IM - Infrastructure Manager client

IM is a tool that ease the access and the usability of IaaS clouds by automating the VMI selection, deployment, configuration, software installation, monitoring and update of Virtual Appliances. It supports APIs from a large number of virtual platforms, making user applications cloud-agnostic. In addition it integrates a contextualization system to enable the installation and configuration of all the user required applications providing the user with a fully functional infrastructure.

Usage: im_client.py \
         [-u|--xmlrpc-url <url>] \
         [-r|--restapi-url <url>] \
         [-v|--verify-ssl] \
         [-a|--auth_file <filename>] \
         operation op_parameters

Achievements

This software has received a gold badge according to the Software Quality Baseline criteria defined by the EOSC-Synergy project.

1 INSTALLATION

1.1 REQUISITES

IM is based on python, so Python 3 or higher runtime and standard library must be installed in the system.

It is also required the RADL parser, available in pip as the 'RADL' package. It is also required the Python Requests library available as 'python-requests' in O.S. packages or 'requests' in pip.

1.2 OPTIONAL PACKAGES

In case of using the SSL secured version of the XMLRPC API the SpringPython framework must be installed.

1.3 INSTALLING

1.3.1 FROM PIP

You only have to call the install command of the pip tool with the IM-client package.

pip install IM-client

1.3.2 FROM SOURCE

You only need to install the tar-gziped file to any directoy:

tar xvzf IM-client-X.XX.tar.gz

1.3.3 IM-Client Docker image

The IM Client has an official Docker container image available in Github Container Registry that can be used instead of installing the CLI. You can download it by typing:

sudo docker pull ghcr.io/grycap/im-client

You can exploit all the potential of the IM Client as if you download the CLI and run it on your computer:

docker run --rm -ti -v "$PWD:/tmp/im" ghcr.io/grycap/im-client \
      -r https://server.com:8800 -a /tmp/im/auth.dat list
docker run --rm -ti -v "$PWD:/tmp/im" ghcr.io/grycap/im-client \
      -r https://server.com:8800 -a /tmp/im/auth.dat create /tmp/im/somefile.radl

1.4 CONFIGURATION

To avoid typing the parameters in all the client calls. The user can define a config file "im_client.cfg" in the current directory or a file ".im_client.cfg" in their home directory. In the config file the user can specify the following parameters:

[im_client]
# only set one of the urls
#xmlrpc_url=http://localhost:8899
restapi_url=http://localhost:8800
auth_file=auth.dat
xmlrpc_ssl_ca_certs=/tmp/pki/ca-chain.pem

1.4.1 AUTH FILE

The authorization file stores in plain text the credentials to access the cloud providers, the IM service and the VMRC service. Each line of the file is composed by pairs of key and value separated by semicolon, and refers to a single credential. The key and value should be separated by " = ", that is an equals sign preceded and followed by one white space at least, like this:

id = id_value ; type = zzzz ; username = xxxx ; password = yyyy

Remember that the InfrastructureManager auth line is mandatory, like this:

id = im ; type = InfrastructureManager ; username = xxxx ; password = yyyy

or using an OIDC token:

id = im ; type = InfrastructureManager ; token = xxxxxx

Values can contain "=", and "\n" is replaced by carriage return. The available keys are:

• type indicates the service that refers the credential. The services supported are InfrastructureManager, VMRC, OpenNebula, EC2 , FogBow, OpenStack, OCCI, LibCloud, Docker, GCE, Azure, AzureClassic and Kubernetes.

• username indicates the user name associated to the credential. In EC2 it refers to the Access Key ID. In GCE it refers to Service Account's Email Address.

• password indicates the password associated to the credential. In EC2 it refers to the Secret Access Key. In GCE it refers to Service Private Key (either in JSON or PKCS12 formats). See how to get it and how to extract the private key file from here. In OpenStack sites using 3.x_oidc_access_token authentication it indicates the OIDC access token.

• tenant indicates the tenant associated to the credential. This field is only used in the OpenStack plugin.

• host indicates the address of the access point to the cloud provider. This field is not used in IM, GCE, Azure, and EC2 credentials.

• proxy indicates the content of the proxy file associated to the credential. To refer to a file you must use the function "file(/tmp/proxyfile.pem)" as shown in the example. This field is used in the OCCI and OpenStack plugins.

• project indicates the project name associated to the credential. This field is only used in the GCE plugin.

• public_key indicates the content of the public key file associated to the credential. To refer to a file you must use the function "file(cert.pem)" as shown in the example. This field is used in the Azure Classic and Docker plugins. For Azure Classic see how to get it here.

• private_key indicates the content of the private key file associated to the credential. To refer to a file you must use the function "file(key.pem)" as shown in the example. This field is used in the Azure Classic and Docker plugins. For Azure Classic see how to get it here.

• id associates an identifier to the credential. The identifier should be used as the label in the deploy section in the RADL.

