Cosmetics

Customization of step’s static resources (images, stylesheets, …) is made possible in a generic way using the two following properties:

• cosmetics.folder: This property points to a local folder on the controller’s host in which alternate cosmetic resources are stored (conf/cosmetics/ by default).

• cosmetics.interceptedResourceList: A comma-separated list of resources which are supposed to be replaces with custom resources. Defaults to step-variables.css.

Using these two properties, step will intercept frontend requests to the resources in cosmetics.interceptedResourceList and redirect them to the corresponding resources in cosmetics.folder.

Simple example

One of the most classical need for a custom theme is to modify the background in order to easily identify different environments.

For example, if you modify the existing step-variables.css as follow:

:root {
  /* Production */
  --main-header-bg-color: #992600;
}

You will then be able to easily distinguish your productive environment:

Complex example

You may want to customize the theme more in detail. Below is a complete example overriding all the step logos of the default style:

First, we need to custom the css even more:

• We override the image on the right side of the login page via background: url(’login_page.png’). This is needed as the name of this image will depend on the codename of the major release you use.
• We also increase the size of the company logo on the login and home pages.

:root {
  /* Production */
  --main-header-bg-color: #992600;
}
.enterprise-theme .step-login-flex-child.step-mountain-background {
  /* Override the login image */
  background: url('login_page.png') no-repeat no-repeat right;
  background-size: cover;
}
.enterprise-theme .step-login-flex-child div.step-login-content div.step-login-logo img {
  /* Increasing the company logo on the login page */
  max-width: 240px;
}
.enterprise-theme #sidebar-logo-and-burgermenu .sidebar-logo {
  /* Increasing the company logo on the home page */
  height: 60px;
}

Then we intercept all the images using the step branding by changing the cosmetics.interceptedResourceList as follow:

cosmetics.interceptedResourceList=step-variables.css,favicon.png,logologin.png,logotopleft.png

We also change the window title via the ui.title property, so we can easily identify which tab is our productive environment:

ui.title=PROD - Step Automation

Finally, we add the needed png files in the resource folder. These images are:

• favicon.png: This is the website icon
• login_page.png: the main image displayed at the right of the login page
• logologin.png: This is the image displayed at the top of the login form
• logotopleft.png: This is the image displayed to the top left of the main page

Restart the controller and the customization will be in place:

• Login page:

• Home page:

Modifying the Screens Templates

Model

Step entities are displayed based on their model. This model defines which properties (i.e. name, application, type, …) an entity has. To do so go to “Settings -> Screens” and select the entity on the dropdown. Models are set per project, so the intended project needs to be selected before changing the model.

Tables

Each entity has its table in Step (i.e. the plan table shows all plans), while the table is open the user can use drag & drop to sort the displayed properties of this entity and also select / deselect which properties are shown in three vertical dots in the right.

How to modify models

These models can be modified:

ExecutionParameters

Create an execution Parameter screen. (note: unlike with other models the ID directly contains the parameter id without prepending “attributes.”)

This adds a new input field to the plan execution dialog and allow you to use the new parameter when executing your plan:

In the execution you will find the parameter here:

Keyword

Add a new property to your keywords and compare masks. It will be useful for example for identifying multiple version of the same keyword:

Example of how to add a property to the keyword model:

Example how the keyword list looks like with a custom keyword model (See table customization of how to show the column):

Additional models

Other models work the same as the keyword:

• plan: Define extra properties for plans

Value Cell Renderer

Value Cell Renderer is an advanced property of the entity models. It allows to manage how the property is rendered in the table. The following renderers are currently available:

• planLink: for planTable: will display the plans name as a link that opens the plan editor
• functionLink: for functionTable: will display the functions name as a link that opens the functions settings
• functionLinkEditor: for functionTable: will display the functions name as a link that opens the functions editor
• functionPackageLink: for functionTable: will display the functions name as a link that opens the functions editor
• schedulerTaskLink: for schedulerTable: will display the schedules name as a link that opens the schedules settings
• parameterLastModification: for parameterTable: renders the lastModificationDate as a date and shows the last modifications user as a tooltip
• parameterKey: for parameterTable: will display the parameter name as a link that opens the parameter settings
• htmlDescription: will render attributes.description as html in the table

How to modify entity table columns

By clicking the three vertical dots on each table, the shown columns can be selected:

After changing the column order (by drag & drop) or selecting/deselecting shown columns the changes have to be saved. Depending on the users permissions four options are available:

• Saving for yourself in all projects
• Saving for yourself in the current project
• Saving for all users in the current project
• Saving for all users in all projects

These settings will be applied in the order in which they are set. This means if a user first sets the columns for project A and then for all projects, the latter settings will be also applied for project A. The same is true if an admin sets settings for all users and projects, and a user later sets his own settings. To prevent users from doing so, the permissions can be set to prevent users from having individual settings, see: Rights Matrix

Customizing the menu

Starting with Step 21.2 Enterprise Edition, the navigation menu can be customized based on user roles. To enable the customization you’ll need to create a configuration file and reference it in the step.properties.

