Online Data Manager (ODM) overview and installation

Online Data Manager (ODM) overview and installation

This overview and installation document provides information about licensing Online Data Manager (ODM) products, software notes and recommendations, and procedures for initializing the VERITAS^® Volume Manager. It also includes a complete list of ODM documentation.

Licensing ODM products

When you install the ODM software, the VERITAS Volume Manager (VxVM) and Visual Administrator (VxVA) are installed but not fully enabled. A license is required to fully enable the ODM products. You can obtain the ODM license from your software vendor or from the site where you downloaded the ODM software.

Without a license, the following ODM features are supported:

The Volume Manager includes the following functionality:
- Concatenation (data is mapped linearly onto one or more subdisks in a plex)
- Spanning (concatenation using subdisks that reside on more than one VM disk)
The Visual Administrator will execute only those tasks that apply to the Volume Manager features described above.

Once you license ODM, the following features are enabled:

Mirroring (using multiple plexes to duplicate the information contained in a volume)
Striping (mapping data so that the data is interleaved among two or more physical disks)
RAID-5 (a method of providing data redundancy through the use of parity -- a calculated value that can be used to reconstruct data after a failure)
Logs for mirror recovery
All features of the Visual Administrator

For instructions on how to apply your ODM license to the system, see ``Licensing and registering SCO OpenServer products''.

ODM documentation

In addition to this overview, the following ODM guides are provided.

VERITAS Volume Manager System Administrator's Guide
This guide provides information about VxVM configuration, performance, and command line usage. It also includes VxVM error messages and recovery procedures.
VERITAS Volume Manager User's Guide
This guide provides basic procedures for volume management and system administration using the command line and the Volume Manager Visual Administrator (VxVA), the graphical interface to ODM.

ODM manual pages

These manual pages describe the ODM commands, system files, and interfaces.

vxassist(ADM)
vxbootsetup(ADM)
vxconfigd(ADM)
vxconvert(ADM)
vxdctl(ADM)
vxdg(ADM)
vxdisk(ADM)
vxdiskadd(ADM)
vxdiskadm(ADM)
vxdisksetup(ADM)
vxdump(ADM)
vxedit(ADM)
vxencap(ADM)
vxevac(ADM)
vxinfo(ADM)
vxintro(ADM)
vxiod(ADM)
vxmake(ADM)
vxmend(ADM)
vxmirror(ADM)
vxnotify(ADM)
vxplex(ADM)
vxprint(ADM)
vxreattach(ADM)
vxrecover(ADM)
vxrelocd(ADM)
vxresize(ADM)
vxrestore(ADM)
vxrootmir(ADM)
vxsd(ADM)
vxserial(ADM)
vxsparecheck(ADM)
vxstat(ADM)
vxtrace(ADM)
vxtunefs(ADM)
vxunroot(ADM)
vxupgrade(ADM)
vxvol(ADM)
plexrec(F)
sdrec(F)
tunefstab(F)
vol_pattern(F)
volrec(F)
vxmake(F)
vxconfig(HW)
vxfsio(HW)
vxinfo(HW)
vxio(HW)
vxiod(HW)
vxtrace(HW)

VxVM basic and Mirroring products documentation

In general, the information in the VERITAS Volume Manager (VxVM) User's Guide and the VERITAS Volume Manager (VxVM) System Administrator's Guide applies to the VxVM ``basic'' product and to the Mirroring product, with some exceptions.

The exceptions are listed in the following tables. All the listed topics do not apply to the Volume Manager ``basic'' product. In addition, the tables list which topics do not apply to the Mirroring product.

For example, in the ``Volume Manager (VxVM) User's Guide'' table, the topic ``Creating a File System on a Mirrored Volume,'' does not apply to the VxVM ``basic'' product, but does apply to the Mirroring product.

Chapters and topics that are not listed in the tables are applicable for all products, or are general, descriptive information.

Volume Manager (VxVM) System Administrator's Guide

Topics
(none of these apply to the basic VxVM product)

Applies to
Mirroring product?

