Two command line tools are provided to interface to the Regional Controller via the bash command line:
rc_cli - which allows you to perform individual API calls
rc_loaddata - used to do batch initialization of the RC database
Both tools require Python 3 be installed on your machine.
RC_CLI - Perform individual API calls
The options and arguments to the rc_cli command may be displayed by invoking it with –help:
$ ./rc_cli --help
usage: rc_cli [-h] [-H HOST] [-u USER] [-p PASSWORD] [-s SORT] [-F] [-Y]
obj op [args [args ...]]
Perform API calls against a Regional Controller.
positional arguments:
obj The object to manipulate
op The operation to perform on the object
args Extra arguments for the operation
optional arguments:
-h, --help show this help message and exit
-H HOST the host of the regional controller (default localhost)
-u USER the username to use (default admin)
-p PASSWORD the password to use
-s SORT the field to sort list views by
-F forcibly delete a POD (use with "pod delete")
-Y display YAML columns in the output of list commands
rc_cli will first login to the RC using the host/user/password combination you specify, and then will perform the operation (specified by op) on the object (obj) listed. The user must have the appropriate permissions in the database to perform the requested operation. For example, in order to run the pod create command, the user must have a user role that includes the create-pod or create-* role attributes.
Objects correspond to the objects that the RC controls; e.g. blueprint, edgesite, hardware, node, pod, and region. Operations are create, delete, list, and show.
The create operation takes as arguments the names of one or more YAML files containing the objects to be created. For successfully created objects, it will print the URL of the newly created object, which contains the object’s UUID. The YAML files provided must contain only YAML descriptions of the appropriate type specified on the command line. The format of these YAML objects can be determined by referencing Regional Controller API Documentation.
$ ./rc_cli -u admin -p admin123 hardware create hardware.yaml https://localhost/api/v1/hardware/36a72db1-f2d9-424b-b94e-b72ca7a50cfe
delete takes as arguments a list of UUIDs of the objects to be deleted. If the specified objects are deleted, it produces no output.
list takes no arguments and just lists all objects of the specified type in the RC. The returned list will be sorted based on UUID, unless a different sort column is passed (via the -s sort option). The list operation will not display the YAML column of any object that has a YAML column unless the -Y option is provided.
$ ./rc_cli -u admin -p admin123 region list UUID Name Description Parent ------------------------------------ --------- --------------------------------------------------------- ------------------------------------ 00000000-0000-0000-0000-000000000000 Universal The Parent of all Regions 00000000-0000-0000-0000-000000000000 3e2c604a-6ee7-4694-bbdc-9d19b146c6a0 Lab The main region that all nodes in the lab will all be in. 00000000-0000-0000-0000-000000000000 efa82a42-c3b0-49c7-9db3-bbcd85927bbf OEzone A special region for all OpenEdge hardware in the lab. 00000000-0000-0000-0000-000000000000
show takes as arguments a list of UUIDs of the objects to show the details of.
$ ./rc_cli -u admin -p admin123 hardware show c1dfa1ac-53e0-11e9-86c2-c313482f1fdb --- description: Standard Dell configuration for Rover/Unicycle name: Dell PowerEdge R740 uuid: c1dfa1ac-53e0-11e9-86c2-c313482f1fdb yaml: cpu: 2x22 Cores @ 2.1GHz Skylake 6152 CPU disk: - 4x480G SSD - 6x2.4T HDD lom: 4x10G Intel 710 nic: - 2x1G LOM Intel 5xx - 2x25G PCI3 Intel 710 ps: 2 ram: 12x32GB
RC_LOADDATA - Batch Load Data into the RC DB
The options and arguments to the rc_loaddata command may be displayed by invoking it with –help:
usage: rc_loaddata [-h] [-H HOST] [-u USER] [-p PASSWORD] [-s {hardware,regions,nodes,edgesites,blueprints,pods}] [-A] [-D] [-M] YAML [YAML ...] Add/delete objects from a Regional Controller. The objects are retrieved from one or more YAML files. positional arguments: YAML YAML files containing RC objects to be added/deleted/matched optional arguments: -h, --help show this help message and exit -H HOST the host of the regional controller (default localhost) -u USER the username to use (default admin) -p PASSWORD the password to use -s {hardware,regions,nodes,edgesites,blueprints,pods} only process the named section of the YAMLs (default all) -A add new entries in YAMLs to the RC -D delete entries in YAMLs from the RC -M match entries in YAMLs to the RC
The purpose of rc_loaddata is to perform operations on large number of RC objects at once. The operation to perform is specified by the -A/-D/-M options, only one of which may be provided.
-A - add all entries in the YAML files to the RC database.
-D - delete all entries in the YAML files from the RC database.
-M - match entries between the YAML files and the RC database. That is, make the RC database identical to the contents of the catenated YAML files.
You can restrict the objects that the above operations are performed against by specifying one or more -s section options. Normally all sections are scanned.
When adding objects to the RC, the objects are added in a “bottom up” order; that is: ‘hardware’, ‘regions’, ‘nodes’, ‘edgesites’, ‘blueprints’, and finally ‘pods’. When deleting objects, the reverse order (top down) is used.
As with rc_cli, rc_loaddata will first login to the RC using the host/user/password combination you specify.
The YAML files themselves consists of stanzas for any of the six sections needed, followed by named objects describing the objects of each type to be added/deleted/matched. For example:
hardware: Dell_740xd: uuid: 9897a008-71d4-11e9-8bda-0017f20dbff8 description: Dell 740xd hardware for the REC Blueprint yaml: todo: put hardware details here rack_layout: height: 2U regions: zone1: description: The main region that all nodes in the MT lab will all be in. edgesites: REC_OpenEdge1: uuid: 60ab1298-7769-11e9-92b3-373d9b2f2476 description: The first OE REC cluster nodes: [ aknode201, aknode202, aknode203, aknode204, aknode205 ] regions: [ zone1 ] blueprints: nodes: aknode201: hardware: Nokia_OE19 yaml: oob_ip: 172.26.16.201 rack_location: { name: Row4_Rack6, slot: 2, unit: 1 }