When laying out file systems, remember that hard drives transfer data faster from the outer tracks to the inner. Thus, smaller and heavier-accessed file systems should be closer to the outside of the drive, while larger partitions like /usr should be placed toward the inner parts of the disk. It is a good idea to create partitions in an order similar to: /, swap, /var, and /usr.

The size of the /var partition reflects the intended machine's usage. This partition is used to hold mailboxes, log files, and printer spools. Mailboxes and log files can grow to unexpected sizes depending on the number of users and how long log files are kept. On average, most users rarely need more than about a gigabyte of free disk space in /var.

The /usr partition holds many of the files which support the system, including the FreeBSD Ports Collection and system source code. At least 2 gigabytes is recommended for this partition.

When selecting partition sizes, keep the space requirements in mind. Running out of space in one partition while barely using another can be a hassle.

As a rule of thumb, the swap partition should be about double the size of physical memory (RAM). Systems with minimal RAM may perform better with more swap. Configuring too little swap can lead to inefficiencies in the VM page scanning code and might create issues later if more memory is added.

On larger systems with multiple SCSI disks or multiple IDE disks operating on different controllers, it is recommended that swap be configured on each drive, up to four drives. The swap partitions should be approximately the same size. The kernel can handle arbitrary sizes but internal data structures scale to 4 times the largest swap partition. Keeping the swap partitions near the same size will allow the kernel to optimally stripe swap space across disks. Large swap sizes are fine, even if swap is not used much. It might be easier to recover from a runaway program before being forced to reboot.

By properly partitioning a system, fragmentation introduced in the smaller write heavy partitions will not bleed over into the mostly read partitions. Keeping the write loaded partitions closer to the disk's edge will increase I/O performance in the partitions where it occurs the most. While I/O performance in the larger partitions may be needed, shifting them more toward the edge of the disk will not lead to a significant performance improvement over moving /var to the edge.

When this method is selected, a menu will display the available disk(s). If multiple disks are connected, choose the one where FreeBSD is to be installed.

Figure 2.13. Selecting from Multiple Disks

Once the disk is selected, the next menu prompts to install to either the entire disk or to create a partition using free space. If [ Entire Disk ] is chosen, a general partition layout filling the whole disk is automatically created. Selecting [ Partition ] creates a partition layout from the unused space on the disk.

Figure 2.14. Selecting Entire Disk or Partition

After the partition layout has been created, review it to ensure it meets the needs of the installation. Selecting [ Revert ] will reset the partitions to their original values and pressing [ Auto ] will recreate the automatic FreeBSD partitions. Partitions can also be manually created, modified, or deleted. When the partitioning is correct, select [ Finish ] to continue with the installation.

Figure 2.15. Review Created Partitions

Highlight the installation drive (ada0 in this example) and select [ Create ] to display a menu of available partition schemes:

Figure 2.17. Manually Create Partitions

GPT is usually the most appropriate choice for amd64 computers. Older computers that are not compatible with GPT should use MBR. The other partition schemes are generally used for uncommon or older computers.

After the partitioning scheme has been selected and created, select [ Create ] again to create the partitions.

Figure 2.18. Manually Create Partitions

A standard FreeBSD GPT installation uses at least three partitions:

Another partition type worth noting is freebsd-zfs, used for partitions that will contain a FreeBSD ZFS file system (Chapter 20, The Z File System (ZFS)). Refer to gpart(8) for descriptions of the available GPT partition types.

Multiple file system partitions can be created and some people prefer a traditional layout with separate partitions for /, /var, /tmp, and /usr. See Example 2.1, “Creating Traditional Split File System Partitions” for an example.

The Size may be entered with common abbreviations: K for kilobytes, M for megabytes, or G for gigabytes.

A Mountpoint is needed if the partition will contain a file system. If only a single UFS partition will be created, the mountpoint should be /.

The Label is a name by which the partition will be known. Drive names or numbers can change if the drive is connected to a different controller or port, but the partition label does not change. Referring to labels instead of drive names and partition numbers in files like /etc/fstab makes the system more tolerant to hardware changes. GPT labels appear in /dev/gpt/ when a disk is attached. Other partitioning schemes have different label capabilities and their labels appear in different directories in /dev/.

After the custom partitions have been created, select [ Finish ] to continue with the installation.

Support for automatic creation of root-on-ZFS installations was added in FreeBSD 10.0-RELEASE. This partitioning mode only works with whole disks and will erase the contents of the entire disk. The installer will automatically create partitions aligned to 4k boundaries and force ZFS to use 4k sectors. This is safe even with 512 byte sector disks, and has the added benefit of ensuring that pools created on 512 byte disks will be able to have 4k sector disks added in the future, either as additional storage space or as replacements for failed disks. The installer can also optionally employ GELI disk encryption as described in Section 18.12.2, “Disk Encryption with geli”. If encryption is enabled, a 2 GB unencrypted boot pool containing the /boot directory is created. It holds the kernel and other files necessary to boot the system. A swap partition of a user selectable size is also created, and all remaining space is used for the ZFS pool.

The main ZFS configuration menu offers a number of options to control the creation of the pool.

Figure 2.19. ZFS Partitioning Menu

Select T to configure the Pool Type and the disk(s) that will constitute the pool. The automatic ZFS installer currently only supports the creation of a single top level vdev, except in stripe mode. To create more complex pools, use the instructions in Section 2.6.5, “Shell Mode Partitioning” to create the pool. The installer supports the creation of various pool types, including stripe (not recommended, no redundancy), mirror (best performance, least usable space), and RAID-Z 1, 2, and 3 (with the capability to withstand the concurrent failure of 1, 2, and 3 disks, respectively). while selecting the pool type, a tooltip is displayed across the bottom of the screen with advice about the number of required disks, and in the case of RAID-Z, the optimal number of disks for each configuration.

Figure 2.20. ZFS Pool Type

Once a Pool Type has been selected, a list of available disks is displayed, and the user is prompted to select one or more disks to make up the pool. The configuration is then validated, to ensure enough disks are selected. If not, select <Change Selection> to return to the list of disks, or <Cancel> to change the pool type.

Figure 2.21. Disk Selection

If one or more disks are missing from the list, or if disks were attached after the installer was started, select - Rescan Devices to repopulate the list of available disks. To ensure that the correct disks are selected, so as not to accidently destroy the wrong disks, the - Disk Info menu can be used to inspect each disk, including its partition table and various other information such as the device model number and serial number, if available.

Figure 2.23. Analysing a Disk

The main ZFS configuration menu also allows the user to enter a pool name, disable forcing 4k sectors, enable or disable encryption, switch between GPT (recommended) and MBR partition table types, and select the amount of swap space. Once all options have been set to the desired values, select the >>> Install option at the top of the menu.

If GELI disk encryption was enabled, the installer will prompt twice for the passphrase to be used to encrypt the disks.

Figure 2.24. Disk Encryption Password

The installer then offers a last chance to cancel before the contents of the selected drives are destroyed to create the ZFS pool.

Figure 2.25. Last Chance

The installation then proceeds normally.

When creating advanced installations, the bsdinstall paritioning menus may not provide the level of flexibility required. Advanced users can select the Shell option from the partitioning menu in order to manually partition the drives, create the file system(s), populate /tmp/bsdinstall_etc/fstab, and mount the file systems under /mnt. Once this is done, type exit to return to bsdinstall and continue the installation.