Chapter 4 - Disk Operations

4.3.3 Adding a VM Disk to the Hot-Relocation Pool No

4.3.4 Removing a VM Disk from the Hot-Relocation Pool No

Chapter 5 - File System Operations

5.3.2 Creating a File System on a Striped Volume No

5.3.3 Creating a File System on a RAID-5 Volume No

5.3.4 Creating a File System on a Mirrored Volume Yes

5.12.2.2 Striped Volume/FS Create Form No

5.12.2.3 RAID-5 Volume/FS Create Form No

Chapter 6 - Volume Operations

6.2.2 Creating a Striped Volume No

6.2.3 Creating a RAID-5 Volume No

6.2.4 Creating a Mirrored Volume Yes

6.2.5 Creating a Striped and Mirrored Volume No

6.4 Adding a Mirror to a Volume Yes

6.5 Adding a Log No

6.6 Removing a Mirror or a Log Yes

6.8 Preparing a Volume for Online Backup No

6.11.2.2 Striped Volume/FS Create Form No

6.11.2.3 RAID-5 Volume/FS Create Form No

6.11.2.5 Snapshot Form No

Chapter 7 - Plex Operations

7.5 Detaching a Plex No

7.6 Attaching a Plex No

Chapter 8 - Subdisk Operations

8.4.1 Log Subdisks No

Chapter 10 - Disk CLI Operations

10.3.2 Adding a VM Disk to the Hot-Relocation Pool No

10.3.3 Removing a VM Disk From the Hot-Relocation Pool No

10.4 Mirroring the Boot Disk Yes

Chapter 12 - Volume CLI Operations

12.2.2 Creating a Striped Volume No

12.2.3 Creating a RAID-5 Volume No

12.3 Removing a Volume No

12.4 Mirroring a Volume Yes

12.5 Adding a RAID-5 Log No

12.6 Adding a DRL Log No

12.7 Removing a Mirror No

12.8 Removing a Raid-5 Log No

12.9 Extending a Volume No

12.10 Shrinking a Volume No

12.11 Preparing a Volume for Online Backup No

Chapter 13 - Menu Interface Operations

13.3.4 Adding a VM Disk to the Hot-Relocation Pool No

13.3.5 Removing a VM Disk From the Hot-Relocation Pool No

13.4.1 Mirroring Volumes on a VM Disk Yes

13.4.2 Moving Volumes from a VM Disk No

Topics (none of these apply to the basic VxVM product)	Applies to Mirroring product?
Chapter 4 - Disk Operations
4.3.3 Adding a VM Disk to the Hot-Relocation Pool	No
4.3.4 Removing a VM Disk from the Hot-Relocation Pool	No
Chapter 5 - File System Operations
5.3.2 Creating a File System on a Striped Volume	No
5.3.3 Creating a File System on a RAID-5 Volume	No
5.3.4 Creating a File System on a Mirrored Volume	Yes
5.12.2.2 Striped Volume/FS Create Form	No
5.12.2.3 RAID-5 Volume/FS Create Form	No
Chapter 6 - Volume Operations
6.2.2 Creating a Striped Volume	No
6.2.3 Creating a RAID-5 Volume	No
6.2.4 Creating a Mirrored Volume	Yes
6.2.5 Creating a Striped and Mirrored Volume	No
6.4 Adding a Mirror to a Volume	Yes
6.5 Adding a Log	No
6.6 Removing a Mirror or a Log	Yes
6.8 Preparing a Volume for Online Backup	No
6.11.2.2 Striped Volume/FS Create Form	No
6.11.2.3 RAID-5 Volume/FS Create Form	No
6.11.2.5 Snapshot Form	No
Chapter 7 - Plex Operations
7.5 Detaching a Plex	No
7.6 Attaching a Plex	No
Chapter 8 - Subdisk Operations
8.4.1 Log Subdisks	No
Chapter 10 - Disk CLI Operations
10.3.2 Adding a VM Disk to the Hot-Relocation Pool	No
10.3.3 Removing a VM Disk From the Hot-Relocation Pool	No
10.4 Mirroring the Boot Disk	Yes
Chapter 12 - Volume CLI Operations
12.2.2 Creating a Striped Volume	No
12.2.3 Creating a RAID-5 Volume	No
12.3 Removing a Volume	No
12.4 Mirroring a Volume	Yes
12.5 Adding a RAID-5 Log	No
12.6 Adding a DRL Log	No
12.7 Removing a Mirror	No
12.8 Removing a Raid-5 Log	No
12.9 Extending a Volume	No
12.10 Shrinking a Volume	No
12.11 Preparing a Volume for Online Backup	No
Chapter 13 - Menu Interface Operations
13.3.4 Adding a VM Disk to the Hot-Relocation Pool	No
13.3.5 Removing a VM Disk From the Hot-Relocation Pool	No
13.4.1 Mirroring Volumes on a VM Disk	Yes
13.4.2 Moving Volumes from a VM Disk	No

Volume Manager (VxVM) User's Guide

Topics
(none of these apply to the basic VxVM product)

Applies to
Mirroring product?

