Table des matières
SO303 - Storage Administration
Preparing your Solaris 11 VM
Before continuing further, shutdown your Solaris 11 VM. Using the Storage section of the Oracle VM VirtualBox Manager, add the following .vmdk disks to the existing SATA controller of your Solaris 11 VM:
| Disk | Size | Name |
|---|---|---|
| c7t2d0 | 200 Mb | Disk1.vmdk |
| c7t3d0 | 200 Mb | Disk2.vmdk |
| c7t4d0 | 200 Mb | Disk3.vmdk |
| c7t5d0 | 200 Mb | Disk4.vmdk |
| c7t6d0 | 200 Mb | Disk5.vmdk |
| c7t7d0 | 20 Gb | Mirror.vmdk |
Using the System section of the Oracle VM VirtualBox Manager, add a second processor to your Solaris 11 VM.
Finally, boot your Solaris 11 VM.
Introduction
All previous versions of Solaris, including Solaris 10 use the Unix File System ( UFS ) as their default file system. Solaris 11 uses the Zettabyte File System ( ZFS ), introduced with Solaris 10, as its default file system.
Solaris 11 and ZFS
The Solaris 11 implementation of ZFS includes the following capabilities:
- 128-bit addressing,
- data integrity assurance,
- automated data corruption detection and repair,
- encryption,
- compression,
- de-duplication,
- quotas,
- file system migration between pools,
- snapshots.
ZFS Vocabulary
The introduction of ZFS was obviously accompanied by a new vocabulary:
| Term | Description |
|---|---|
| pool | A storage element regrouping one or more disk partitions containing one or more file systems |
| file system | A dataset containing directories and files |
| clone | A copy of a file system |
| snapshot | A read-only copy of the state of a file system |
| compression | The reduction of storage space achieved by the removal of duplicate data blocks |
| de-duplication | The reduction of storage space achieved by the removal of redundant data patterns |
| checksum | A 256-bit number used to validate data when read or written |
| encryption | The protection of data using a password |
| quota | The maximum amount of disk space used by a group or user |
| reservation | A preallocated amount of disk space assigned to a user or file system |
| mirror | An exact duplicate of a disk or partition |
| RAID-Z | ZFS implementation of RAID-5 |
| RAID-Z2 | ZFS implementation of RAID-6 |
| RAID-Z3 | ZFS implementation of Triple Parity RAID |
ZFS Commands
The ZFS commands are as follows:
| Command | Description |
|---|---|
| zpool | Used to manage ZFS pools |
| zfs | Used to manage ZFS file systems |
The zpool Command
The zpool command uses a set of subcommands:
| Command | Description |
|---|---|
| create | Creates a storage pool and configures its mount point |
| destroy | Destroys a storage pool |
| list | Displays the health and storage usage of a pool |
| get | Displays a list of pool properties |
| set | Sets a property for a pool |
| status | Displays the health of a pool |
| history | Displays the commands issued for a pool since its creation |
| add | Adds a disk to an existing pool |
| remove | Removes a disk from an existing pool |
| replace | Replaces a disk in a pool by another disk |
| scrub | Verifies the checksums of a pool and repairs any damaged data blocks |
The zfs Command
The zfs command use a set of subcommands:
| Command | Description |
|---|---|
| create | Creates a ZFS file system, sets its properties and automatically mounts it |
| destroy | Destroys a ZFS file system or snapshot |
| list | Displays the properties and storage usage of a ZFS file system |
| get | Displays a list of ZFS file system properties |
| set | Sets a property for a ZFS file system |
| snapshot | Creates a read-only copy of the state of a ZFS file system |
| rollback | Returns the file system to the state of the last snapshot |
| send | Creates a file from a snapshot in order to migrate it to another pool |
| receive | Retrieves a file created by the subcommand send |
| clone | Creates a copy of a snapshot |
| promote | Transforms a clone into a ZFS file system |
| diff | Displays the file differences between two snapshots or a snapshot and its parent file system |
| mount | Mounts a ZFS file system at a specific mount point |
| unmount | Unmounts a ZFS file system |
Solaris Slices
Those familiar with UFS on Solaris will remember having to manipulate Solaris slices. Those slices still exist:
root@solaris:~# format
Searching for disks...done
AVAILABLE DISK SELECTIONS:
0. c7t0d0 <ATA-VBOX HARDDISK-1.0-20.00GB>
/pci@0,0/pci8086,2829@d/disk@0,0
1. c7t2d0 <ATA-VBOX HARDDISK-1.0-200.00MB>
/pci@0,0/pci8086,2829@d/disk@2,0
2. c7t3d0 <ATA-VBOX HARDDISK-1.0-200.00MB>
/pci@0,0/pci8086,2829@d/disk@3,0
3. c7t4d0 <ATA-VBOX HARDDISK-1.0-200.00MB>
/pci@0,0/pci8086,2829@d/disk@4,0
4. c7t5d0 <ATA-VBOX HARDDISK-1.0-200.00MB>
/pci@0,0/pci8086,2829@d/disk@5,0
5. c7t6d0 <ATA-VBOX HARDDISK-1.0-200.00MB>
/pci@0,0/pci8086,2829@d/disk@6,0
6. c7t7d0 <ATA-VBOX HARDDISK-1.0 cyl 2608 alt 2 hd 255 sec 63>
/pci@0,0/pci8086,2829@d/disk@7,0
Specify disk (enter its number): 0
selecting c7t0d0
[disk formatted]
/dev/dsk/c7t0d0s1 is part of active ZFS pool rpool. Please see zpool(1M).
