cinder/doc/source/configuration/block-storage/drivers/dell-storagecenter-driver.rst

Dell EMC SC Series Fibre Channel and iSCSI drivers

The Dell EMC Storage Center volume driver interacts with configured Storage Center arrays.

The Dell EMC Storage Center driver manages a Storage Center array via the Dell EMC Storage Manager (DSM) Data Collector or by directly connecting to the Storage Center at the cost of replication and Live Volume functionality. Also note that the directly connecting to the Storage Center is only supported with Storage Center OS 7.1.1 or later. Any version of Storage Center OS supported by DSM is supported if connecting via the Data Collector.

Driver configuration settings and Storage Center options are defined in the cinder.conf file.

Prerequisites:

• Storage Center OS version 7.1.1 or later and OpenStack Ocata or later must be used if connecting directly to the Storage Center.
• Dell EMC Storage Manager 2015 R1 or later if connecting through DSM.

Supported operations

The Dell EMC Storage Center volume driver provides the following Cinder volume operations:

• Create, delete, attach (map), and detach (unmap) volumes.
• Create, list, and delete volume snapshots.
• Create a volume from a snapshot.
• Copy an image to a volume.
• Copy a volume to an image.
• Clone a volume.
• Extend a volume.
• Create, delete, list and update a consistency group.
• Create, delete, and list consistency group snapshots.
• Manage an existing volume.
• Replication (Requires DSM.)
• Failover-host for replicated back ends. (Requires DSM.)
• Create a replication using Live Volume. (Requires DSM.)

Extra spec options

Volume type extra specs can be used to enable a variety of Dell EMC Storage Center options. Selecting Storage Profiles, Replay Profiles, enabling replication, replication options including Live Volume and Active Replay replication. (Replication options are available when connected via DSM.)

Storage Profiles control how Storage Center manages volume data. For a given volume, the selected Storage Profile dictates which disk tier accepts initial writes, as well as how data progression moves data between tiers to balance performance and cost. Predefined Storage Profiles are the most effective way to manage data in Storage Center.

By default, if no Storage Profile is specified in the volume extra specs, the default Storage Profile for the user account configured for the Block Storage driver is used. The extra spec key storagetype:storageprofile with the value of the name of the Storage Profile on the Storage Center can be set to allow to use Storage Profiles other than the default.

For ease of use from the command line, spaces in Storage Profile names are ignored. As an example, here is how to define two volume types using the High Priority and Low Priority Storage Profiles:

$ openstack volume type create "GoldVolumeType"
$ openstack volume type set --property storagetype:storageprofile=highpriority "GoldVolumeType"
$ openstack volume type create "BronzeVolumeType"
$ openstack volume type set --property storagetype:storageprofile=lowpriority "BronzeVolumeType"

Replay Profiles control how often the Storage Center takes a replay of a given volume and how long those replays are kept. The default profile is the daily profile that sets the replay to occur once a day and to persist for one week.

The extra spec key storagetype:replayprofiles with the value of the name of the Replay Profile or profiles on the Storage Center can be set to allow to use Replay Profiles other than the default daily profile.

As an example, here is how to define a volume type using the hourly Replay Profile and another specifying both hourly and the default daily profile:

$ openstack volume type create "HourlyType"
$ openstack volume type set --property storagetype:replayprofile=hourly "HourlyType"
$ openstack volume type create "HourlyAndDailyType"
$ openstack volume type set --property storagetype:replayprofiles=hourly,daily "HourlyAndDailyType"

Note the comma separated string for the HourlyAndDailyType.

Replication for a given volume type is enabled via the extra spec replication_enabled.

To create a volume type that specifies only replication enabled back ends:

$ openstack volume type create "ReplicationType"
$ openstack volume type set --property replication_enabled='<is> True' "ReplicationType"

Extra specs can be used to configure replication. In addition to the Replay Profiles above, replication:activereplay can be set to enable replication of the volume's active replay. And the replication type can be changed to synchronous via the replication_type extra spec can be set.