• subscription_id indicates the subscription_id name associated to the credential. This field is only used in the Azure and Azure Classic plugins. To create a user to use the Azure (ARM) plugin check the documentation of the Azure python SDK: here

• token indicates the OpenID token associated to the credential. This field is used in the OCCI and also to authenticate with the InfrastructureManager. To refer to the output of a command you must use the function "command(command)" as shown in the examples.

OpenStack addicional fields

OpenStack has a set of addicional fields to access a cloud site:

• auth_version the auth version used to connect with the Keystone server. The possible values are: 2.0_password or 3.X_password. The default value is 2.0_password.

• base_url base URL to the OpenStack API endpoint. By default, the connector obtains API endpoint URL from the server catalog, but if this argument is provided, this step is skipped and the provided value is used directly. The value is: http://cloud_server.com:8774/v2/<tenant_id>.

• service_region the region of the cloud site (case sensitive). It is used to obtain the API endpoint URL. The default value is: RegionOne.

• service_name the service name used to obtain the API endpoint URL. The default value is: Compute.

• auth_token token which is used for authentication. If this argument is provided, normal authentication flow is skipped and the OpenStack API endpoint is directly hit with the provided token. Normal authentication flow involves hitting the auth service (Keystone) with the provided username and password and requesting an authentication token.

An example of the auth file:

# OpenNebula site
id = one; type = OpenNebula; host = osenserver:2633; username = user; password = pass
# OpenStack site using standard user, password, tenant format
id = ost; type = OpenStack; host = https://ostserver:5000; username = user; password = pass; tenant = tenant
# OpenStack site using VOMS proxy authentication
id = ostvoms; type = OpenStack; proxy = file(/tmp/proxy.pem); host = https://keystone:5000; tenant = tname
# OpenStack site using OIDC authentication for EGI Sites
id = ost; type = OpenStack; host = https://ostserver:5000; username = egi.eu; tenant = openid; password = command(oidc-token OIDC_ACCOUNT); auth_version = 3.x_oidc_access_token; domain = project_name_or_id
# IM auth data
id = im; type = InfrastructureManager; username = user; password = pass
# VMRC auth data
id = vmrc; type = VMRC; host = http://server:8080/vmrc; username = user; password = pass
# EC2 auth data
id = ec2; type = EC2; username = ACCESS_KEY; password = SECRET_KEY
# Google compute auth data
id = gce; type = GCE; username = username.apps.googleusercontent.com; password = pass; project = projectname
# Docker site with certificates
id = docker; type = Docker; host = http://host:2375; public_key = file(/tmp/cert.pem); private_key = file(/tmp/key.pem)
# Docker site without SSL security
id = docker; type = Docker; host = http://host:2375
# OCCI VOMS site auth data
id = occi; type = OCCI; proxy = file(/tmp/proxy.pem); host = https://server.com:11443
# OCCI OIDC site auth data
id = occi; type = OCCI; token = token; host = https://server.com:11443
# Azure (RM) site auth data
id = azure; type = Azure; subscription_id = subscription-id; username = [email protected]; password = pass
# Kubernetes site auth data
id = kub; type = Kubernetes; host = http://server:8080; username = user; password = pass
# FogBow auth data
id = fog; type = FogBow; host = http://server:8182; proxy = file(/tmp/proxy.pem)
# Azure Classic auth data
id = azurecla; type = AzureClassic; subscription_id = subscription_id; public_key = file(/tmp/cert.pem); private_key = file(/tmp/key.pem)

1.4 INVOCATION

The programim_client is called like this:

Usage: im_client.py \
         [-u|--xmlrpc-url <url>] \
         [-r|--restapi-url <url>] \
         [-v|--verify-ssl] \
         [-a|--auth_file <filename>] \
         operation op_parameters

• option: -u|--xmlrpc-url url

  URL to the XML-RPC service. This option or the -r one must be specified.

• option:: -r|--rest-url url

  URL to the REST API on the IM service. This option or the -u one must be specified.

• option:: -v|--verify-ssl

  Verify the certificates of the SSL connection. The default value is False,

• option: -a|--auth_file filename

  Path to the authorization file, see here. This option is compulsory.

• option: -f|--force

  Force the deletion of the infrastructure. Only for destroy operation. The default value is False.

• option: -q|--quiet

  Work in quiet mode. Avoid all unnecessary prints. The default value is False.

• option: -n|--name

  Show/use Infrastructure name in the selected operation. In case of list operation it will show the name of each infrastructure (if available). In other operations if this flag is set the user should specify the name of the infrastructure instead of the ID. The default value is False.

