Managing Your ZFS Swap and Dump Devices

During an initial Oracle Solaris OS installation or after performing a Live Upgrade migration from a UFS file system, a swap area is created on a ZFS volume in the ZFS root pool. For example:

# swap -l
swapfile                  dev  swaplo  blocks   free
/dev/zvol/dsk/rpool/swap 256,1      16 4194288 4194288

During an initial Oracle Solaris OS installation or a Live Upgrade from a UFS file system, a dump device is created on a ZFS volume in the ZFS root pool. In general, a dump device requires no administration because it is set up automatically at installation time. For example:

# dumpadm
      Dump content: kernel pages
       Dump device: /dev/zvol/dsk/rpool/dump (dedicated)
Savecore directory: /var/crash/t2000
  Savecore enabled: yes
   Save compressed: on

If you disable and remove the dump device, then you must enable it with the dumpadm command after it is re-created. In most cases, you will only have to adjust the size of the dump device by using the zfs command.

For information about the swap and dump volume sizes that are created by the installation programs, see Oracle Solaris Installation and Live Upgrade Requirements for ZFS Support.

Both the swap volume size and the dump volume size can be adjusted during and after installation. For more information, see Adjusting the Sizes of Your ZFS Swap Device and Dump Device.

Consider the following issues when working with your ZFS swap and dump devices:

• Separate ZFS volumes must be used for the swap area and the dump device.

• Currently, using a swap file on a ZFS file system is not supported.

• If you need to change your swap area or dump device after the system is installed or upgraded, use the swap and dumpadm commands as in previous releases. For more information, see Chapter 16, Configuring Additional Swap Space (Tasks), in System Administration Guide: Devices and File Systems and Chapter 17, Managing System Crash Information (Tasks), in System Administration Guide: Advanced Administration.

See the following sections for more information:

• Adjusting the Sizes of Your ZFS Swap Device and Dump Device

• Troubleshooting ZFS Dump Device Issues

Adjusting the Sizes of Your ZFS Swap Device and Dump Device

You might need to adjust the size of swap and dump devices after installation or possibly, recreate the swap and dump volumes.

• You can adjust the size of your swap and dump volumes during an initial installation. For more information, see Example 4-1.

• You can create and size your swap and dump volumes before you perform a Live Upgrade operation. For example:

  1. Create your storage pool.

     # zpool create rpool mirror c0t0d0s0 c0t1d0s0

  2. Create your dump device.

     # zfs create -V 2G rpool/dump

  3. Enable the dump device.

     # dumpadm -d /dev/zvol/dsk/rpool/dump
           Dump content: kernel pages
            Dump device: /dev/zvol/dsk/rpool/dump (dedicated)
     Savecore directory: /var/crash/t2000
       Savecore enabled: yes
        Save compressed: on

  4. Create your swap volume:

     # zfs create -V 2G rpool/swap

  5. You must enable the swap area when a new swap device is added or changed.

     # swap -a /dev/zvol/dsk/rpool/swap

  6. Add an entry for the swap volume to the /etc/vfstab file.

  Live Upgrade does not resize existing swap and dump volumes.

• You can reset the volsize property of the dump device after a system is installed. For example:

  # zfs set volsize=2G rpool/dump
  # zfs get volsize rpool/dump
  NAME        PROPERTY  VALUE       SOURCE
  rpool/dump  volsize   2G          -

• If the current swap area is not in use, you can resize the size of the current swap volume, but you must reboot the system to see the increased swap space size.

  # zfs get volsize rpool/swap
  NAME        PROPERTY  VALUE    SOURCE
  rpool/swap  volsize   4G       local
  # zfs set volsize=8g rpool/swap
  # zfs get volsize rpool/swap
  NAME        PROPERTY  VALUE    SOURCE
  rpool/swap  volsize   8G       local
  # init 6

• You can attempt to resize the swap volume, but it might be best to remove the swap device. Then, re-create it. For example:

  # swap -d /dev/zvol/dsk/rpool/swap
  # zfs create -V 2g rpool/swap
  # swap -a /dev/zvol/dsk/rpool/swap

• You can adjust the size of the swap and dump volumes in a JumpStart profile by using profile syntax similar to the following:

  install_type initial_install
  cluster SUNWCXall
  pool rpool 16g 2g 2g c0t0d0s0

  In this profile, two 2g entries set the size of the swap volume and dump volume to 2 GB each.

• If you need more swap space on a system that is already installed, just add another swap volume. For example:

  # zfs create -V 2G rpool/swap2

  Then, activate the new swap volume. For example:

  # swap -a /dev/zvol/dsk/rpool/swap2
  # swap -l
  swapfile                  dev  swaplo   blocks   free
  /dev/zvol/dsk/rpool/swap  256,1      16 1058800 1058800
  /dev/zvol/dsk/rpool/swap2 256,3      16 4194288 4194288

  Finally, add an entry for the second swap volume to the /etc/vfstab file.

Customizing ZFS Swap and Dump Volumes

Keep the following points in the mind if you remove the default swap and dump volumes and recreate them in a non-root (data) pool:

• If you want to create swap and dump devices in a non-root pool, do not create swap and dump volumes in a RAIDZ pool. If a pool includes swap and dump volumes, it must be a one-disk pool or a mirrored pool.

• If you use Live Upgrade to update your system, use the -P option to preserve the dump device from the PBE to ABE. For example:

  # lucreate -n newBE -P

Troubleshooting ZFS Dump Device Issues

Review the following if you have problems either capturing a system crash dump or resizing the dump device.

• If a crash dump was not created automatically, you can use the savecore command to save the crash dump.

• A dump volume is created automatically when you initially install a ZFS root file system or migrate to a ZFS root file system. In most cases, you only need to adjust the size of the dump volume if the default dump volume size is too small. For example, on a large-memory system, the dump volume size is increased to 40 GB as follows:

  # zfs set volsize=40G rpool/dump

  Resizing a large dump volume can be a time-consuming process.

  If, for any reason, you need to enable a dump device after you manually create a dump device, use syntax similar to the following:

  # dumpadm -d /dev/zvol/dsk/rpool/dump
        Dump content: kernel pages
         Dump device: /dev/zvol/dsk/rpool/dump (dedicated)
  Savecore directory: /var/crash/t2000
    Savecore enabled: yes

• A system with 128 GB or greater memory night need a larger dump device than the dump device that is created by default. If the dump device is too small to capture an existing crash dump, a message similar to the following is displayed:

  # dumpadm -d /dev/zvol/dsk/rpool/dump
  dumpadm: dump device /dev/zvol/dsk/rpool/dump is too small to hold a system dump
  dump size 36255432704 bytes, device size 34359738368 bytes

  For information about sizing the swap and dump devices, see Planning for Swap Space in System Administration Guide: Devices and File Systems.

• You cannot currently add a dump device to a pool with multiple top-level devices. You will see a message similar to the following:

  # dumpadm -d /dev/zvol/dsk/datapool/dump 
  dump is not supported on device '/dev/zvol/dsk/datapool/dump': 'datapool' has multiple top level vdevs

  Add the dump device to the root pool, which cannot have multiple top-level devices.