To create a volume type that enables replication of the active replay:

$ openstack volume type create "ReplicationType"
$ openstack volume type key --property replication_enabled='<is> True' "ReplicationType"
$ openstack volume type key --property replication:activereplay='<is> True' "ReplicationType"

To create a volume type that enables synchronous replication :

$ openstack volume type create "ReplicationType"
$ openstack volume type key --property replication_enabled='<is> True' "ReplicationType"
$ openstack volume type key --property replication_type='<is> sync' "ReplicationType"

To create a volume type that enables replication using Live Volume:

$ openstack volume type create "ReplicationType"
$ openstack volume type key --property replication_enabled='<is> True' "ReplicationType"
$ openstack volume type key --property replication:livevolume='<is> True' "ReplicationType"

If QOS options are enabled on the Storage Center they can be enabled via extra specs. The name of the Volume QOS can be specified via the storagetype:volumeqos extra spec. Likewise the name of the Group QOS to use can be specified via the storagetype:groupqos extra spec. Volumes created with these extra specs set will be added to the specified QOS groups.

To create a volume type that sets both Volume and Group QOS:

$ openstack volume type create "StorageCenterQOS"
$ openstack volume type key --property 'storagetype:volumeqos'='unlimited' "StorageCenterQOS"
$ openstack volume type key --property 'storagetype:groupqos'='limited' "StorageCenterQOS"

Data reduction profiles can be specified in the storagetype:datareductionprofile extra spec. Available options are None, Compression, and Deduplication. Note that not all options are available on every Storage Center.

To create volume types that support no compression, compression, and deduplication and compression respectively:

$ openstack volume type create "NoCompressionType"
$ openstack volume type key --property 'storagetype:datareductionprofile'='None' "NoCompressionType"
$ openstack volume type create "CompressedType"
$ openstack volume type key --property 'storagetype:datareductionprofile'='Compression' "CompressedType"
$ openstack volume type create "DedupType"
$ openstack volume type key --property 'storagetype:datareductionprofile'='Deduplication' "DedupType"

Note: The default is no compression.

iSCSI configuration

Use the following instructions to update the configuration file for iSCSI:

default_volume_type = delliscsi
enabled_backends = delliscsi

[delliscsi]
# Name to give this storage back-end
volume_backend_name = delliscsi
# The iSCSI driver to load
volume_driver = cinder.volume.drivers.dell_emc.sc.storagecenter_iscsi.SCISCSIDriver
# IP address of the DSM or the Storage Center if attaching directly.
san_ip = 172.23.8.101
# DSM user name
san_login = Admin
# DSM password
san_password = secret
# The Storage Center serial number to use
dell_sc_ssn = 64702

# ==Optional settings==

# The DSM API port
dell_sc_api_port = 3033
# Server folder to place new server definitions
dell_sc_server_folder = devstacksrv
# Volume folder to place created volumes
dell_sc_volume_folder = devstackvol/Cinder

Fibre Channel configuration

Use the following instructions to update the configuration file for fibre channel:

default_volume_type = dellfc
enabled_backends = dellfc

[dellfc]
# Name to give this storage back-end
volume_backend_name = dellfc
# The FC driver to load
volume_driver = cinder.volume.drivers.dell_emc.sc.storagecenter_fc.SCFCDriver

# IP address of the DSM or the Storage Center if attaching directly.
san_ip = 172.23.8.101
# DSM user name
san_login = Admin
# DSM password
san_password = secret
# The Storage Center serial number to use
dell_sc_ssn = 64702

# ==Optional settings==

# The DSM API port
dell_sc_api_port = 3033
# Server folder to place new server definitions
dell_sc_server_folder = devstacksrv
# Volume folder to place created volumes
dell_sc_volume_folder = devstackvol/Cinder

Dual DSM

It is possible to specify a secondary DSM to use in case the primary DSM fails.

Configuration is done through the cinder.conf. Both DSMs have to be configured to manage the same set of Storage Centers for this backend. That means the dell_sc_ssn and any Storage Centers used for replication or Live Volume.

Add network and credential information to the backend to enable Dual DSM.

[dell]
# The IP address and port of the secondary DSM.
secondary_san_ip = 192.168.0.102
secondary_sc_api_port = 3033
# Specify credentials for the secondary DSM.
secondary_san_login = Admin
secondary_san_password = secret

The driver will use the primary until a failure. At that point it will attempt to use the secondary. It will continue to use the secondary until the volume service is restarted or the secondary fails at which point it will attempt to use the primary.

Note: Requires two DSM Data Collectors.

