ssm man page

ssm — System Storage Manager: a single tool to manage your storage

Synopsis

ssm [-h] [--version] [-v] [-f] [-b BACKEND] [-n] {check,resize,create,list,add,remove,snapshot,mount} ...

ssm create [-h] [-s SIZE] [-n NAME] [--fstype FSTYPE] [-r LEVEL] [-I STRIPESIZE] [-i STRIPES] [-p POOL] [-e [{luks,plain}]] [device [device ...]] [mount]

ssm list [-h] [{volumes,vol,dev,devices,pool,pools,fs,filesystems,snap,snapshots}]

ssm remove [-h] [-a] [items [items ...]]

ssm resize [-h] [-s SIZE] volume [device [device ...]]

ssm check [-h] device [device ...]

ssm snapshot [-h] [-s SIZE] [-d DEST | -n NAME] volume

ssm add [-h] [-p POOL] device [device ...]

ssm mount [-h] [-o Options] volume directory

Description

System Storage Manager provides easy to use command line interface to manage your storage using various technologies like lvm, btrfs, encrypted volumes and more.

In more sophisticated enterprise storage environments, management with Device Mapper (dm), Logical Volume Manager (LVM), or Multiple Devices (md) is becoming increasingly more difficult. With file systems added to the mix, the number of tools needed to configure and manage storage has grown so large that it is simply not user friendly. With so many options for a system administrator to consider, the opportunity for errors and problems is large.

The btrfs administration tools have shown us that storage management can be simplified, and we are working to bring that ease of use to Linux filesystems in general.

Options

-h, --help
show this help message and exit
--version
show program's version number and exit
-v, --verbose
Show aditional information while executing.
-f, --force
Force execution in the case where ssm has some doubts or questions.
-b BACKEND, --backend BACKEND
Choose backend to use. Currently you can choose from (lvm,btrfs,crypt).
-n, --dry-run
Dry run. Do not do anything, just parse the command line options and gather system information if necessary. Note that with this option ssm will not perform all the check as some of them are done by the backends themselves. This option is mainly used for debugging purposes.

System Storage Manager Commands

Introduction

System Storage Manager have several commands you can specify on the command line as a first argument to the ssm. They all have specific use and its own arguments, but global ssm arguments are propagated to all commands.

Create command

ssm create [-h] [-s SIZE] [-n NAME] [--fstype FSTYPE] [-r LEVEL] [-I STRIPESIZE] [-i STRIPES] [-p POOL] [-e [{luks,plain}]] [device [device ...]] [mount]

This command creates a new volume with defined parameters. If device is provided it will be used to create a volume, hence it will be added into the pool prior the volume creation (See Add command section). More devices can be used to create a volume.

If the device is already used in the different pool, then ssm will ask you whether you want to remove it from the original pool. If you decline, or the removal fails, then the volume creation fails if the SIZE was not provided. On the other hand, if the SIZE is provided and some devices can not be added to the pool the volume creation might succeed if there is enough space in the pool.

POOL name can be specified as well. If the pool exists new volume will be created from that pool (optionally adding device into the pool). However if the POOL does not exist ssm will attempt to create a new pool with provided device and then create a new volume from this pool. If --backend argument is omitted, the default ssm backend will be used. Default backend is lvm.

ssm also supports creating RAID configuration, however some back-ends might not support all the levels, or it might not support RAID at all. In this case, volume creation will fail.

If mount point is provided ssm will attempt to mount the volume after it is created. However it will fail if mountable file system is not present on the volume.

-h, --help
show this help message and exit
-s SIZE, --size SIZE
Gives the size to allocate for the new logical volume A size suffix K|k, M|m, G|g, T|t, P|p, E|e can be used to define 'power of two' units. If no unit is provided, it defaults to kilobytes. This is optional if if not given maximum possible size will be used.
-n NAME, --name NAME
The name for the new logical volume. This is optional and if omitted, name will be generated by the corresponding backend.
--fstype FSTYPE
Gives the file system type to create on the new logical volume. Supported file systems are (ext3, ext4, xfs, btrfs). This is optional and if not given file system will not be created.
-r LEVEL, --raid LEVEL
Specify a RAID level you want to use when creating a new volume. Note that some backends might not implement all supported RAID levels. This is optional and if no specified, linear volume will be created. You can choose from the following list of supported levels (0,1,10).
-I STRIPESIZE, --stripesize STRIPESIZE
Gives the number of kilobytes for the granularity of stripes. This is optional and if not given, backend default will be used. Note that you have to specify RAID level as well.
-i STRIPES, --stripes STRIPES
Gives the number of stripes. This is equal to the number of physical volumes to scatter the logical volume. This is optional and if stripesize is set and multiple devices are provided stripes is determined automatically from the number of devices. Note that you have to specify RAID level as well.
-p POOL, --pool POOL
Pool to use to create the new volume.
-e [{luks,plain}], --encrypt [{luks,plain}]
Create encrpted volume. Extension to use can be specified.

