Name: nova-flavor-create
Owner: CERN Operations
Description: null
Created: 2015-05-09 19:55:19.0
Updated: 2015-05-09 21:47:58.0
Pushed: 2015-05-09 22:14:49.0
Homepage: null
Size: 128
Language: Python
GitHub Committers
User | Most Recent Commit | # Commits |
---|
Other Committers
User | Most Recent Commit | # Commits |
---|
It's a small tool that creates flavors in a cell environment.
It creates a flavor with the same row ID in all available cells.
The flavors in the children cells are always created as public.
In order to delete, disable or manage the the flavor with spefic tenants use
nova api in the parent cell.
To see the available options run:
$python cern-flavor-create -h
usage: cern-flavor-create [-h] [–ephemeral EPHEMERAL] [–swap SWAP]
[--rxtx_factor RXTX_FACTOR] [--is_public IS_PUBLIC]
[--config CONFIG]
name id ram disk vcpus
positional arguments:
name - Name of the new flavor
id - UniqueID (integer or UUID) for the new flavor. If specifying 'auto', a UUID will be generated as id
ram - Memory size in MB
disk - Disk size in GB
vcpus - Number of vcpus
optional arguments:
-h, –help - show this help message and exit
–ephemeral EPHEMERAL - Ephemeral space size in GB (default 0)
–swap SWAP - Swap space size in MB (default 0)
–rxtx_factor RXTX_FACTOR - RX/TX factor (default 1)
–is_public IS_PUBLIC - Make flavor accessible to the public (default False)
–config CONFIG - Configuration file
The configuration file should contain the DBs locations for all cells and
should be explicitly defined with the option –config
Example for an infrastructure with a parent cell and one child.
[default]
available_cells=cell_parent,cell_child
[cell_parent]
cell_type=parent
db_type=mysql
db_location=parent_db.cern.ch
db_port=3306
user=nova_user
password=db_parent_password
database=nova_parent
[cell_child]
cell_type=child
db_type=mysql
db_location=child_db.cern.ch
db_port=3306
user=nova_user
password=db_child_password
database=nova_child
To create a flavor:
./cern-flavor-create –config cern-flavor-create.conf my_flavor auto 2048 20 2
We use it in Icehouse and Juno.
Bugs? Oh, almost certainly.
This tool was written to be used in the CERN Cloud Infrastructure and it has been tested only in our environment.
Since it updates nova DB use it with extremely caution.