# # This script is setup to use the username and API provided as commandline arguments (as a simple way of keeping them out of the codebase) # RacksJS ships with coffeescript examples. # You can run these by installing coffee-script: # # sudo npm install -g coffee-script # # and then by running them with your API details: # # coffee nextgen_server.coffee -n API_USERNAME API_KEY # #request = require 'request' RacksJS = require '../dist/racks.js' new RacksJS { username: process.argv[3] apiKey: process.argv[4] verbosity: 5 # To auth against LON #endpoint: 'https://lon.identity.api.rackspacecloud.com/v2.0' }, (rs) -> log = console.log # You can change datacenters or network targets at any time. # Every call makes its own lookup via a .target() method in a _racksmeta object # which is appended to every instance of every product object # .target() is called before every HTTP request, so simply change the # rs.datacenter = 'ORD' # valid values are ORD, IAD, DFW, LON, AUS, HKG # # or the # rs.network = 'public' # valid values are public and internal # # at any time. #rs.cloudImages.images.assume('some_uuid_goes_here').members (out) -> # console.log out #console.log rs.cloudFiles.containers.assume('cdn.example.com')._racksmeta.target() # Utilies #rs.utils.rebuildLB 'LB_ID', (result) -> # if result # console.log 'done' #rs.nextgen.servers.all (reply) -> # log reply # Lets get to some cool stuff # CLOUD FILES: # # Stream a file to cloud files: # rs.cloudFiles.containers.assume('SOME_CONTAINER_NAME').upload { # stream: request("SOME_URL"), # path: 'myUploadedFilename.ext' # }, (reply) -> # console.log reply.statusCode # # Upload a file to a cloud files container with an assumed named: # rs.cloudFiles.containers.assume('CONTAINER_NAME').upload { # file: 'path_of_file_to_upload' # }, (reply) -> # console.log reply.statusCode # Enable a CORS header on a container # rs.cloudFiles.containers.assume('container_name').enableCORSHeader 'http://www.example.com http://stage.example.com', (reply) -> # console.log reply # rs.cloudFiles.containers.new { name: 'somenewTestContainer' } # CLOUD MONITORING #rs.cloudMonitoring.agents.assume('SOME_UUID').connections (reply) -> log reply #rs.cloudMonitoring.entities.assume('SOME_UUID').details (reply) -> log reply #rs.cloudMonitoring.entities.assume('SOME_UUID').update { # ip_addresses: {}, # label: 'Some Server Name', # agent_id: 'SOME_UUID' #}, (reply) -> # log reply #rs.nextgen.servers.assume('SOME_UUID').serverActions (list) -> # console.log list #rs.cloudDatabases.instances.assume('SOME_UUID').enableRoot (reply) -> # console.log reply #rs.cloudMonitoring.entities.assume('SOME_ID').details (reply) -> # console.log reply #rs.cloudMonitoring.entities.assume('SOME_ID').update { # agent_id: 'SOME_ID' #}, (reply) -> # console.log 'update reply:', reply #rs.cloudMonitoring.entities.all (entities) -> ... #rs.cloudMonitoring.entities.new { # label: '' # ip_addresses: { # public0_v4: '' # private0_v4: '' # access_ip1_v4: '' # } # managed: true # uri: '' # metadata: {} #}, (reply) -> # console.log arguments # Get Check #rs.cloudMonitoring.entities.assume(process.argv[5]).getCheck checkID, (reply) -> log reply # Create Check #rs.cloudMonitoring.entities.assume(process.argv[5]).createCheck { ..jsonstuff.. }, (reply) -> log reply # Get Alarm #rs.cloudMonitoring.entities.assume(process.argv[5]).getAlarm alarmID, (reply) -> log reply # Create Alarm #rs.cloudMonitoring.entities.assume(process.argv[5]).createAlarm { ..jsonstuff.. }, (reply) -> log reply #rs.cloudFiles.containers.all (containers) -> # console.log containers # A lot of products have aliases, for instance, instead of # rs.cloudServersOpenStack, you can just say "nextgen" # #rs.nextgen.servers.all (servers) -> # console.log servers # #rs.nextgen.images.all (images) -> # console.log images # # They're just links - so you get the same endpoint URLs: #console.log rs.nextgen.images._racksmeta.target() # # Cloud load balancers... #rs.clbs.assume('CLB_ID').details (reply) -> # console.log reply # Some services which normally return ugly plaintext (like OpenStack's Swift AKA CloudFiles) # are wrapped nicely into json and still work with things like RacksJS's .target() # see the "plaintext" option within the code # #rs.cloudFiles.containers.all (reply) -> # console.log reply # Override the Rackconnect automation level status for a server #rs.servers.assume('SOME_UUID').setMetadata # 'rax_service_level_automation': 'Complete' #rs.nextgen.servers.all (servers) -> # servers.forEach (server) -> # server.changePassword 'some_new_password', (reply) -> # console.log 'server:', server.name, reply ## NextGen, also known as rs.cloudServersOpenStack, includes all modern Rackspace servers. ## It's by far the most polished of all the Racksjs components ## Images and flavors! #rs.nextgen.images.all (images) -> console.log images #rs.nextgen.flavors.all (flavors) -> console.log flavors ## Create a new server! #rs.nextgen.servers.new { # 'name': 'racksjs_test' # 'flavorRef': 'performance1-1' # 'imageRef': 'SOME_UUID' #}, (server) -> # ## We can poll until the build is complete # server.systemActive (details) -> # console.log 'server object:', server, 'server details', details # server.delete() # You can also create a user with a bootable volume quite easily: # rs.servers.new { # 'name': 'bootable_volume_test' # 'flavorRef': 'compute1-4' # 'imageRef': '' # 'block_device_mapping_v2': [ # { # "boot_index":"0" # "uuid":"753a7703-4960-488b-aab4-a3cdd4b276dc" # "volume_size":"100" # "source_type":"image" # "destination_type":"volume" # "delete_on_termination":false # } # ] # "max_count":1 # "min_count":1 # }, (server) -> # You dont have to make needless HTTP calls if you already know an assets ID. Just "assume" it exists. #rs.servers.assume('SOME_UUID').rebuild { # imageRef: 'SOME_UUID' # flavorRef: '5' #}, (reply) -> # console.log reply #rs.servers.assume('SOME_UUID').setMetadata { 'RackConnectPublicIP': 'IPV4_HERE }