Case: you want to create a large number of volumes
Problem: you do not want to create them all simultaneously to not overload unnecessarily the storage backend.
Solution 1: you write a quick for loop in shell that calls cinder create... for each volume and then sleep before creating the next one. This kind of quick shell script works fine, but has obvious limitations: it does not check for errors, it does not wait for the previous volume to be created before continuing, it merely sleeps an arbitrary amount of time then goes on.
Solution 2: write a proper shell script or other proper python script that loops through the creation of each volume and waits each time for the volume to be created before moving on to the next.
Synopsis: Easily create a series of volumes with the same properties one after the other.
Usage:
usage: create_volumes_sequentially.py
[-h] [--os-username username]
[--os-password password]
[--os-tenant-name tenant_name]
[--os-auth-url auth_URL]
[--name template] --size GB_size
--volume-type voltype_name
[--poll-frequency seconds]
--number-volumes number
[--start-index index]
[--image-id ID] [--snapshot-id ID]
[--source-volid ID]
[--availability-zone zone]
[--version]
Create multiple similar cinder volumes one after the other.
optional arguments:
-h, --help
show this help message and exit
--os-username username
The OpenStack user's username to use to connect to
OpenStack Keystone
--os-password password
The OpenStack user's password to use to connect to
OpenStack Keystone
--os-tenant-name tenant_name
The OpenStack user's tenant name to use to connect to
OpenStack Keystone
--os-auth-url auth_URL
The OpenStack Keystone authentication URL
--name template
The volume name template to use, after which we will
add a dash and an index number. Defaults to 'volume'
--size GB_size
The size in GB of each volume to create
--volume-type voltype_name
The cinder volume-type name to use for volume creation
--poll-frequency seconds
The frequency in seconds at which to poll cinder for
completion. Defaults to 30.
--number-volumes number
The number of cinder volumes to create
--start-index index
The index to start with for volume creation, this
index is incremented at each volume created. Defaults
to 1.
--image-id ID
Create volume from image id. Defaults to None
--snapshot-id ID
Create volume from snapshot id. Default None
--source-volid ID
Create volume from volume id. Defaults to None.
--availability-zone zone
Availability zone for volume. Defaults to None.
--version
show program's version number and exit
Copyright (c) 2015 Kibin Labs Pte Ltd. Released under MIT license.
Example:
# ./create_volumes_sequentially.py --size 1 --volume-type cinder_iSCSI --number-volumes 3 --start-index 3
2015-02-13 09:49:31,079 [INFO] Initiated creation of volume volume-3 with volume-id 84d4ce85-3461-480d-b8bf-b910af41257b
2015-02-13 09:49:32,159 [INFO] Waiting for volume volume-3. Current state is 'creating'. Sleeping 30 seconds.
2015-02-13 09:50:03,872 [INFO] Successfully created volume volume-3
2015-02-13 09:50:06,480 [INFO] Initiated creation of volume volume-4 with volume-id 223a1f0d-2c88-4436-b517-2555b002db44
2015-02-13 09:50:07,825 [INFO] Waiting for volume volume-4. Current state is 'creating'. Sleeping 30 seconds.
2015-02-13 09:50:39,071 [INFO] Successfully created volume volume-4
2015-02-13 09:50:40,371 [INFO] Initiated creation of volume volume-5 with volume-id 139c8663-c01e-4994-a46e-16311f3c3f72
2015-02-13 09:50:41,598 [INFO] Waiting for volume volume-5. Current state is 'creating'. Sleeping 30 seconds.
2015-02-13 09:51:12,822 [INFO] Successfully created volume volume-5