openstack/openstack-manuals
Code Issues Proposed changes

Support for Cinder volume replication

Browse Source 
Documenting information on how cloud administrators
can adjust the cinder volume driver to enable
volume replication. This feature safeguards environments
from data loss.

Change-Id: I7ba72a14914cb1b0359e532e619ea36dfe1b73e0
Closes-Bug:#1360768
This commit is contained in:
Joseph Robinson 2015-09-03 14:56:16 +10:00
parent d6d10a5fb7
commit dd784ae0ea
1 changed files with 32 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

32
doc/admin-guide-cloud/source/blockstorage_over_subscription.rst
View File

@ -97,6 +97,38 @@ The above extra specs are used by the scheduler to find a back end that
supports thin provisioning, thick provisioning, or both to match the needs
of a specific volume type.
Volume replication extra specs
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
OpenStack Block Storage has the ability to create volume replicas.
Cloud administrators can define a storage policy that includes
replication by adjusting the cinder volume driver. Volume replication
for OpenStack Block Storage helps safeguard OpenStack environments from
data loss during disaster recovery.
To enable replication when creating volume types, configure the cinder
volume with ``capabilities:replication="<is> True"``.
Each volume created with the replication capability set to `True`
generates a copy of the volume on a storage back end.
One use case for replication involves an OpenStack cloud environment
installed accross two data centers located nearby each other. The
distance between the two data centers in this use case is the length of
a city.
At each data center, a cinder host supports the Block Storage service.
Both data centers include storage back ends.
Depending on the storage requirements, there can be one or two cinder
hosts. The cloud administrator accesses the
:file:`/etc/cinder/cinder.conf` configuration file and sets
``capabilities:replication="<is> True"``.
If one data center experiences a service failure, cloud administrators
can redeploy the VM. The VM will run using a replicated, backed up
volume on a host in the second data center.
Capacity filter
~~~~~~~~~~~~~~~