• operation:

  list [filter]: List the infrastructure IDs created by the user. The filter parameter is optional and is a regex that will be used to filter the list of infrastructures.

  create <radlfile> [async_flag] Create an infrastructure using RADL specified in the file with path radlfile. The async_flag parameter is optional and is a flag to specify if the creation call will wait the resources to be created or return immediately the id of the infrastructure.

  destroy <infId> Destroy the infrastructure with ID infId.

  getinfo <infId> Show the information about all the virtual machines associated to the infrastructure with ID infId.

  getcontmsg <infId> Show the contextualization message of the infrastructure with ID infId.

  getstate <infId> Show the state of the infrastructure with ID infId.

  getoutputs <infId> Show the outputs of infrastructure with ID infId (Only in case of TOSCA docs with REST API).

  getvminfo <infId> <vmId> Show the information associated to the virtual machine with ID vmId associated to the infrastructure with ID infId.

  getvmcontmsg <infId> <vmId> Show the contextualization message of the virtual machine with ID vmId associated to the infrastructure with ID infId.

  addresource <infId> <radlfile> [ctxt_flag] Add to infrastructure with ID infId the resources specifies in the RADL file with path radlfile. The ctxt_flag parameter is optional and is a flag to specify if the contextualization step will be launched just after the VM addition. If not specified the contextualization step will be launched.

  removeresource <infId> <vmId> [ctxt_flag] Destroy the virtual machine with ID vmId in the infrastructure with ID infId. The ctxt_flag parameter is optional and is a flag to specify if the contextualization step will be launched just after the VM addition. If not specified the contextualization step will be launched.

  start <infId> Resume all the virtual machines associated to the infrastructure with ID infId, stopped previously by the operation stop.

  stop <infId> Stop (but not remove) the virtual machines associated to the infrastructure with ID infId.

  alter <infId> <vmId> <radlfile> Modify the specification of the virtual machine with ID vmId associated to the infrastructure with ID vmId, using the RADL specification in file with path radlfile.

  reconfigure <infId> [radl_file] [vm_list] Reconfigure the infrastructure with ID infId and also update the configuration data. The last vm_list parameter is optional and is a list integers specifying the IDs of the VMs to reconfigure. If not specified all the VMs will be reconfigured.

  startvm <infId> <vmId> Resume the specified virtual machine vmId associated to the infrastructure with ID infId, stopped previously by the operation stop.

  stopvm <infId> <vmId> Stop (but not remove) the specified virtual machine vmId associated to the infrastructure with ID infrastructure with ID infId.

  rebootvm <infId> <vmId> Reboot the specified virtual machine vmId associated to the infrastructure with ID infId.

  sshvm <infId> <vmId> [show_only] Connect with SSH with the specified virtual machine vmId associated to infrastructure with ID infId. In case that the specified VM does not have public IP the client will try to connect using the virtual machine with ID 0 as SSH proxy. The show_only parameter is optional and is a flag to specify if ssh command will only be shown in stdout instead of executed.

  ssh <infId> [show_only] Connect with SSH with the virtual machine with ID 0 associated to infrastructure with ID infId. The show_only parameter is optional and is a flag to specify if ssh command will only be shown in stdout instead of executed.

  export <infId> [delete] Export the data of the infrastructure with ID infId. The delete parameter is optional and is a flag to specify if the infrastructure will be deleted from the IM service (the VMs are not deleted).

  import <json_file> Import the data of an infrastructure previously exported with the previous function. The json_file is a file with the data generated with the export function.

  wait <infId> <maxTime> Wait the infrastructure with ID infId to be be in a final state ("configured", "unconfigured" or "failed"). It has an optional parameter maxTime with the max time to wait. It returns 0 if the infrastructure ends with a "configured" state or 1 otherwise.

  create_wait_outputs <radlfile> This operation is a combination of the create, wait and getoutputs functions. First it creates the infrastructure using the specified inputfile, then waits for it to be configured, and finally gets the TOSCA outputs. In case of failure in then infrastructure creation step only the error message will be returned. The results will be returned to stdout in json format::

         {"infid": "ID", "error": "Error message"}

change_auth <infId> <newAuthFile> [overwrite] This operation enables to change the owner of infrastructure with ID infId using the authentication data from file newAuthFile. The overwrite parameter is optional and is a flag to specify if the authentication data will be overwrited or will be appended. The default value is 0.

1.5 PYTHON CLIENT

IMClient can also be used as a Python library to access IM (since version 1.7.0):

   from imclient import IMClient

   auth = IMClient.read_auth_data("/path/auth.dat")
   client = IMClient.init_client("https://im.egi.eu/im", auth)
   inf_desc = """
      network public (outbound = 'yes')

      system node (
      cpu.count>=2 and
      memory.size>=4g and
      net_interface.0.connection = 'public' and
      disk.0.os.name='linux' and
      disk.0.image.url = 'appdb://SCAI/egi.ubuntu.20.04?vo.access.egi.eu'
      )

      configure wn (
      @begin
      ---
      - tasks:
         - debug: msg="Configured!"
      @end
      )

      deploy node 1
   """
   success, inf_id = client.create(inf_desc)
   ...
   success, err = client.destroy(inf_id)