FORMAT MENU:
disk - select a disk
type - select (define) a disk type
partition - select (define) a partition table
current - describe the current disk
format - format and analyze the disk
fdisk - run the fdisk program
repair - repair a defective sector
label - write label to the disk
analyze - surface analysis
defect - defect list management
backup - search for backup labels
verify - read and display labels
inquiry - show disk ID
volname - set 8-character volume name
!<cmd> - execute <cmd>, then return
quit
format> part
PARTITION MENU:
0 - change `0' partition
1 - change `1' partition
2 - change `2' partition
3 - change `3' partition
4 - change `4' partition
5 - change `5' partition
6 - change `6' partition
select - select a predefined table
modify - modify a predefined partition table
name - name the current table
print - display the current table
label - write partition map and label to the disk
!<cmd> - execute <cmd>, then return
quit
partition> pri
Current partition table (original):
Total disk sectors available: 41926589 + 16384 (reserved sectors)
Part Tag Flag First Sector Size Last Sector
0 BIOS_boot wm 256 255.88MB 524287
1 usr wm 524288 19.74GB 41913087
2 unassigned wm 0 0 0
3 unassigned wm 0 0 0
4 unassigned wm 0 0 0
5 unassigned wm 0 0 0
6 unassigned wm 0 0 0
8 reserved wm 41913088 8.00MB 41929471
partition>
Note the following line in the above output:
/dev/dsk/c7t0d0s1 is part of active ZFS pool rpool. Please see zpool(1M).
Since you are using ZFS for storage management, you no longer need to bother about slices !
iSCSI Storage
In Solaris 10 the configuration of iSCSI LUNs was accomplished using the iscsitadm command and the ZFS shareiscsi property. In Solaris 11 these have been replaced by by the use of Common Multiprotocol SCSI Target ( COMSTAR ). COMSTAR is a framework that turns a Solaris host into a SCSI target.
COMSTAR includes the following features:
- scalability,
- compatibility with generic host adapters,
- multipathing,
- LUN masking and mapping functions.
An iSCSI target is an endpoint waiting for connections from clients called initiators. A target can provide multiple Logical Units which provides classic read and write data operations.
Each logical unit is backed by a storage device. You can create a logical unit backed by any one of the following:
- a file,
- a thin-provisioned file,
- a disk partition,
- a ZFS volume.
LAB #1 - Managing ZFS Storage
Displaying Online Help
Both the zpooland zfs commands have built-in online help:
root@solaris:~# zpool help The following commands are supported: add attach clear create destroy detach export get help history import iostat list offline online remove replace scrub set split status upgrade For more info, run: zpool help <command> root@solaris:~# zfs help The following commands are supported: allow clone create destroy diff get groupspace help hold holds inherit key list mount promote receive release rename rollback send set share snapshot unallow unmount unshare upgrade userspace For more info, run: zfs help <command>
Note that you can get help on subcommands by either using zpool help <subcommand> or zfs help <subcommand>.
Checking Pool Status
Use the zpool command with the list subcommand to display the details of your pool:
root@solaris:~# zpool list NAME SIZE ALLOC FREE CAP DEDUP HEALTH ALTROOT rpool 19.6G 6.96G 12.7G 35% 1.00x ONLINE -
Now use the status subcommand:
root@solaris:~# zpool status pool: rpool state: ONLINE scan: none requested config: NAME STATE READ WRITE CKSUM rpool ONLINE 0 0 0 c7t0d0s1 ONLINE 0 0 0 errors: No known data errors
Creating a Mirrored Pool
Create a ZFS mirrored pool called mypool using the first two of the five disks you recently created:
root@solaris:~# zpool create mypool mirror c7t2d0 c7t3d0
Check that your pool has been created:
root@solaris:~# zpool list NAME SIZE ALLOC FREE CAP DEDUP HEALTH ALTROOT mypool 187M 346K 187M 0% 1.00x ONLINE - rpool 19.6G 6.97G 12.7G 35% 1.00x ONLINE -
Display the file systems using the zfs command and the list subcommand:
root@solaris:~# zfs list NAME USED AVAIL REFER MOUNTPOINT mypool 272K 155M 31K /mypool rpool 7.02G 12.3G 4.58M /rpool rpool/ROOT 4.87G 12.3G 31K legacy rpool/ROOT/solaris 4.86G 12.3G 3.92G / rpool/ROOT/solaris-backup-1 2.47M 12.3G 1.98G / rpool/ROOT/solaris-backup-1/var 46K 12.3G 758M /var rpool/ROOT/solaris/var 865M 12.3G 207M /var rpool/VARSHARE 100K 12.3G 100K /var/share rpool/dump 1.03G 12.3G 1.00G - rpool/export 87.6M 12.3G 32K /export rpool/export/home 87.5M 12.3G 32K /export/home rpool/export/home/trainee 87.5M 12.3G 87.5M /export/home/trainee rpool/swap 1.03G 12.3G 1.00G -
Note that the zpool command automatically creates a file system on mypool and mounts it at /mypool.