Replication configuration

Add the following to the back-end specification to specify another Storage Center to replicate to.

[dell]
replication_device = target_device_id: 65495, qosnode: cinderqos

The target_device_id is the SSN of the remote Storage Center and the qosnode is the QoS Node setup between the two Storage Centers.

Note that more than one replication_device line can be added. This will slow things down, however.

A volume is only replicated if the volume is of a volume-type that has the extra spec replication_enabled set to <is> True.

Warning: replication_device requires DSM. If this is on a backend that is directly connected to the Storage Center the driver will not load as it is unable to meet the replication requirement.

Replication notes

This driver supports both standard replication and Live Volume (if supported and licensed). The main difference is that a VM attached to a Live Volume is mapped to both Storage Centers. In the case of a failure of the primary Live Volume still requires a failover-host to move control of the volume to the second controller.

Existing mappings should work and not require the instance to be remapped but it might need to be rebooted.

Live Volume is more resource intensive than replication. One should be sure to plan accordingly.

Failback

The failover-host command is designed for the case where the primary system is not coming back. If it has been executed and the primary has been restored it is possible to attempt a failback.

Simply specify default as the backend_id.

$ cinder failover-host cinder@delliscsi --backend_id default

Non trivial heavy lifting is done by this command. It attempts to recover as best it can but if things have diverged too far it can only do so much. It is also a one time only command so do not reboot or restart the service in the middle of it.

Failover and failback are significant operations under OpenStack Cinder. Be sure to consult with support before attempting.

Server type configuration

This option allows one to set a default Server OS type to use when creating a server definition on the Dell EMC Storage Center.

When attaching a volume to a node the Dell EMC Storage Center driver creates a server definition on the storage array. This definition includes a Server OS type. The type used by the Dell EMC Storage Center cinder driver is "Red Hat Linux 6.x". This is a modern operating system definition that supports all the features of an OpenStack node.

Add the following to the back-end specification to specify the Server OS to use when creating a server definition. The server type used must come from the drop down list in the DSM.

[dell]
dell_server_os = 'Red Hat Linux 7.x'

Note that this server definition is created once. Changing this setting after the fact will not change an existing definition. The selected Server OS does not have to match the actual OS used on the node.

Excluding a domain

This option excludes a list of Storage Center ISCSI fault domains from the ISCSI properties returned by the initialize_connection call. This only applies to the ISCSI driver.

Add the excluded_domain_ips option into the backend config for several fault domains to be excluded. This option takes a comma separated list of Target IP addresses listed under the fault domain. Older versions of DSM (EM) may list this as the Well Known IP Address.

Note that the included_domain_ips takes precedance over excluded_domain_ips. When included_domain_ips is not an empty list, the option excluded_domain_ips is ignored.

Add the following to the back-end specification to exclude the domains at 172.20.25.15 and 172.20.26.15.

[dell]
excluded_domain_ips=172.20.25.15, 172.20.26.15, 0:0:0:0:0:ffff:c0a8:15

Including domains

This option includes or will whitelist a list of Storage Center ISCSI fault domains from the ISCSI properties returned by the initialize_connection call. This only applies to the ISCSI driver.

Add the included_domain_ips option into the backend config for several default domains to be included or whitelisted. This option takes a comma separated list of Target IP addresses listed under the fault domain. Older versions of DSM (EM) may list this as the Well Known IP Address.

Note that the included_domain_ips takes precedance over excluded_domain_ips. When included_domain_ips is not an empty list, the option excluded_domain_ips is ignored.

Add the following to the back-end specification to include or whitelist the domains at 172.20.25.15 and 172.20.26.15.

[dell]
included_domain_ips=172.20.25.15, 172.20.26.15, 0:0:0:0:0:ffff:c0a8:15

Setting Dell EMC SC REST API timeouts

The user can specify timeouts for Dell EMC SC REST API calls.

To set the timeout for ASYNC REST API calls in seconds.

[dell]
dell_api_async_rest_timeout=15

To set the timeout for SYNC REST API calls in seconds.

[dell]
dell_api_sync_rest_timeout=30

Generally these should not be set without guidance from Dell EMC support.

Driver options

The following table contains the configuration options specific to the Dell EMC Storage Center volume driver.

cinder.volume.drivers.dell_emc.sc.storagecenter_common