Menu configuration file

The first section of the file describes the list of menu entries: title, id and icon. The second section describes the menu hierarchy by role. If no hierarchy is defined for a given role, the one defined under ‘default’ is used. Attached file is an example containing the whole menu as default: menu-config.yaml.

Enable the menu customization

In step.properties add following line to reference your configuration file:

ui.menu.config=[path_to_menu_config_yaml_file]

Agent Routing and Affinity

Agent token affinity evaluators are used during Keyword routing to determine the best agent token to execute a Keyword on. During Keyword routing all the available agent tokens are iterated over. The role of the affinity evaluator is to give each agent token a score depending on how good each token matches the token selection criteria. The token with the highest score is selected for Keyword execution. A score of -1 means that the token doesn’t match the selection criteria and will not be selected for Keyword execution.

Configuring a custom evaluator

Since 3.11 Step make it possible to use custom affinity evaluators. The default affinity evaluator (TokenWrapperAffinityEvaluatorImpl) should be enough in the majority of cases and you should only need custom affinity evaluators in some special cases. Before trying to use custom affinity evaluators be sure you have a very good understanding of Keyword routing and of the role of affinity evaluators.

Custom affinity evaluators are defined in step.properties with the following property: grid.tokens.affinityevaluator.classname=full.qualified.name.of.the.affinity.evaluator.Class

Affinity evaluators might require specific properties. These custom properties are defined in step.properties with the properties grid.tokens.affinityevaluator.x where x is the name of the custom property.

Default evaluator

The default affinity evaluator works as follows:

• returns the count of the matching token selection criteria (mandatory and optional criteria)
• returns -1 if not all the mandatory token selection criteria are matching

This means that the token matching the highest number of criteria is selected. See class TokenWrapperAffinityEvaluatorImpl for more information.

CapacityAware example

The default affinity evaluator selects tokens independently of other parameters like the current load (token usage) of the agent and thus doesn’t guaranty an even distribution of the load on the different tokens. In some situation we might want to be sure that the load is distributed evenly over the agents. If you have this requirement the CapacityAwareTokenWrapperAffinityEvaluatorImpl would be better suited than the default affinity evaluator.

There are also situations where you need to run more than one agent instance per agent host (for instance 1 java agent and 1 .NET agent). In this case you might want to define the token capacity accross all the agents of a same host. You might for instance want to ensure that not more than one Keyword is executed on the same agent host at the same time. Doing this with the default affinity evaluator isn’t possible. The CapacityAwareTokenWrapperAffinityEvaluatorImpl makes it possible to define a token capacity per agent host (and not per agent instance).

To enable the CapacityAwareTokenWrapperAffinityEvaluatorImpl:

• add the following line to step.properties:

  grid.tokens.affinityevaluator.classname=step.grid.tokenpool.affinityevaluator.capacityaware.CapacityAwareTokenWrapperAffinityEvaluatorImpl

• create a file with following sample content. this file configures the desired token a capacity per agent host:

  {"agents":[
  {"hostname":"my-multi-agent-host-1","capacity":1},
  {"hostname":"my-multi-agent-host-2","capacity":10}
  ]}

• set the path to the above mentioned file in step.properties:

  grid.tokens.affinityevaluator.configuration=/path/to/the/configuration.json

Configuring the CapacityAware evaluator

The configuration file defined by grid.tokens.affinityevaluator.configuration configures the CapacityAwareTokenWrapperAffinityEvaluatorImpl. The configuration file is formatted as json. It defines the maximal token capacity per agent host. The maximal token capacity defines the maximum number of Keyword executions for a specific agent host. To define the token capacity of an agent host, add the following entry to the array “agents” of the configuration file: {“hostname”:“my-multi-agent-host-XYZ”,“capacity”:1},

The hostname should be equals to the hostname of the agent displayed in the grid view in the URL column. If for instance the URL of one of the agents of the agent host displays:

http://my-agent-01.my.net:50056

my-agent-01.my.net

A capacity of -1 means unlimited.