Adding File Systems to an Existing Pool
Now create two file systems in your pool called /home and /home/user1 and then display the results :
root@solaris:~# zfs create mypool/home root@solaris:~# zfs create mypool/home/user1 root@solaris:~# zfs list NAME USED AVAIL REFER MOUNTPOINT mypool 184K 155M 32K /mypool mypool/home 63K 155M 32K /mypool/home mypool/home/user1 31K 155M 31K /mypool/home/user1 rpool 7.02G 12.3G 4.58M /rpool rpool/ROOT 4.87G 12.3G 31K legacy rpool/ROOT/solaris 4.86G 12.3G 3.92G / rpool/ROOT/solaris-backup-1 2.47M 12.3G 1.98G / rpool/ROOT/solaris-backup-1/var 46K 12.3G 758M /var rpool/ROOT/solaris/var 865M 12.3G 207M /var rpool/VARSHARE 100K 12.3G 100K /var/share rpool/dump 1.03G 12.3G 1.00G - rpool/export 87.6M 12.3G 32K /export rpool/export/home 87.5M 12.3G 32K /export/home rpool/export/home/trainee 87.5M 12.3G 87.5M /export/home/trainee rpool/swap 1.03G 12.3G 1.00G -
Note that the two file systems share the same disk space as the parent pool.
Changing the Pool Mount Point
Suppose that you want the /home file system mounted elsewhere rather than under the /mypool mount point. With ZFS, this is very simple:
root@solaris:~# mkdir /users root@solaris:~# zfs set mountpoint=/users mypool/home root@solaris:~# zfs list NAME USED AVAIL REFER MOUNTPOINT mypool 196K 155M 31K /mypool mypool/home 63K 155M 32K /users mypool/home/user1 31K 155M 31K /users/user1 rpool 7.02G 12.3G 4.58M /rpool rpool/ROOT 4.87G 12.3G 31K legacy rpool/ROOT/solaris 4.86G 12.3G 3.92G / rpool/ROOT/solaris-backup-1 2.47M 12.3G 1.98G / rpool/ROOT/solaris-backup-1/var 46K 12.3G 758M /var rpool/ROOT/solaris/var 865M 12.3G 207M /var rpool/VARSHARE 100K 12.3G 100K /var/share rpool/dump 1.03G 12.3G 1.00G - rpool/export 87.6M 12.3G 32K /export rpool/export/home 87.5M 12.3G 32K /export/home rpool/export/home/trainee 87.5M 12.3G 87.5M /export/home/trainee rpool/swap 1.03G 12.3G 1.00G -
Note that ZFS has automatically and transparently unmounted /mypool/home and re-mounted it at /users.
Adding a Hot Spare
To display all of the properties associated with mypool, use the zpool command and the get subcommand:
root@solaris:~# zpool get all mypool NAME PROPERTY VALUE SOURCE mypool allocated 673K - mypool altroot - default mypool autoexpand off default mypool autoreplace off default mypool bootfs - default mypool cachefile - default mypool capacity 0% - mypool dedupditto 0 default mypool dedupratio 1.00x - mypool delegation on default mypool failmode wait default mypool free 186M - mypool guid 6502877439742337134 - mypool health ONLINE - mypool listshares off default mypool listsnapshots off default mypool readonly off - mypool size 187M - mypool version 34 default
Note that the autoreplace property is set to off. In order to use a hot spare, this property needs to be set to on.
Set the autoreplace property to on:
root@solaris:~# zpool set autoreplace=on mypool root@solaris:~# zpool get autoreplace mypool NAME PROPERTY VALUE SOURCE mypool autoreplace on local
Add the fourth 200 Mb disk that you have created to mypool as a spare:
root@solaris:~# zpool add mypool spare c7t5d0 root@solaris:~# zpool status mypool pool: mypool state: ONLINE scan: none requested config: NAME STATE READ WRITE CKSUM mypool ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 c7t2d0 ONLINE 0 0 0 c7t3d0 ONLINE 0 0 0 spares c7t5d0 AVAIL errors: No known data errors
Observing Pool Activity
Create a random data file in /users/user1:
root@solaris:~# cat /dev/urandom > /users/user1/randomfile & [1] 2617
Write down the PID, you will need it in 2 minutes to kill the process you have just started.