Chapter 2 - Performance Monitoring No

2.2 Performance Guidelines No

2.2.1 Data Assignment No

2.2.2 Striping No

2.2.3 Mirroring Yes

2.2.4 Mirroring and Striping No

2.2.5 Using RAID-5 No

2.2.6 Hot-Relocation No

2.3.1 Performance Priorities No

2.3.3 Using Performance Data No

2.4.2.8 voldrl_min_regionsz No

2.4.2.9 voldrl_max_drtregs No

Chapter 3 - Disk and Disk Group Administration

3.6 Detecting and Replacing Failed Disks No

Chapter 4 - Volume Administration

4.4.4.1 Associating Log Subdisks No

4.5.1 Creating Plexes Yes

4.5.2 Backup Using Mirroring Yes

4.5.4 Dissociating and Removing Plexes Yes

4.5.8 Moving Plexes No

4.5.9 Copying Plexes No

4.6.7.3 Changing Volume Read Policy Yes

4.6.9 Mirroring Existing Volumes Yes

4.7 RAID-5 Volume Operations No

4.8 Performing Online Backup No

Appendix A - Volume Manager Error Messages

A.3.2.37 System boot disk does not have a valid root plex No

A.3.4.2 Detached log for volume No

A.3.4.7 Volume entering degraded mode Yes

A.4.1.3 Read error on object of mirror in volume corrected Yes

A.4.2.4 Read error on mirror of volume Yes

A.4.2.5 Write error on mirror of volume offset length Yes

A.4.2.7 Overlapping mirror detached from volume No

A.4.2.10 Detaching RAID-5 volume No

A.4.2.11 Object detached from RAID-5 volume No

A.4.2.12 RAID-5 volume entering degraded mode operation No

A.4.2.13 Double failure condition detected on RAID-5 volume No

A.4.2.14 Failure in RAID-5 logging operation No

Appendix B - Recovery No

Topics (none of these apply to the basic VxVM product)	Applies to Mirroring product?
Chapter 2 - Performance Monitoring	No
2.2 Performance Guidelines	No
2.2.1 Data Assignment	No
2.2.2 Striping	No
2.2.3 Mirroring	Yes
2.2.4 Mirroring and Striping	No
2.2.5 Using RAID-5	No
2.2.6 Hot-Relocation	No
2.3.1 Performance Priorities	No
2.3.3 Using Performance Data	No
2.4.2.8 voldrl_min_regionsz	No
2.4.2.9 voldrl_max_drtregs	No
Chapter 3 - Disk and Disk Group Administration
3.6 Detecting and Replacing Failed Disks	No
Chapter 4 - Volume Administration
4.4.4.1 Associating Log Subdisks	No
4.5.1 Creating Plexes	Yes
4.5.2 Backup Using Mirroring	Yes
4.5.4 Dissociating and Removing Plexes	Yes
4.5.8 Moving Plexes	No
4.5.9 Copying Plexes	No
4.6.7.3 Changing Volume Read Policy	Yes
4.6.9 Mirroring Existing Volumes	Yes
4.7 RAID-5 Volume Operations	No
4.8 Performing Online Backup	No
Appendix A - Volume Manager Error Messages
A.3.2.37 System boot disk does not have a valid root plex	No
A.3.4.2 Detached log for volume	No
A.3.4.7 Volume entering degraded mode	Yes
A.4.1.3 Read error on object of mirror in volume corrected	Yes
A.4.2.4 Read error on mirror of volume	Yes
A.4.2.5 Write error on mirror of volume offset length	Yes
A.4.2.7 Overlapping mirror detached from volume	No
A.4.2.10 Detaching RAID-5 volume	No
A.4.2.11 Object detached from RAID-5 volume	No
A.4.2.12 RAID-5 volume entering degraded mode operation	No
A.4.2.13 Double failure condition detected on RAID-5 volume	No
A.4.2.14 Failure in RAID-5 logging operation	No
Appendix B - Recovery	No

Software notes and recommendations

The following notes and recommendations apply to the Online Data Manager Volume Manager.

Volume Manager (VxVM)

The following notes and recommendations apply to the Volume Manager.

Disks with insufficient space (less than 1024 disk blocks) for the allocation of an on-disk database copy cannot be encapsulated. The database requires at least the same space as is allocated for other disks in the same disk group. This size defaults to 1024 blocks. A way to work around this is to relocate the data on the last partition of the disk to a volume on a different disk, and free the space by reducing the partition size to 0. The space for this database must be allocated from the beginning or the end of the disk, with the exception of the root disk. The root disk can be encapsulated by carving out space from the swap slice if there is no space at the beginning or at the end of the disk. This is done by creating a subdisk for the private partition in the space obtained from the swap slice. There is no workaround to the problem of not having space on a disk to store private VxVM information. VxVM requires at least a small region of private storage (1024 blocks) for proper disk identification.
The Volume Manager tracks disks using long unique identifiers that VxVM stores on each disk. VxVM expects each disk to have a different unique identifier, and does not effectively guard against the situation where two disks have the same unique identifier. Duplicate identifiers should only occur as the result of the administrator using dd or some other utility to perform physical copies of the contents of an entire disk. The only effective workaround is that the administrator should not do exact physical disk copying.
It should be possible to prevent any access of a disk by VxVM. For example, startup of VxVM could be severely impacted by a disk with errors that result in I/Os that take a long time to fail. However, when VxVM starts up, it accesses every disk on the system, by reading its VTOC and possibly a few blocks from one partition. There is currently no mechanism to prevent this. A disk can be off-lined persistently, but the offline state is only recognized after the probe of all disks.
RAID-5 cannot currently be mirrored or backed up online using any mechanisms provided in VxVM.
The disk hot-sparing mechanism currently used in VxVM does not work well for partial disk failures. The model was written using the presumption that disks fail totally, rather than partially, and that partial errors can usually be fixed by writing back a failing sector (which is done automatically when using mirroring or RAID-5). Usually this is a good assumption, but some users have encountered situations where only a few sectors failed and hot sparing did not occur.
On machines with low memory (32 megabytes or less), under heavy I/O stress conditions against high memory usage volumes (that is, RAID-5 volumes), a situation occurred in which the system could not allocate physical memory pages any more. For example, such a situation may result when exercising heavy I/O stress against RAID-5 volumes for 24 hours on a 32-megabyte machine.
the OS does not support raw interfaces for devices greater than 2 gigabytes.
VxVM assumes there are not more than 16 partitions defined on any disk VTOC. It is assumed that the slices are s0 through s15.
Note that there is no protection built into vxassist to prevent the user from shrinking the swap volume without first shrinking what the system sees as available swap space. If it is necessary to shrink the swap volume, this operation should be done in single-user mode and the system should be rebooted immediately. Failing to take these precautions could result in unknown system behavior or lock-up.
Using vxdg free with a non-existent disk-media-name does not print an appropriate error message. It simply prints a header.
If you are mirroring your swap device, we recommend that you set aside a dedicated dump slice to store possible system dumps. Otherwise, the system may not detect the presence of a system dump on the swap slice.

NOTE: At least one disk must remain in the root disk group (rootdg) while VxVM is running.

Visual Administrator (VxVA)

The following notes and recommendations apply to the Volume Manager Visual Administrator.

Spurious commands will be created if a plex is associated with a volume in another vxva session or from the command line. These commands are innocuous.
Any volumes can have the LOGDRL set as the logging type, whether or not there are any logs currently attached. The properties form indicates the state of this flag, which means that the volume can accept a DRL.
vxva popup warning windows appear relative to the root window. In a virtual desktop environment, this can be a problem if the vxva session has windows on multiple virtual screens. It is recommended that users use only one virtual screen per vxva session.
Hidden warning popup messages may need to be dismissed before a session in multiple virtual screens can continue.
The volume properties form may be used to rename a volume. If this is done when a filesystem that resides on that volume is mounted, then the mount point will no longer be displayed in the view below the volume. This is due to inconsistencies in the maintenance of the /etc/vfstab file by the Visual Administrator.

