physvol and wait for mirror revives to complete before stealing the physvol.

To take control of a foreign disk, do the following:

If you add to a running system a disk that was already labeled as a physvol, you must manually probe the disk in order for the system to recognize it as an XVM disk. (You do not have to do this when you label a new XVM disk because XVM automatically probes the disk as part of the label process.)

---

Note: It is assumed that the disk to be probed is available in the hardware inventory (that is, the controller that it is connected to has been probed outside of XVM).

---

To probe one or more disks for XVM label information with the GUI, do the following:

If the disk being probed has not been previously labeled by XVM, an error is returned.

You can dump configuration information for an individual volume element or all of its children. You can also use dump configuration commands for a physvol; you must explicitly dump the physvol topology tree separately from a volume element topology tree. Dumping configuration information allows you to replace a disk in a running system and to regenerate the XVM configuration on the new disk without rebooting the system.

When you dump and regenerate a device, you do not regenerate the data on the disk you are replacing, but rather you regenerate the XVM configuration on the new disk.

When you dump a volume element, a new UUID is generated for the object being dumped in order to avoid any possible name collision issues when the object is later re-created.

To dump the XVM configuration commands, do the following:

This task lets you create a filesystem on a volume that is online but not open. To create filesystems on multiple volume elements, use the Browse button.

---

Caution: Clicking OK will erase all data that exists on the target volume.

---

To make a filesystem, do the following:

This task lets you grow a mounted filesystem.

---

Note: Before you can grow a filesystem, you must first increase the size of the volume on which the filesystem is mounted. You can launch the Insert Mirrors or Concats above Volume Elements task to add a concat, or you can use the drag-and-drop mechanism to attach a slice to an existing concat. You cannot add a slice to an existing volume element if this changes the way that the data is laid out in that volume element or in any of its ancestor.

---

To grow a filesystem, do the following:

This task lets you mount a filesystem only on the node to which the GUI is connected (the local node).

To mount a filesystem locally, do the following:

For more information, see the mount(8) man page.

To unmount a filesystem from the local node, do the following:

This task lets you delete a local filesystem's mount information in the /etc/fstab file.

---

Note: The filesystem will still be present on the volume.

---

Do the following:

You can grant access to a specific task to one or more users at a time.

Do the following:

Click Next to move to the next page.

To see which tasks a specific user can access, select View: Users. Select a specific user to see details about the tasks available to that user.

To see which users can access a specific task, select View: Task Privileges. Select a specific task to see details about the users who can access it and the privileged commands it requires.

You can revoke task access from one or more users at a time.

Do the following:

To see which tasks a specific user can access, select View: Users. Select a specific user to see details about the tasks available to that user.

To see which users can access a specific task, select View: Task Privileges. Select a specific task to see details about the users who can access it.