Now display the writes to the pool using the iostat subcommand of the zpool command:
root@solaris:~# zpool iostat -v 3
capacity operations bandwidth
pool alloc free read write read write
---------- ----- ----- ----- ----- ----- -----
mypool 29.6M 157M 0 5 184 50.6K
mirror 29.6M 157M 0 5 184 50.6K
c7t2d0 - - 0 5 1.46K 59.8K
c7t3d0 - - 0 5 1.33K 56.9K
---------- ----- ----- ----- ----- ----- -----
rpool 6.96G 12.7G 0 14 27.3K 124K
c7t0d0s1 6.96G 12.7G 0 14 27.3K 124K
---------- ----- ----- ----- ----- ----- -----
capacity operations bandwidth
pool alloc free read write read write
---------- ----- ----- ----- ----- ----- -----
mypool 39.9M 147M 0 127 0 2.89M
mirror 39.9M 147M 0 127 0 2.89M
c7t2d0 - - 0 113 0 2.78M
c7t3d0 - - 0 110 0 2.89M
---------- ----- ----- ----- ----- ----- -----
rpool 6.96G 12.7G 0 13 0 29.5K
c7t0d0s1 6.96G 12.7G 0 13 0 29.5K
---------- ----- ----- ----- ----- ----- -----
capacity operations bandwidth
pool alloc free read write read write
---------- ----- ----- ----- ----- ----- -----
mypool 53.3M 134M 0 128 0 2.78M
mirror 53.3M 134M 0 128 0 2.78M
c7t2d0 - - 0 112 0 2.44M
c7t3d0 - - 0 113 0 2.78M
---------- ----- ----- ----- ----- ----- -----
rpool 6.96G 12.7G 0 77 0 500K
c7t0d0s1 6.96G 12.7G 0 77 0 500K
---------- ----- ----- ----- ----- ----- -----
capacity operations bandwidth
pool alloc free read write read write
---------- ----- ----- ----- ----- ----- -----
mypool 65.5M 121M 0 171 0 3.08M
mirror 65.5M 121M 0 171 0 3.08M
c7t2d0 - - 0 153 0 3.08M
c7t3d0 - - 0 153 0 3.08M
---------- ----- ----- ----- ----- ----- -----
rpool 6.96G 12.7G 0 21 0 45.1K
c7t0d0s1 6.96G 12.7G 0 21 0 45.1K
---------- ----- ----- ----- ----- ----- -----
capacity operations bandwidth
pool alloc free read write read write
---------- ----- ----- ----- ----- ----- -----
mypool 75.8M 111M 0 172 0 2.88M
mirror 75.8M 111M 0 172 0 2.88M
c7t2d0 - - 0 149 0 2.85M
c7t3d0 - - 0 149 0 2.88M
---------- ----- ----- ----- ----- ----- -----
rpool 6.96G 12.7G 0 0 0 882
c7t0d0s1 6.96G 12.7G 0 0 0 882
---------- ----- ----- ----- ----- ----- -----
^C
Is your mirror functioning ?
Now kill the process creating the file randomfile :
# kill -9 PID [Entrée]
Delete the file /users/user1/randomfile:
root@solaris:~# rm -rf /users/user1/randomfile [1]+ Killed cat /dev/urandom > /users/user1/randomfile
Setting a User Quota
To set a user quota, you need to use the set subcommand of zpool:
root@solaris:~# zfs set quota=50M mypool/home/user1 root@solaris:~# zfs get quota mypool NAME PROPERTY VALUE SOURCE mypool quota none default root@solaris:~# zfs list NAME USED AVAIL REFER MOUNTPOINT mypool 380K 155M 31K /mypool mypool/home 63K 155M 32K /users mypool/home/user1 31K 50.0M 31K /users/user1 rpool 7.03G 12.3G 4.58M /rpool rpool/ROOT 4.87G 12.3G 31K legacy rpool/ROOT/solaris 4.86G 12.3G 3.92G / rpool/ROOT/solaris-backup-1 2.47M 12.3G 1.98G / rpool/ROOT/solaris-backup-1/var 46K 12.3G 758M /var rpool/ROOT/solaris/var 865M 12.3G 207M /var rpool/VARSHARE 101K 12.3G 101K /var/share rpool/dump 1.03G 12.3G 1.00G - rpool/export 89.3M 12.3G 32K /export rpool/export/home 89.3M 12.3G 32K /export/home rpool/export/home/trainee 89.3M 12.3G 89.3M /export/home/trainee rpool/swap 1.03G 12.3G 1.00G -
Note that the quota of 50 Mb has been set on /mypool/home/user1.
Now create a random data file in /users/user1:
root@solaris:~# cat /dev/urandom > /users/user1/testfile cat: output error (0/131072 characters written) Disc quota exceeded
After a few minutes, you will see the Disc quota exceeded message.