Enabling DMAPI access to VXFS

Data Management API (DMAPI) access to the kernel is not enabled by default. Do the following to enable DMAPI kernel access.

NOTE: Besides enabling DMAPI, you must also license it before you can use it.

VXFS must be configured into the system, otherwise the system will not build with DMAPI.

Edit /etc/conf/sdevice.d/vxportal to change the second column of the last line from a N to a Y. Change this line:

   vxportal    N   0   0   0   0   0   0   0   0   -1

to this:

   vxportal    Y   0   0   0   0   0   0   0   0   -1

Rebuild the system:

/etc/conf/bin/idbuild -B
Reboot the system.

Pre-initialization instructions

Consider these items in preparation for Volume Manager initialization:

Most of the commands used in the initialization are in the /sbin, /usr/sbin, /usr/bin/X11, or /opt/vxvm-va/bin directory. You should add these directories to your PATH environment variable. For example, using the Bourne or Korn Shell, modify your .profile with:

PATH=/sbin:/usr/sbin:$PATH:/usr/bin/X11:/opt/vxva/bin export PATH
Before initializing the Volume Manager, make sure you are aware of the contents of each disk. During the initialization process, you will have to decide the following:
- Do you wish to place all or just some of the disks on the system and/or controller under Volume Manager control?
- Do you wish to place the system's root disk under Volume Manager control? (This is necessary in the case of a one-disk system.)
- When you place disks under Volume Manager control, do you wish to preserve data in any or all existing filesystems and partitions through encapsulation?
- For each disk, do you want to choose whether to encapsulate, initialize, or leave the disk alone (Custom Installation)? Otherwise, do you want to either encapsulate or initialize all disks on a controller together (Quick Installation)?
  NOTE: The software initialization may require repeated reboots. You are prompted for permission to reboot when the reboots are needed. If you choose not to reboot, the initialization cannot complete. After the first reboot, the initialization process takes control of the system until the initialization is complete.
Any disk to be managed by Volume Manager should have two free partitions and a small amount of free space (1024 sectors). The free space should be at the beginning or end of the disk and should not belong to a partition. This space is used for storing disk group configurations and a disk label that ensures that the Volume Manager can identify the disk, even if it is moved to a different address or controller. It is also used to keep track of disk configuration and to ensure correct recovery. The Volume Manager allocates approximately 1024 sectors (512K) from each disk for the disk group configurations and the disk label. This space is sufficient to handle normal disk group configurations for up to approximately 20 disks.
The boot (root) disk is a special case; if no other space is available, the Volume Manager will attempt to allocate space usually reserved for swap by shrinking the swap slice. This process is known as swap relocation and, if necessary, will happen automatically during root disk encapsulation.
NOTE: Although it is possible to put disks with no free space under Volume Manager control, this should only be used as a migration strategy because many VxVM capabilities based on disk identity are available only for disks with the required free space. See vxdisk(1M) for information about the nopriv disk type.
If you plan to encapsulate any disks, make sure that those disks have the required minimum free space (1024 sectors).

Initializing the Volume Manager

The following describes how to initialize the Volume Manager using the vxinstall program. The vxinstall program places all specified disks under Volume Manager control. These disks are automatically placed in the default disk group, rootdg. If you wish to place some of your disks in another disk group, you must use vxinstall to initialize at least one disk into rootdg and then use vxdiskadm(1M) to initialize or encapsulate additional disks into other disk groups. Refer to the VERITAS Volume Manager User's Guide for information on how to create other disk groups in which to place your disks.

NOTE: At least one disk must remain in rootdg at all times while VxVM is running.

Disks that are not initially placed under Volume Manager control by vxinstall can be added later using a Volume Manager interface (such as vxdiskadm). However, you must add at least one disk to Volume Manager control with vxinstall before you can add disks using other utilities or interfaces.

Once you have completed the package install, initialize the Volume Manager as follows:

Log in as root.
If you want to exclude one or more disks from Volume Manager control, add the names of those disks to the /etc/vx/disks.exclude file.
vxinstall ignores any disks listed in this file and leaves them untouched. The following is an example of the contents of the disks.exclude file (which you have to create):
```
   c0b0t0d0
   c0b0t1d0
```
Similarly, you can exclude all disks on an entire controller by adding the name of that controller to the /etc/vx/cntrls.exclude file. The following is an example of the contents of the cntrls.exclude file:
```
   c0
   c1
```

Enter the following to start vxinstall:

   vxinstall

The vxinstall program then examines all controllers it finds attached to the system and then lists them:

Generating list of attached controllers.... Volume Manager Installation Menu: VolumeManager/Install The Volume Manager names disks on your system using the controller and disk number of the disk, substituting them into the following pattern: c<controller>b<bus>t<target>d<disk>

Some examples would be: c0b0t0d0 - first controller, bus 0, first target, first disk c1b0t0d0 - second controller, bus 0, first target, first disk c1b0t1d0 - second controller, bus 0, second target, first disk

The Volume Manager has detected the following controllers on your system:

c0: c1: c2:

Hit RETURN to continue.