If a host isn’t specified in this configuration file the capacity is unlimited per default.

The configuration file reloaded automatically on every change. A restart of the controller is therefore not required to take the changes into account.

Custom groovy macros

Installation

In order to use Groovy macros, three steps have to be done:

• First, you have to create a folder on the controller. This folder will contain the Groovy file defining your macros.
• The startup script should be modified by adding the new folder in the class path:
  “%JAVA_PATH%java.exe” %JAVA_OPTS% -cp “..\lib*;..\templates” step.controller.ControllerServer -config=../conf/step.properties
• A new line has to be added in the step.properties file. This option will allow you to specify the name of the Groovy class containing your marcos:
  tec.expressions.scriptbaseclass=Templates

When these changes are done, you can then create your macros file in the new folder.

Note that:

• The controller has to be restarted each time you modify these options or the macro file
• The base name of macro file should have the same name used in tec.expressions.scriptbaseclass and have the .groovy extension (…\step-controller-X.Y.Z\templates\Templates.groovy in our example)

Validation

• In order to validate our groovy macro, we will download the following file and save it under …\step-controller-X.Y.Z\templates\Templates.groovy: Template.groovy
• You should now restart your controller for the new macro to be taken into account
• You can now re-run the example test case. The function now() should be known and return the current date:

• As a final check, you can also ensure that the new test status is properly reflected in ALM:

Troubleshooting

The following exceptions can be faced when using Groovy Macros:

• Script1.groovy: 1: unable to resolve class WrongTemplate:
  • The name in tec.expressions.scriptbaseclass didn’t match the name of your Groovy class
  • The name of the folder didn’t match the name in the Java path
• groovy.lang.MissingMethodException: No signature of method: Script1.unknownMethod():
  • The method name is misspelled
  • Or our didn’t restart your controller after defining the new macros

Password strength policies

Starting with version 3.18, it is possible to define a number of policies that user passwords must adhere to.

Note that by default, no policies are active. The following policies can be defined:

Minimum password length

The minimum required password length can be specified as shown below.

security.password.minimumLength=4

Require character classes

To require that (at least one) upper case (A-Z), lower case (a-z), or numeric (0-9) character must be present, use these configuration options:

security.password.requireUpperCase=true
security.password.requireLowerCase=true
security.password.requireNumeric=true

Require special characters

If you want to enforce that (at least one) special character is present in the password, provide a list of the special characters to be checked, for example:

security.password.requireSpecialCharacter=!@#$%&*

Controlling Messaging Recipients

Version 22 of Step introduced a new notification mechanism which allows the system to send messages to users. Currently, this is used for license-related events.

Starting with Step 22.1, it is possible to specify the audience for messages, i.e., which messages should be delivered to which users. Each message has an associated category and optional subcategory, while audiences consist of a (comma-separated) list of roles.

Here is an example configuration excerpt:

messaging.audience.licensing.licenseUsage = admin, manager

This defines that any message in the licensing category and licenseUsage subcategory will be dispatched to all users that have a primary role of admin or manager.

Configuration properties

Here is the list of supported configuration properties:

• messaging.audience : default audience. If not specified, implicitly uses the default admin.
• messaging.audience.licensing : default audience for all licensing events
• messaging.audience.licensing.licenses : messages related to licenses themselves, e.g. new licenses available, license expiry events etc.
• messaging.audience.licensing.licenseUsage : messages related to license usage, in particular license violations.
• messaging.audience.licensing.licensingError : messages related to unexpected licensing exceptions.

Behavior

Conceptually, the configuration keys are evaluated in a hierarchical way, bottom-up. That is, to determine the audience for a particular message with category x and subcategory y:

1. If messaging.audience.x.y is defined, its content will be used.
2. Otherwise, if messaging.audience.x is defined, its content will be used.
3. Otherwise, if messaging.audience is defined, its content will be used.
4. Otherwise, the built-in fallback admin is used as the role to receive messages.

Notes:

• The system does not perform validation of the role names specified. It is possible to provide invalid role names – they will just not match any user.
• If a specific configuration key is present, but contains no value at all, or only invalid role names, messages which are affected by this configuration key will simply not be delivered to anybody.