Looking at the available disk space on /users/user1 you will notice that the value is now 0:
root@solaris:~# zfs list mypool/home/user1 NAME USED AVAIL REFER MOUNTPOINT mypool/home/user1 50.1M 0 50.1M /users/user1
Delete the testfile file:
root@solaris:~# rm -f /users/user1/testfile
Setting a User Reservation
As with setting quotas, setting a reservation is very simple:
root@solaris:~# zfs set reservation=25M mypool/home/user1 root@solaris:~# zfs get reservation mypool/home/user1 NAME PROPERTY VALUE SOURCE mypool/home/user1 reservation 25M local
Using Snapshots
Create a file in /users/user1:
root@solaris:~# echo "This is a test file for the first snapshot" > /users/user1/snapshot1 root@solaris:~# ls /users/user1 snapshot1
To create a snapshot of a ZFS file system, you need to use the snapshot subcommand of the zfs command:
root@solaris:~# zfs snapshot mypool/home/user1@Dec13
The snapshot is located in a hidden directory under /users/user1 called .zfs/snapshot:
root@solaris:~# ls -l /users/user1/.zfs/snapshot total 3 drwxr-xr-x 2 root root 3 Dec 13 10:24 Dec13
As you can see, the snapshot contains the snapshot1 file:
root@solaris:~# ls -l /users/user1/.zfs/snapshot/Dec13/ total 2 -rw-r--r-- 1 root root 43 Dec 13 10:24 snapshot1
It is important to note here that the .zfs directory is also hidden from the ls command, even when using the -a switch:
root@solaris:~# ls -laR /users/user1 /users/user1: total 8 drwxr-xr-x 2 root root 3 Dec 13 10:24 . drwxr-xr-x 3 root root 3 Dec 12 16:09 .. -rw-r--r-- 1 root root 43 Dec 13 10:24 snapshot1
You can also create a recursive snapshot of all file systems in a pool:
root@solaris:~# zfs snapshot -r mypool@Dec13-1
The snapshots are stored in their respective .zfs directories:
root@solaris:~# ls /users/.zfs/snapshot Dec13-1 root@solaris:~# ls /users/user1/.zfs/snapshot Dec13 Dec13-1
You can list all snapshots as follows:
root@solaris:~# zfs list -t snapshot -r mypool NAME USED AVAIL REFER MOUNTPOINT mypool@Dec13-1 0 - 31K - mypool/home@Dec13-1 0 - 32K - mypool/home/user1@Dec13 0 - 31.5K - mypool/home/user1@Dec13-1 0 - 31.5K -
Create another file in /users/user1:
root@solaris:~# echo "This is a test file for the second snapshot" > /users/user1/snapshot2 root@solaris:~# ls -l /users/user1 total 4 -rw-r--r-- 1 root root 43 Dec 13 10:44 snapshot1 -rw-r--r-- 1 root root 44 Dec 13 10:45 snapshot2 root@solaris:~# cat /users/user1/snapshot1 This is a test file for the first snapshot root@solaris:~# cat /users/user1/snapshot2 This is a test file for the second snapshot
Now take a second recursive snapshot of mypool:
root@solaris:~# zfs snapshot -r mypool@Dec13-2 root@solaris:~# zfs list -t snapshot -r mypool NAME USED AVAIL REFER MOUNTPOINT mypool@Dec13-1 0 - 31K - mypool@Dec13-2 0 - 31K - mypool/home@Dec13-1 0 - 32K - mypool/home@Dec13-2 0 - 32K - mypool/home/user1@Dec13 0 - 31.5K - mypool/home/user1@Dec13-1 0 - 31.5K - mypool/home/user1@Dec13-2 0 - 33K -
The diff subcommand of the zfs command displays the differences between two snapshots:
root@solaris:~# zfs diff mypool/home/user1@Dec13-1 mypool/home/user1@Dec13-2 M /users/user1/ M /users/user1/snapshot1 + /users/user1/snapshot2
The above out put shows that /users/user1/snapshot2 has been added and included in the second snapshot as it appears in the command line.
This output can contain the following characters:
| Character | Description |
|---|---|
| M | Modification |
| R | Renamed |
| + | Added |
| - | Deleted |
Note that you cannot compare the snapshots in the reverse order:
root@solaris:~# zfs diff mypool/home/user1@Dec13-2 mypool/home/user1@Dec13-1 Unable to obtain diffs: mypool/home/user1@Dec13-1 is not a descendant dataset of mypool/home/user1@Dec13-2
Rolling Back to a Snapshot
In the case that you wish to rollback to a specific snapshot, note that you can only roll back to the last snapshot as shown by the output of zfs list:
root@solaris:~# zfs list -t snapshot -r mypool NAME USED AVAIL REFER MOUNTPOINT mypool@Dec13-1 0 - 31K - mypool@Dec13-2 0 - 31K - mypool/home@Dec13-1 0 - 32K - mypool/home@Dec13-2 0 - 32K - mypool/home/user1@Dec13 0 - 31.5K - mypool/home/user1@Dec13-1 0 - 31.5K - mypool/home/user1@Dec13-2 0 - 33K - root@solaris:~# zfs rollback mypool/home/user1@Dec13-1 cannot rollback to 'mypool/home/user1@Dec13-1': more recent snapshots exist use '-r' to force deletion of the following snapshots: mypool/home/user1@Dec13-2
Delete the Dec13-2 snapshot as follows:
root@solaris:~# zfs destroy mypool/home/user1@Dec13-2 root@solaris:~# zfs list -t snapshot -r mypool NAME USED AVAIL REFER MOUNTPOINT mypool@Dec13-1 0 - 31K - mypool@Dec13-2 0 - 31K - mypool/home@Dec13-1 0 - 32K - mypool/home@Dec13-2 0 - 32K - mypool/home/user1@Dec13 0 - 31.5K - mypool/home/user1@Dec13-1 0 - 31.5K -
Now roll back to Dec13-1:
root@solaris:~# zfs rollback mypool/home/user1@Dec13-1 root@solaris:~# ls -l /users/user1 total 2 -rw-r--r-- 1 root root 43 Dec 13 10:24 snapshot1
Note that the snapshot2 file has obviously disappeared since it was not in the Dec13-1 snapshot.