Next, vxinstall displays a brief introduction to the initialization process:

Volume Manager Installation Menu: VolumeManager/Install You will now be asked if you wish to use Quick Installation or Custom Installation. Custom Installation allows you to select how the Volume Manager will handle the installation of each disk attached to your system. Quick Installation examines each disk attached to your system and attempts to create volumes to cover all disk partitions that might be used for file systems or for other similar purposes. If you do not wish to use some disks with the Volume Manager, or if you wish to reinitialize some disks, use the Custom Installation option. Otherwise, we suggest that you use the Quick Installation option.

Hit RETURN to continue.

vxinstall then displays a menu with the following options

1 Quick Installation 2 Custom Installation

? Display help about menu ?? Display help about menuing system q Exit from menus

Select an operation to perform:

The options behave as follows:

Quick Installation uses default options for installation.
Custom Installation allows you to control whether and how each disk will be brought under Volume Manager control.
? displays a help file describing the current operation or menu choices.
?? displays general information about using vxinstall.
q exits from the current operation, or from the vxinstall program.

Select either 1 Quick Installation or 2 Custom Installation to proceed with the initialization.
NOTE: Quick Installation will bring every disk attached to your system under Volume Manager control unless you have chosen to exclude some disks (through the disks.exclude or .cntrls.exclude file, described earlier). If you have any disks on your system that you do not want placed under Volume Manager control, either list them in the disks.exclude file, or choose the Custom Installation option (which allows you to leave certain disks alone).
- If you chose Quick Installation, go to ``Quick installation''.
- If you chose Custom Installation, go to ``Custom installation''.

Quick installation

The Quick Installation option is the easiest to use. For each controller, Quick Installation gives you the option of either initializing or encapsulating all disks. If you want to encapsulate some disks on a given controller and initialize others, use Custom Installation instead.

Quick installation allows you to initialize disks on a controller, even if valid partitions are found on those disks. However, you should avoid selecting the initialize option for any disks containing valid data.

With encapsulation, the /etc/vfstab and /etc/swaptab files are updated to ensure that filesystems previously mounted on disk partitions will be mounted on volumes instead.

To use the Quick Installation option:

Select menu item 1 Quick Installation from the vxinstall main menu.
vxinstall asks you a series of questions for each controller and disk it finds connected to your system. When a default response is displayed in parentheses, press <Enter> to accept that default. At any of the vxinstall prompts, you can select q to completely abandon the initialization and start over.
All disks are encapsulated or initialized (according to your instructions) at the end of the vxinstall procedure. If you quit vxinstall before it enters its final phase of actually initializing or encapsulating the disks, all disks will be left as they were before vxinstall started.
At the following prompt, indicate whether to encapsulate your boot disk (this example assumes your boot disk is named c0b0t0d0):
```
   The c0b0t0d0 disk is your Boot Disk. You can not add
   it as a new disk. If you encapsulate it, you will make
   your root filesystem and other system areas on the
   Boot Disk into volumes. This is required if you wish
   to mirror your root file system or system swap area.
   
   Encapsulate Boot Disk [y,n,q,?] (default: n)
```
- If you enter n, your root and other filesystems will be left unaltered.
- If you enter y, vxinstall asks you to enter a disk name for the boot disk and press <Enter> to continue. vxinstall encapsulates your root filesystem as a volume, along with your swap device and all other disk partitions found on your boot disk. /home, /var, and any other filesystems on your boot disk are also encapsulated.
NOTE: If you are mirroring your swap device, we recommend that you set aside a dedicated dump slice to store possible system dumps. Otherwise, the system may not detect the presence of a system dump on the swap slice.

vxinstall now goes through each controller and asks you how to handle the disks contained on that controller. vxinstall begins this process for each controller by identifying the controller and generating a list of its disks:
```
   Generating list of attached disks on c1....
   The Volume Manager has detected the following disks
   on controller c1:
   c1b0t0d0 c1b0t1d0 c1b0t2d0 c1b0t3d0 c1b0t4d0
   
   Hit RETURN to continue.
```
If any disks are listed in the /etc/vx/disks.exclude file, they are listed here separately as excluded disks.

At the following prompt, indicate whether you want to initialize all the disks on the controller.

   Volume Manager Quick Installation For Controller c1 Menu:
   VolumeManager/Install/QuickInstall/c1
   
   Initialize all disks on this controller?
   (destroys data on these disks) [y,n,q,?] (default: n)

If you enter n, vxinstall assumes that you want to encapsulate all disks on the controller instead and displays the following:

Volume Manager will now try to encapsulate all the disks on this controller. Disks not having valid partitions will be initialized.

Hit RETURN to continue.

If you enter y, vxinstall prompts you for confirmation and warns you that this will destroy any existing data on the disks.
```
   Are you sure (destroys data on these disks)
   [y,n,q,?] (default: n)
```
NOTE: Be careful when using this option as it can result in data loss.

Enter y to continue with the disk initialization. The following message appears:
```
   Volume Manager will now initialize all the disks
   on this controller including disks having valid
   partitions.
   
   Hit RETURN to continue.
```