List command

ssm list [-h] [{volumes,vol,dev,devices,pool,pools,fs,filesystems,snap,snapshots}]

List informations about all detected devices, pools, volumes and snapshots found in the system. list command can be used either alone to list all the information, or you can request specific section only.

Following sections can be specified:

{volumes | vol}
List information about all volumes found in the system.
{devices | dev}
List information about all devices found in the system. Some devices are intentionally hidden, like for example cdrom, or DM/MD devices since those are actually listed as volumes.
{pools | pool}
List information about all pools found in the system.
{filesystems | fs}
List information about all volumes containing filesystems found in the system.
{snapshots | snap}
List information about all snapshots found in the system. Note that some back-ends does not support snapshotting and some can not distinguish between snapshot and regular volume. in this case ssm will try to recognize volume name in order to identify snapshot, but if the ssm regular expression does not match the snapshot pattern, this snapshot will not be recognized.
-h, --help
show this help message and exit

Remove command

ssm remove [-h] [-a] [items [items ...]]

This command removes item from the system. Multiple items can be specified. If the item can not be removed for some reason, it will be skipped.

item can represent:

device
Remove device from the pool. Note that this can not be done in some cases where the device is used by pool. You can use -f argument to force removal. If the device does not belong to any pool, it will be skipped.
pool
Remove the pool from the system. This will also remove all volumes created from that pool.
volume
Remove the volume from the system. Note that this will fail if the volume is mounted and it can not be forced with -f.
-h, --help
show this help message and exit
-a, --all
Remove all pools in the system.

Resize command

ssm resize [-h] [-s SIZE] volume [device [device ...]]

Change size of the volume and file system. If there is no file system only the volume itself will be resized. You can specify device to add into the volume pool prior the resize. Note that device will only be added into the pool if the volume size is going to grow.

If the device is already used in the different pool, then ssm will ask you whether you want to remove it from the original pool.

In some cases file system has to be mounted in order to resize. This will be handled by ssm automatically by mounting the volume temporarily.

Note that resizing btrfs subvolume is not supported, only the whole file system can be resized.

-h, --help
show this help message and exit
-s SIZE, --size SIZE
New size of the volume. With the + or - sign the value is added to or subtracted from the actual size of the volume and without it, the value will be set as the new volume size. A size suffix of [k|K] for kilobytes, [m|M] for megabytes, [g|G] for gigabytes, [t|T] for terabytes or [p|P] for petabytes is optional. If no unit is provided the default is kilobytes.

Check command

ssm check [-h] device [device ...]

Check the file system consistency on the volume. You can specify multiple volumes to check. If there is no file system on the volume, this volume will be skipped.

In some cases file system has to be mounted in order to check the file system This will be handled by ssm automatically by mounting the volume temporarily.

-h, --help
show this help message and exit

Snapshot command

ssm snapshot [-h] [-s SIZE] [-d DEST | -n NAME] volume

Take a snapshot of existing volume. This operation will fail if back-end which the volume belongs to does not support snapshotting. Note that you can not specify both NAME and DESC since those options are mutually exclusive.

In some cases file system has to be mounted in order to take a snapshot of the volume. This will be handled by ssm automatically by mounting the volume temporarily.

-h, --help
show this help message and exit
-s SIZE, --size SIZE
Gives the size to allocate for the new snapshot volume A size suffix K|k, M|m, G|g, T|t, P|p, E|e can be used to define 'power of two' units. If no unit is provided, it defaults to kilobytes. This is option and if not give, the size will be determined automatically.
-d DEST, --dest DEST
Destination of the snapshot specified with absolute path to be used for the new snapshot. This is optional and if not specified default backend policy will be performed.
-n NAME, --name NAME
Name of the new snapshot. This is optional and if not specified default backend policy will be performed.

Add command

ssm add [-h] [-p POOL] device [device ...]

This command adds device into the pool. The device will not be added if it's already part of different pool by default, but user will be asked whether to remove the device from it's pool. When multiple devices are provided, all of them are added into the pool. If one of the devices can not be added into the pool for any reason, add command will fail. If no pool is specified, default pool will be chosen. In the case of non existing pool, it will be created using provided devices.

-h, --help
show this help message and exit
-p POOL, --pool POOL
Pool to add device into. If not specified the default pool is used.