Cloning a Snapshot
Snapshots are read-only. To convert a snapshot to a writable file system, you can use the clone subcommand of the zfs command:
root@solaris:~# zfs clone mypool/home/user1@Dec13-1 mypool/home/user3 root@solaris:~# zfs list NAME USED AVAIL REFER MOUNTPOINT mypool 25.5M 129M 31K /mypool mypool/home 25.1M 129M 34K /users mypool/home/user1 50.5K 50.0M 31.5K /users/user1 mypool/home/user3 18K 129M 31.5K /users/user3 rpool 7.03G 12.3G 4.58M /rpool rpool/ROOT 4.87G 12.3G 31K legacy rpool/ROOT/solaris 4.86G 12.3G 3.92G / rpool/ROOT/solaris-backup-1 2.47M 12.3G 1.98G / rpool/ROOT/solaris-backup-1/var 46K 12.3G 758M /var rpool/ROOT/solaris/var 865M 12.3G 207M /var rpool/VARSHARE 102K 12.3G 102K /var/share rpool/dump 1.03G 12.3G 1.00G - rpool/export 90.5M 12.3G 32K /export rpool/export/home 90.5M 12.3G 32K /export/home rpool/export/home/trainee 90.4M 12.3G 90.4M /export/home/trainee rpool/swap 1.03G 12.3G 1.00G -
Display the contents of /users/user3:
root@solaris:~# ls -l /users/user3 total 2 -rw-r--r-- 1 root root 43 Dec 13 10:24 snapshot1
Using Compression
In order to minimize storage space, you can make a file system use compression. Compression can be activated either at creation time or after creation. Compression only works for new data. Any existing data in the file system at the time of activating compression remains uncompressed.
To activate compression on an existing file system, you need to change the file system's compression property from off to on:
root@solaris:~# zfs set compression=on mypool/home/user1 root@solaris:~# zfs get compression mypool/home/user1 NAME PROPERTY VALUE SOURCE mypool/home/user1 compression on local
Using De-duplication
Another space saving property of ZFS file systems is De-duplication:
root@solaris:~# zfs set dedup=on mypool/home/user1 root@solaris:~# zfs get dedup mypool/home/user1 NAME PROPERTY VALUE SOURCE mypool/home/user1 dedup on local
Using Encryption
Unlike Compression and De-duplication, Encryption can only be set on a file system at the time of creation:
root@solaris:~# zfs create -o encryption=on mypool/home/user2 Enter passphrase for 'mypool/home/user2': fenestros Enter again: fenestros
Note that the passphrase is not shown in the real output of the command. It is in the above example only for the purposes of this lesson.
To check if encryption is active on a file system, use the following command:
root@solaris:~# zfs get encryption mypool/home/user1 NAME PROPERTY VALUE SOURCE mypool/home/user1 encryption off - root@solaris:~# zfs get encryption mypool/home/user2 NAME PROPERTY VALUE SOURCE mypool/home/user2 encryption on local
Replacing a Faulty Disk
In the case of a faulty disk and no hot spares, replacing the disk is a one-line operation using the replace subcommand of the zpool command:
root@solaris:~# zpool status mypool pool: mypool state: ONLINE scan: none requested config: NAME STATE READ WRITE CKSUM mypool ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 c7t2d0 ONLINE 0 0 0 c7t3d0 ONLINE 0 0 0 spares c7t5d0 AVAIL errors: No known data errors root@solaris:~# zpool replace mypool c7t2d0 c7t4d0
Use the status subcommand of the zpool command again to see what has happened:
root@solaris:~# zpool status mypool pool: mypool state: ONLINE scan: resilvered 601K in 0h0m with 0 errors on Thu Dec 13 11:45:49 2012 config: NAME STATE READ WRITE CKSUM mypool ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 c7t4d0 ONLINE 0 0 0 c7t3d0 ONLINE 0 0 0 spares c7t5d0 AVAIL errors: No known data errors
ZFS Resilvering is the equivalent of UFS re-synchronization.