At the following prompt, indicate whether you want default disk names to be assigned to all of the disks on the controller.
```
   Use default disk names for these disks? [y,n,q,?] (default: y)
```
If you enter y, vxinstall automatically assigns and lists default disk names for each disk:
```
   The c1b0t0d0 disk will be given disk name disk01
   The c1b0t1d0 disk will be given disk name disk02
   The c1b0t2d0 disk will be given disk name disk03
   The c1b0t3d0 disk will be given disk name disk04
   
   Hit RETURN to continue.
```
If encapsulation was selected, this list also includes the following message for each disk that can be encapsulated
```
   The c1b0t0d0 disk has been configured for encapsulation.
```
or the following message for empty disks that will be initialized instead of encapsulated:
```
   The c3b0t2d0 disk appears to be empty.
   Adding as a new disk.
```
If you enter n, vxinstall prompts you for a disk name for each disk on the controller individually:
```
   Enter disk name for c1b0t0d0 [<name>,q,?] (default: disk01)
```
For each disk, respond with the desired disk name and press <Enter>.
When all of the disks on the current controller have been named, press <Enter> and repeat Steps 3 and 4 for the remaining controllers on your system.
When you have completed the vxinstall procedure for all controllers on your system, vxinstall displays a summary of the disks you have designated for initialization (New Disk) or encapsulation (Encapsulate) on each controller:
```
   The following is a summary of your choices.
   c0b0t5d0 New Disk
   c1b0t0d0 New Disk
   c1b0t1d0 New Disk
   c1b0t2d0 New Disk
   c2b0t2d3 New Disk
   c3b0t0d0 Encapsulate
   c3b0t1d0 Encapsulate
   c3b0t2d0 New Disk
   Is this correct [y,n,q,?] (default: y)
```
This is your last chance to review and alter your choices about how to handle any of the disks to be placed under Volume Manager control.
- If you enter y, vxinstall will encapsulate all disks listed with Encapsulate and initialize (destroying any existing data) all disks listed with New Disk.
- If you enter n, vxinstall prompts you for the name of a disk to be removed from the list and excluded from Volume Manager control:
```
   Enter disk to be removed from your choices.
   
   Hit return when done.
   

   [<name>,q,?]
```
  Enter the name of the disk to be removed from the list and press <Enter>. Repeat this for each disk to be removed. vxinstall should now display an updated summary without the disks chosen for removal. If you are now satisfied with the summary, enter y.
  After running vxinstall, you should shut down and reboot your machine immediately or at the earliest opportunity. Refer to ``Rebooting after vxinstall''.
  NOTE: Do not make any changes to your disk or filesystem configurations before shutting down and rebooting your system.
  
  If a shutdown is not required, vxinstall indicates successful completion of the installation:
```
   Installation is complete.
```

Custom installation

Custom Installation allows you to control which disks are added to Volume Manager control and how they are added. This is important if you do not want all your disks under Volume Manager control. Custom Installation also gives you the option of initializing all disks on a controller, encapsulating all disks on a controller, or initializing some disks on a controller and encapsulating others.

To use the Custom Installation option:

Select menu item 2 Custom Installation from the vxinstall main menu.
vxinstall asks a series of questions for each controller and disk it finds connected to your system. When a default response is displayed in parentheses, press <Enter> to accept that default. At any of the vxinstall prompts, select q to completely abandon the initialization and start over.
All disks are encapsulated or initialized (according to your instructions) at the end of the vxinstall procedure. If you quit vxinstall before it enters its final phase of actually initializing or encapsulating the disks, all disks will be left as they were before vxinstall started.
At the following prompt, indicate whether to encapsulate your boot disk (this example assumes your boot disk is named c0b0t0d0):
```
   The c0b0t0d0 disk is your Boot Disk. You can not
   add it as a new disk. If you encapsulate it, you
   will make your root filesystem and other system
   areas on the Boot Disk into volumes. This is
   required if you wish to mirror your root file
   system or system swap area.
   Encapsulate Boot Disk [y,n,q,?] (default: n)
```
- If you enter n, your root and other filesystems will be left unaltered.
- If you enter y, vxinstall encapsulates your root filesystem as a volume, along with your swap device and all other disk partitions found on your boot disk. /home, /var, and any other filesystems on your boot disk are also encapsulated.
NOTE: If you are mirroring your swap device, we recommend that you set aside a dedicated dump slice to store possible system dumps. Otherwise, the system may not detect the presence of a system dump on the swap slice.

vxinstall now goes through each controller and asks you how to handle the disks contained on that controller. vxinstall begins this process for each controller by identifying the controller and generating a list of its disks:
```
   Generating list of attached disks on c1....
   The Volume Manager has detected the following disks
   on controller c1:
   c1b0t0d0 c1b0t1d0 c1b0t2d0 c1b0t3d0 c1b0t0d4
   
   Hit RETURN to continue.
```
If any disks are listed in the /etc/vx/disks.exclude file, they are listed here separately as excluded disks.

At the following prompt, select the option that matches how you want the disks on this controller to be handled:

Installation options for controller c1 Menu: VolumeManager/Install/Custom/c1

1 Install all disks as pre-existing disks. (encapsulate) 2 Install all disks as new disks. (discards data on disks!) 3 Install one disk at a time. 4 Leave these disks alone.

? Display help about menu ?? Display help about the menuing system q Exit from menus

Select an operation to perform:

To encapsulate all disks on the controller, select 1.
Volumes will be created to encapsulate any partitions on all disks on this controller. The /etc/vfstab file will be updated to ensure that file systems previously mounted on disk partitions will be mounted as volumes instead.
At the following prompt, indicate whether you want default disk names to be assigned to all of the disks on the controller
```
   Use default disk names for these disks? [y,n,q,?] (default: y)
```
- If you enter y, vxinstall automatically assigns and lists default disk names for each disk:
```
   The c1b0t0d0 disk will be given disk name disk01
   The c1b0t0d0 disk has been configured for encapsulation.
   The c1b0t1d0 disk will be given disk name disk02
   The c1b0t1d0 disk has been configured for encapsulation.
   
   Hit RETURN to continue.
```
  If any disks are found to be empty, the following message indicates that they will be initialized instead of encapsulated:
```
   The c3b0t2d0 disk appears to be empty.
   Adding as a new disk.
```
If you enter n, vxinstall prompts you for a disk name for each disk on the controller individually:
```
   Enter disk name for c1b0t0d0 [<name>,q,?] (default: disk01)
```
For each disk, respond with the desired disk name and press <Enter>.
When all of the disks on the current controller have been named, press <Enter> to move on to the next controller.
To initialize all disks on the controller, select 2.
NOTE: Be careful when using this option as it can result in data loss.