Mount command

ssm mount [-h] [-o Options] volume directory

This command will mount the volume at specified directory. The volume can be specified in the same way as with mount(8), however in addition one can also specify volume in the format as it appear in the ssm list table.

For example, instead of finding out what the device and subvolume id of the btrfs subvolume "btrfs_pool:vol001" is in order to mount it, on can simply call ssm mount btrfs_pool:vol001 /mnt/test.

One can also specify Options in the same way as with mount(8).

-h, --help
show this help message and exit
-o OPTIONS, --options OPTIONS
Options are specified with a -o flag followed by a comma separated string of options. This option is equivalent to the same mount(8) option.

Back-Ends

Introduction

Ssm aims to create unified user interface for various technologies like Device Mapper (dm), Btrfs file system, Multiple Devices (md) and possibly more. In order to do so we have a core abstraction layer in ssmlib/main.py. This abstraction layer should ideally know nothing about the underlying technology, but rather comply with device, pool and volume abstraction.

Various backends can be registered in ssmlib/main.py in order to handle specific storage technology implementing methods like create, snapshot, or remove volumes and pools. The core will then call these methods to manage the storage without needing to know what lies underneath it. There are already several backends registered in ssm.

Btrfs backend

Btrfs is the file system with many advanced features including volume management. This is the reason why btrfs is handled differently than other conventional file systems in ssm. It is used as a volume management back-end.

Pools, volumes and snapshots can be created with btrfs backend and here is what it means from the btrfs point of view:

pool
Pool is actually a btrfs file system itself, because it can be extended by adding more devices, or shrink by removing devices from it. Subvolumes and snapshots can also be created. When the new btrfs pool should be created ssm simply creates a btrfs file system, which means that every new btrfs pool has one volume of the same name as the pool itself which can not be removed without removing the entire pool. Default btrfs pool name is btrfs_pool.

When creating new btrfs pool, the name of the pool is used as the file system label. If there is already existing btrfs file system in the system without a label, btrfs pool name will be generated for internal use in the following format "btrfs_{device base name}".

Btrfs pool is created when create or add command is used with devices specified and non existing pool name.
volume
Volume in btrfs back-end is actually just btrfs subvolume with the exception of the first volume created on btrfs pool creation, which is the file system itself. Subvolumes can only be created on btrfs file system when it is mounted, but user does not have to worry about that since ssm will automatically mount the file system temporarily in order to create a new subvolume.

Volume name is used as subvolume path in the btrfs file system and every object in this path must exists in order to create a volume. Volume name for internal tracking and for representing to the user is generated in the format "{pool_name}:{volume name}", but volumes can be also referenced with its mount point.

Btrfs volumes are only shown in the list output, when the file system is mounted, with the exception of the main btrfs volume - the file system itself.

Also note that btrfs volumes and subvolumes can not be resized. This is mainly limitation of the btrfs tools which currently does not work reliably.

New btrfs volume can be created with create command.
snapshot
Btrfs file system support subvolume snapshotting, so you can take a snapshot of any btrfs volume in the system with ssm. However btrfs does not distinguish between subvolumes and snapshots, because snapshot actually is just a subvolume with some block shared with different subvolume. It means, that ssm is not able to recognize btrfs snapshot directly, but instead it is trying to recognize special name format of the btrfs volume. However, if the NAME is specified when creating snapshot which does not match the special pattern, snapshot will not be recognized by the ssm and it will be listed as regular btrfs volume.

New btrfs snapshot can be created with snapshot command.
device
Btrfs does not require any special device to be created on.

Lvm backend

Pools, volumes and snapshots can be created with lvm, which pretty much match the lvm abstraction.

pool
Lvm pool is just volume group in lvm language. It means that it is grouping devices and new logical volumes can be created out of the lvm pool. Default lvm pool name is lvm_pool.

Lvm pool is created when create or add command is used with devices specified and non existing pool name.
volume
Lvm volume is just logical volume in lvm language. Lvm volume can be created wit create command.
snapshot
Lvm volumes can be snapshotted as well. When a snapshot is created from the lvm volume, new snapshot volume is created, which can be handled as any other lvm volume. Unlike btrfs lvm is able to distinguish snapshot from regular volume, so there is no need for a snapshot name to match special pattern.
device
Lvm requires physical device to be created on the device, but with ssm this is transparent for the user.

Crypt backend

Crypt backend in ssm uses cryptsetup and dm-crypt target to manage encrypted volumes. Crypt backend can be used as a regular backend for creating encrypted volumes on top of regular block devices, or even other volumes (lvm or md volumes for example). Or it can be used to create encrypted lvm volumes right away in a single step.