Destroying a Pool
Destroying a pool is achieved by using the destroy subcommand of the zpool command:
root@solaris:~# zpool destroy mypool
As you can see by the following output, this operation has also destroyed all the associated snapshots:
root@solaris:~# zfs list NAME USED AVAIL REFER MOUNTPOINT rpool 7.03G 12.3G 4.58M /rpool rpool/ROOT 4.87G 12.3G 31K legacy rpool/ROOT/solaris 4.86G 12.3G 3.92G / rpool/ROOT/solaris-backup-1 2.47M 12.3G 1.98G / rpool/ROOT/solaris-backup-1/var 46K 12.3G 758M /var rpool/ROOT/solaris/var 865M 12.3G 207M /var rpool/VARSHARE 102K 12.3G 102K /var/share rpool/dump 1.03G 12.3G 1.00G - rpool/export 90.5M 12.3G 32K /export rpool/export/home 90.5M 12.3G 32K /export/home rpool/export/home/trainee 90.4M 12.3G 90.4M /export/home/trainee rpool/swap 1.03G 12.3G 1.00G - root@solaris:~# zfs list -t snapshot -r mypool cannot open 'mypool': filesystem does not exist root@solaris:~# ls -l /users total 0
As you have seen above, destroying a pool, all the data in it and all the associated snapshots is disconcertingly simple. You should therefore be very careful when using the destroy subcommand.
Creating a RAID-5 Pool
You can create a RAID-5 pool using the RAID-Z algorithm:
root@solaris:~# zpool create mypool raidz c7t2d0 c7t3d0 c7t4d0 spare c7t5d0 root@solaris:~# zpool status mypool pool: mypool state: ONLINE scan: none requested config: NAME STATE READ WRITE CKSUM mypool ONLINE 0 0 0 raidz1-0 ONLINE 0 0 0 c7t2d0 ONLINE 0 0 0 c7t3d0 ONLINE 0 0 0 c7t4d0 ONLINE 0 0 0 spares c7t5d0 AVAIL errors: No known data errors
Destroy mypool :
root@solaris:~# zpool destroy mypool
Creating a RAID-6 Pool
You can create a RAID-6 pool using the RAID-Z2 algorithm:
root@solaris:~# zpool create mypool raidz2 c7t2d0 c7t3d0 c7t4d0 c7t5d0 spare c7t6d0 root@solaris:~# zpool status mypool pool: mypool state: ONLINE scan: none requested config: NAME STATE READ WRITE CKSUM mypool ONLINE 0 0 0 raidz2-0 ONLINE 0 0 0 c7t2d0 ONLINE 0 0 0 c7t3d0 ONLINE 0 0 0 c7t4d0 ONLINE 0 0 0 c7t5d0 ONLINE 0 0 0 spares c7t6d0 AVAIL errors: No known data errors
Destroy mypool :
root@solaris:~# zpool destroy mypool
Create a triple parity RAID mypool using your five 200MB disks. Do not delete it.
Displaying the Zpool History
You can review everything that has been done to existing pools by using the history subcommand of the zpool command:
root@solaris:~# zpool history History for 'mypool': 2012-12-13.14:02:17 zpool create mypool raidz3 c7t2d0 c7t3d0 c7t4d0 c7t5d0 spare c7t6d0 History for 'rpool': 2012-11-20.19:08:05 zpool create -f -B rpool c7t0d0s1 2012-11-20.19:08:05 zfs create -p -o mountpoint=/export rpool/export 2012-11-20.19:08:05 zfs set mountpoint=/export rpool/export 2012-11-20.19:08:05 zfs create -p rpool/export/home 2012-11-20.19:08:06 zfs create -p -o canmount=noauto -o mountpoint=/var/share rpool/VARSHARE 2012-11-20.19:08:06 zfs set mountpoint=/var/share rpool/VARSHARE 2012-11-20.19:08:07 zfs create -p -V 1024.0m rpool/dump 2012-11-20.19:08:12 zfs create -p -V 1024.0m rpool/swap 2012-11-20.19:08:20 zfs set primarycache=metadata rpool/swap 2012-11-20.19:25:51 zfs set primarycache=metadata rpool/swap 2012-11-20.19:26:25 zfs create rpool/export/home/trainee 2012-11-20.22:45:56 zfs set primarycache=metadata rpool/swap 2012-12-01.14:32:36 zfs set primarycache=metadata rpool/swap 2012-12-03.13:15:45 zfs set primarycache=metadata rpool/swap 2012-12-08.14:33:41 zfs /tmp/be 2012-12-11.15:33:50 zfs set primarycache=metadata rpool/swap 2012-12-12.09:57:00 zfs set primarycache=metadata rpool/swap
Note that the history related to destroyed pools has been deleted.