All disks on this controller will be reinitialized. This destroys all data and partitions on each disk and makes the disk available as free space for allocating new volumes, or mirrors of existing volumes. If you are running vxinstall on new disks that have never been placed under Volume Manager control before, choose this option so that the disks will be repartitioned according to the Volume Manager's specifications.
At the following prompt, indicate whether you want default disk names to be assigned to all of the disks on the controller
```
   Use default disk names for these disks? [y,n,q,?] (default: y)
```
- If you enter y, vxinstall automatically assigns and lists default disk names for each disk:
```
   The c1b0t0d0 disk will be given disk name disk01
   The c1b0t1d0 disk will be given disk name disk02
   
   Hit RETURN to continue.
```
- If you enter n, vxinstall prompts you for a disk name for each disk on the controller:
```
   Enter disk name for c1b0t0d0 [<name>,q,?] (default: disk01)
```
  For each disk, respond with the desired disk name and press <Enter>.
  When all of the disks on the current controller have been named, press <Enter> to move on to the next controller.
To install one disk at a time, select 3.
Each disk will be handled separately and you will be prompted for information on a per-disk basis. This allows you to install a disk as a pre-existing disk, install it as a new disk, or leave it alone.
At the following prompt, select the option that matches how you want this disk to be handled:
```
   Installation options for disk c1b0t0d0
   Menu: VolumeManager/Install/Custom/c1/c1b0t0d0
   
   1 Install as a pre-existing disk. (encapsulate)
   2 Install as a new disk.(discards data on disks!)
   3 Leave this disk alone.
   

   ? Display help about menu
   ?? Display help about the menuing system
   q Exit from menus
   Select an operation to perform:
```
These options are similar to those in the Custom Installation main menu. Once you have selected how to handle this disk, follow the instructions provided for that option in Step 3.
Press <Enter> to move on to the next disk. Repeat the previous step for the remaining disks on the current controller.
When all of the disks on the current controller have been named, press <Enter> to move on to the next controller.
To leave all disks on the controller unaltered, select 4.
No changes will be made to the disks and they will not be placed under Volume Manager control. If applications are currently using these disks and you do not want to upgrade these applications to use the Volume Manager, you can use this option to ensure that your applications can continue to use the disks without modification.
Repeat Step 3 for each of the remaining controllers on your system.
When you have completed the vxinstall procedure for all controllers on your system, vxinstall displays a summary of the disks you have designated for initialization (New Disk) or encapsulation (Encapsulate) on each controller:
```
   The following is a summary of your choices.
   	c0b0t5d0 			New Disk
   	c1b0t0d0 			New Disk
   	c1b0t1d0 			New Disk
   	c1b0t2d0 			New Disk
   	c2b0t2d3 			New Disk
   	c3b0t0d0 			Encapsulate
   	c3b0t1d0 			Encapsulate
   	c3b0t2d0 			New Disk
   
   Is this correct [y,n,q,?] (default: y)
```
This is your last chance to review and alter your choices about how to handle any of the disks to be placed under Volume Manager control.
- If you enter y, vxinstall encapsulates all disks listed with Encapsulate and initialize (destroying any existing data) all disks listed with New Disk.
- If you enter n, vxinstall prompts you for the name of a disk to be removed from the list and excluded from Volume Manager control.
```
   Enter disk to be removed from your choices.
   Hit return when done.
   [<name>,q,?]
```
  Enter the name of the disk to be removed from the list and press <Enter>. Repeat for each disk to be removed and then press <Enter> once more. vxinstall should now display an updated summary without the disks chosen for removal. If you are now satisfied with the summary, enter y.
After running vxinstall, you should shut down and reboot your machine immediately or at the earliest opportunity. Refer to ``Rebooting after vxinstall''.
NOTE: Do not make any changes to your disk or filesystem configurations before shutting down and rebooting your system.
If a shutdown is not required, vxinstall indicates successful completion of the installation.
```
   Installation is complete.
```

Rebooting after vxinstall

After you have specified how vxinstall should process all of the disks attached to your system, you must reboot the system to make changes to your disk partitioning that cannot be made while your disks are in use.

Several reboots may be required, depending on the setup you chose. When it is necessary to shut down and reboot your system, vxinstall displays a message:

   The system now must be shut down and rebooted in order
   to continue the reconfiguration.
   
   Shutdown and reboot now [y,n,q,?] (default: n)

If this happens, respond in either of the following ways:

Enter y to begin an immediate shutdown.
Enter n to cause vxinstall to exit without starting a shutdown. Remember to shutdown and reboot as soon as possible.

NOTE: During the next one or more reboots, you may be asked several times if you wish to continue an operation. Press the <Enter> key at all of these prompts to accept the default answer. If you select a different answer from the default for any of these prompts, or press q, the initialization may fail.

Checking Volume Manager processes

After the Volume Manager has been successfully installed and you have rebooted your system, you can use the ps command to confirm that key Volume Manager processes vxconfigd(1M), vxnotify(1M), and vxsparecheck(1M) are running:

ps -ef | grep vx

See ps(1).