Only volumes can be created with crypt backend. This backend does not support pooling and does not require special devices.

pool
Crypt backend does not support pooling it is not possible to create crypt pool or add a device into a pool.
volume
Volume in crypt backend is the volume created by dm-crypt which represent the data on the original encrypted device in unencrypted form. Crypt backend does not support pooling, so only one device can be used to create crypt volume. It also does not support raid or any device concatenation.

Currently two modes, or extensions are supported luks and plain. Luks is used by default.For more information about the extensions please see cryptsetup manual page.
snapshot
Crypt backend does not support snapshotting, however if the encrypted volume is created on top of the lvm volume, the lvm volume itself can be snapshotted. The snapshot can be then opened by using cryptsetup. It is possible that this might change in the future so that ssm will be able to activate the volume directly without the extra step.
device
Crypt backend does not require any special device to be created on.

MD backend

MD backend in ssm is currently limited to only gather the information about MD volumes in the system. You can not create or manage MD volumes or pools, but it will be extended in the future.

Examples

List system storage information:

# ssm list

List all pools in the system:

# ssm list pools

Create a new 100GB volume with default lvm backend using /dev/sda and /dev/sdb with xfs file system:

# ssm create --size 100G --fs xfs /dev/sda /dev/sdb

Create a new volume with btrfs backend using /dev/sda and /dev/sdb and let the volume to be RAID 1:

# ssm -b btrfs create --raid 1 /dev/sda /dev/sdb

Using lvm backend create a RAID 0 volume with devices /dev/sda and /dev/sdb with 128kB stripe size, ext4 file system and mount it on /home:

# ssm create --raid 0 --stripesize 128k /dev/sda /dev/sdb /home

Extend btrfs volume btrfs_pool by 500GB and use /dev/sdc and /dev/sde to cover the resize:

# ssm resize -s +500G btrfs_pool /dev/sdc /dev/sde

Shrink volume /dev/lvm_pool/lvol001 by 1TB:

# ssm resize -s-1t /dev/lvm_pool/lvol001

Remove /dev/sda device from the pool, remove the btrfs_pool pool and also remove the volume /dev/lvm_pool/lvol001:

# ssm remove /dev/sda btrfs_pool /dev/lvm_pool/lvol001

Take a snapshot of the btrfs volume btrfs_pool:my_volume:

# ssm snapshot btrfs_pool:my_volume

Add devices /dev/sda and /dev/sdb into the btrfs_pool pool:

# ssm add -p btrfs_pool /dev/sda /dev/sdb

Mount btrfs subvolume btrfs_pool:vol001 on /mnt/test:

# ssm mount btrfs_pool:vol001 /mnt/test

Environment Variables

SSM_DEFAULT_BACKEND
Specify which backend will be used by default. This can be overridden by specifying -b or --backend argument. Currently only lvm and btrfs is supported.
SSM_LVM_DEFAULT_POOL
Name of the default lvm pool to be used if -p or --pool argument is omitted.
SSM_BTRFS_DEFAULT_POOL
Name of the default btrfs pool to be used if -p or --pool argument is omitted.
SSM_PREFIX_FILTER
When this is set ssm will filter out all devices, volumes and pools which name does not start with this prefix. It is used mainly in ssm test suite to make sure that we do not scramble local system configuration.

Licence

(C)2011 Red Hat, Inc., Lukas Czerner <lczerner@redhat.com>

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 2 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see <http://www.gnu.org/licenses/>.

Requirements

Python 2.6 or higher is required to run this tool. System Storage Manager can only be run as root since most of the commands requires root privileges.

There are other requirements listed bellow, but note that you do not necessarily need all dependencies for all backends, however if some of the tools required by the backend is missing, the backend would not work.

Python modules

·
os
·
re
·
sys
·
stat
·
argparse
·
datetime
·
threading
·
subprocess

System tools

·
tune2fs
·
fsck.SUPPORTED_FS
·
resize2fs
·
xfs_db
·
xfs_check
·
xfs_growfs
·
mkfs.SUPPORTED_FS
·
which
·
mount
·
blkid
·
wipefs

Lvm backend

·
lvm2 binaries

Btrfs backend

·
btrfs progs

Crypt backend

·
dmsetup
·
cryptsetup

Availability

System storage manager is available from http://storagemanager.sourceforge.net. You can subscribe to storagemanager-devel@lists.sourceforge.net to follow the current development.

Author

Lukáš Czerner <lczerner@redhat.com>

Info

October 02, 2013 0.4 System Storage Manager