LAB #2 - Managing iSCSI Storage
Installing the COMSTAR Server
Start by installing the COMSTAR storage server software:
root@solaris:~# pkg install group/feature/storage-server
Packages to install: 20
Create boot environment: No
Create backup boot environment: Yes
Services to change: 1
DOWNLOAD PKGS FILES XFER (MB) SPEED
Completed 20/20 1023/1023 54.5/54.5 636k/s
PHASE ITEMS
Installing new actions 1863/1863
Updating package state database Done
Updating image state Done
Creating fast lookup database Done
The COMSTAR target mode framework runs as the stmf service. Check to see if it is enabled:
root@solaris:~# svcs \*stmf\* STATE STIME FMRI disabled 15:43:16 svc:/system/stmf:default
Enable the service:
root@solaris:~# svcadm enable stmf root@solaris:~# svcs \*stmf\* STATE STIME FMRI online 16:01:56 svc:/system/stmf:default
You can check the status of the server using the stmfadm command:
root@solaris:~# stmfadm list-state Operational Status: online Config Status : initialized ALUA Status : disabled ALUA Node : 0
Creating SCSI Logical Units
First you need to create your Backing Storage Device within your mypool pool:
root@solaris:~# zfs create -V 100M mypool/iscsi root@solaris:~# zfs list NAME USED AVAIL REFER MOUNTPOINT mypool 103M 51.6M 31K /mypool mypool/iscsi 103M 155M 16K - rpool 7.40G 11.9G 4.58M /rpool rpool/ROOT 5.22G 11.9G 31K legacy rpool/ROOT/solaris 5.22G 11.9G 4.08G / rpool/ROOT/solaris-backup-1 2.47M 11.9G 1.98G / rpool/ROOT/solaris-backup-1/var 46K 11.9G 758M /var rpool/ROOT/solaris-backup-2 127K 11.9G 3.92G / rpool/ROOT/solaris-backup-2/var 58K 11.9G 266M /var rpool/ROOT/solaris/var 980M 11.9G 209M /var rpool/VARSHARE 102K 11.9G 102K /var/share rpool/dump 1.03G 12.0G 1.00G - rpool/export 108M 11.9G 32K /export rpool/export/home 108M 11.9G 32K /export/home rpool/export/home/trainee 108M 11.9G 108M /export/home/trainee rpool/swap 1.03G 12.0G 1.00G -
You can see your raw device in the /dev/zvol/rdsk/mypool/ directory:
root@solaris:~# ls -l /dev/zvol/rdsk/mypool total 0 lrwxrwxrwx 1 root root 0 Dec 14 09:42 iscsi -> ../../../..//devices/pseudo/zfs@0:6,raw
You can now create a logical unit using the create-lu subcommand of the sbdadm command:
root@solaris:~# sbdadm create-lu /dev/zvol/rdsk/mypool/iscsi Created the following LU: GUID DATA SIZE SOURCE -------------------------------- ------------------- ---------------- 600144f0e2a54e00000050cae6d80001 104857600 /dev/zvol/rdsk/mypool/iscsi
Mapping the Logical Unit
In order for the logical unit to be available to initiators, it has to be mapped. In order to map the logical device you need its GUID. You can use either one of the two following commands to get that information:
root@solaris:~# sbdadm list-lu Found 1 LU(s) GUID DATA SIZE SOURCE -------------------------------- ------------------- ---------------- 600144f0e2a54e00000050cae6d80001 104857600 /dev/zvol/rdsk/mypool/iscsi
root@solaris:~# stmfadm list-lu -v
LU Name: 600144F0E2A54E00000050CAE6D80001
Operational Status : Online
Provider Name : sbd
Alias : /dev/zvol/rdsk/mypool/iscsi
View Entry Count : 0
Data File : /dev/zvol/rdsk/mypool/iscsi
Meta File : not set
Size : 104857600
Block Size : 512
Management URL : not set
Vendor ID : SUN
Product ID : COMSTAR
Serial Num : not set
Write Protect : Disabled
Write Cache Mode Select: Enabled
Writeback Cache : Enabled
Access State : Active
Create simple mapping for this logical unit by using the add-view subcommand of the stmfadm command:
root@solaris:~# stmfadm add-view 600144F0E2A54E00000050CAE6D80001
Creating a Target
In order to create a target the svc:/network/iscsi/target:default service must be online. Check if it is:
root@solaris:~# svcs \*scsi\* STATE STIME FMRI disabled 15:42:56 svc:/network/iscsi/target:default online Dec_12 svc:/network/iscsi/initiator:default
Start the service:
root@solaris:~# svcadm enable -r svc:/network/iscsi/target:default root@solaris:~# svcs \*scsi\* STATE STIME FMRI online Dec_12 svc:/network/iscsi/initiator:default online 10:06:54 svc:/network/iscsi/target:default
Now create a target using the create-target subcommand of the itadm command:
root@solaris:~# itadm create-target Target iqn.1986-03.com.sun:02:897fd011-8b3d-cf2b-fc1d-a010bd97d035 successfully created
To list the target(s), use the list-target subcommand of the itadm command:
root@solaris:~# itadm list-target TARGET NAME STATE SESSIONS iqn.1986-03.com.sun:02:897fd011-8b3d-cf2b-fc1d-a010bd97d035 online 0
Configuring the Target for Discovery
Finally, you need to configure the target so it can be discovered by initiators:
root@solaris:~# devfsadm -i iscsi