Starting up VxVA

You can invoke the Volume Manager Visual Administrator either from the SCOadmin VERITAS Volume Manager or from a ``terminal'' window. For information on how to use the Volume Manager Visual Administrator, refer to the VERITAS Volume Manager User's Guide.

NOTE: VxVA will only work in demo mode until vxinstall has been run..

Starting up from the command line

You can start VxVA from a terminal window.

To run VxVA, your PATH environment variable must contain a path to locate vxva and va_watch. If your path does not include /usr/X/bin, change your environment variable to include it.

You can verify your path by typing:

echo $PATH

If your path is incorrect, you must change it.

If you are using a Bourne or Korn shell, type

PATH=${PATH}:/usr/X/bin
If you are using a C Shell (csh or tcsh), type

setenv PATH ${PATH}:/usr/X/bin

Once your PATH is set up correctly, you can start VxVA from a terminal window.

NOTE: You can start VxVA in demo mode or real mode. Demo mode requires no special privileges, but you must have appropriate privileges to run VxVA in real mode.

To start the Volume Manager Visual Administrator from a terminal window:

Bring up a terminal window.
Type:

vxva
To start the Volume Manager Visual Administrator in demo mode, type:

vxva -t

Before running VxVA in demo mode for the first time, enter

/usr/X/bin/vxva_setup

It is only necessary to run this shell script once. This script creates dummy disks (for the VxVA demo mode) and a rootdg disk group for you. This script can, however, be used at any time to clean up the existing configuration and return to the default demo.

On small displays (such as those with a graphical resolution of 640 x 480), some forms may be too long to fit entirely on the screen and the buttons on the bottom of the form may not be visible. If this is the case, the window manager "move" function (<Alt>+<F7>, by default) can be used to move the view or form so all fields and buttons are visible.

Another technique that may allow these forms to fit better on a small screen is to start up VxVA by typing:

vxva -xrm "vxva*propertiesForm*marginHeight: 1"

This resource specification will cause forms to appear shorter than normal.

The system owner or root can use the next command to give the user the privilege to run VxVA.

/opt/vxvm-va/install/SetVAPrivilege user-id

where user-id is the user's login name.

Performing Volume Manager operations

Most Volume Manager operations can be performed using either the Visual Administrator graphical user interface or the Volume Manager command line interface. Some disk operations can be performed using the Volume Manager Support Operations menu interface vxdiskadm(1M). For details on how to use these interfaces, refer to the VERITAS Volume Manager User's Guide. Common operations include:

creating, resizing, or removing volumes
adding mirrors to volumes
creating, mounting, unmounting, or resizing filesystems
adding, removing, or replacing disks
creating new disk groups
adding disks to disk groups

Removing Volumes

This procedure removes all Volume Manager volumes.

The following preparations must be completed before you remove volumes.

All filesystems and other applications that use volume devices must be modified to use the underlying disks.
Any filesystems that have been created since the Volume Manager was initialized must be made accessible through a disk partition.
All data from any volumes that were created from multiple regions of storage, including striped or spanned volumes, must be moved onto a single partition.
Any mirror copies of filesystems and other databases should be removed so that only one copy that starts on a partition boundary remains.
If any volume contains more than one plex (mirror), all but one plex should be removed.
To display a list of all volumes, use the command
```
   vxprint -Ath
```
To remove a plex, use the command:
```
   vxplex -o rm dis plex-name
```
See vxplex(1M).

NOTE: Leave exactly one plex for the special volumes rootvol, swapvol, and standvol. Make the remaining plex for each volume consist of one subdisk on physical disk c0b0t0d0s7. These subdisks must be either the original subdisks created by running vxinstall or they must have been created using vxrootmir(1M) or the vxdiskadd(1M) mirror disk operation.
If any remaining plex has more than one subdisk, consolidate those subdisks into a single subdisk. A simple way to do this is to use the commands

vxplex -o rm dis plex-name

NOTE: This operation will not work if you do not have sufficient space on another disk.
Modify /etc/vfstab to mount all filesystems using a partition instead of a volume.
Create partitions for each volume using the prtvtoc(1M) and edvtoc(1M) utilities to update the Volume Table of Contents (VTOC) on each drive. The entry for /stand in the /etc/vfstab file is updated automatically when you invoke the command /etc/vx/bin/vxunroot.
NOTE: Due to the potential for making mistakes while removing volumes, it is recommended that you perform a full system backup before proceeding.

Removal procedure

To remove volumes:

Convert the volumes for the root, swap, and /stand filesystem devices back to using disk partitions. Once all mirror copies of root have been removed, reconfigure the kernel by running the command
```
   /etc/vx/bin/vxunroot
```
This removes all changes that were made to the system to support booting from a root volume. This also updates the /etc/vfstab file to replace the entry for /stand, if you have not done so already. See vxunroot(1M).
Reboot the system.
After rebooting, there should no longer be any volumes in use. To check for volumes that are still in use:
```
   vxprint -Aht -e v_open
```
See vxprint(1M).
If any volumes are listed, determine what application or filesystem is using each volume. One way you can determine which applications reference a listed volume is to rename the volume, so that the application cannot find the volume under its original name. To rename a volume, enter
```
   vxedit rename old_vol_name  new_vol_name
```
See vxedit(1M).
If you have to rename any volumes, reboot again to ensure that the volumes are no longer in use.
After rebooting, use vxprint once more to ensure that no volumes are in use.
Applications that reference the volume will print error messages to the console.

Shutting down the Volume Manager

To shut down the Volume Manager, enter:

   vxdctl stop
   vxiod -f set 0

See vxdctl(1M) and vxiod(1M).