Release Notes: IBM Aspera Orchestrator 4.1.0

4.1.3

4.1.2

Fixed issues

Orchestrator#4332 - When importing an exported workflow with remote node dependencies a 500 server error is shown.

Orchestrator#4273 - When deleting a step from the Workflow designer view, the deleted step template remains visible in the Change action template page.

Orchestrator#4241 - When configuring a workflow with an input parameter of type password an error appears when selecting the Launch icon in the workflows page.

Orchestrator#4209 - Users are not able to map inputs for duplicated action templates.

Orchestrator#4203 - When using a runtime input parameter the API Initiate a workorder endpoint fails and shows an error.

Orchestrator#4194 - Upgrading from Orchestrator version 4.1.0 to 4.1.1 shows an error when using an external database since the database.yml file is being overwritten while running the rpm command.

Orchestrator#4159 - Sessions are not automatically cleared even after reaching the designated expiration date and time.

Orchestrator#4140 - When using BitBucket to push or pull workflows, selecting the Update option triggers the Configuration updated, but unable to reach hostname: bitbucket.org error.

Other changes

Orchestrator#4292 - A new set of plugins has been included as part of the release.

Orchestrator#4272 - Re-added the button to edit outputs in a workstep in the work order details pane.

Orchestrator#1278 - Added a new UI field within the SAML configuration screen to specify the return host directly in the UI. You will no longer need to manually edit configuration files or restart the server.

Orchestrator#4344 - The Git integration documentation section has been updated for this release.

4.1.1

4.1.0

New features

Orchestrator#2141 - Developed a new script to upgrade workflow YAML files from version 4.0.1 to be compatible with Orchestrator 4.1.0. This resolves issues caused by changes in Ruby and Rails introduced in Orchestrator 4.1.0, which made the previous YAML formats incompatible.

Orchestrator#1178 - Added the Look-up Editor portlet to view and edit tables created by the SimpleParameterLookup plugin.

Orchestrator#327 - Administrators with access to terminal can now remove users from the administrator group.

Orchestrator#240 - Git for Workflow synchronization is now supported.

Breaking changes

Orchestrator#3952 - Orchestrator now enforces the use of unique email addresses for all new users. That means, if you create a new user with an already used email address the process will fail. Make sure to use a unique email address before performing an upgrade.

Orchestrator#3298 - Starting with Orchestrator 4.1.0, you have to store all SSH keys that Orchestrator uses under /home/aspweb (you can place them directly in /home/aspweb or inside any other directory in that same location). This will ensure that the aspweb user has permissions to perform file operations as needed.

after_initialize :set_portlet_name_and_desc
  before_save      :set_portlet_name_and_desc
  def set_portlet_name_and_desc
    self.name = "PORTLET_NAME"
    self.description = "PORTLET_DESCRIPTION"
  end

Orchestrator#2169 - Removed the option to define a new principal.

Orchestrator#1853 - The help page and the API routes have received several changes, including the authentication method (now using bearer token) to improve security with this release.

Orchestrator#1847 - When upgrading to IBM Aspera Orchestrator 4.1.0, all your existing plugins will be updated to their latest versions. If there are no newer supported versions, contact support for assistance.

Orchestrator#1834 - The orchestrator.yml file is no longer part of the application in 4.1.0.

Orchestrator#1509 - All routes have been updated to use REST-style parameter methods for identifying resources and parameters, instead of embedding parameters directly in the URL. For example, work_steps/:id/cancel is now work_steps/cancel?id=WORKSTEP_ID. The only exception is the standard Rails CRUD routes, which continue to embed the resource ID in the URL by default (for example, work_steps/:id/edit).

Orchestrator#1238 - The Capsule manager menu has been removed from the Engines view sidebar.

Other changes

Orchestrator#3932 - The queue insertion logic prioritizes items based on their priority values (in ascending order). An item is inserted into the queue in a way that no higher-priority item (lower numerical value) appears below it (for example, a priority value of 1 has a higher priority than 100). After insertion, items can still be manually rearranged within the queue; however, priority values are no longer considered when reordering items after insertion.

Orchestrator#3864 - If you're upgrading from 4.0.1 to 4.1.0 you cannot reset work orders that were created before the upgrade. To launch a new working work order, relaunch these work orders instead.

Orchestrator#3654 - Plugins that are currently in beta will not be enabled on the plugins page.

Orchestrator#3077 - Clicking on a SubWorkflow work step shows the plugin's configurations, aligning behavior to match other plugins. To see the plugins' sub-workflow, go to the Subworkflows table in the Configurations page.

Orchestrator#2898 - The mount manager feature has been removed.

Orchestrator#2430 - The PluginInstall class has been renamed to Plugin.

Orchestrator#2419 - Removed the Monitors > All Monitors page from the dashboard route.

Orchestrator#2402 - The health_check/dashboard and health_check/refresh routes have been removed.

Orchestrator#2255 - The orch_web_root has been removed from the remote_nodes table.

Orchestrator#2246 - Virtual Catcher has been removed from the application.

Orchestrator#2099 - You can install external JavaScript files in the /opt/aspera/var/config/orchestrator/custom_js directory. This will preserve these custom files even if you uninstall Orchestrator.

Orchestrator#2095 - Replaced orchestrator.yml's web_root variable with Rails.configuration.web_root.

Orchestrator#2286 - Monitors now point to the portable_id column instead of the name to associate with workflows.

Orchestrator#1973 - All credentials are now encoded with a unique salt, instead of a single global secret.

Orchestrator#1834 - Orchestrator settings are now global to all orchestrator nodes that share the same database, as a user this guarantees the same behavioral experience across all orchestrator nodes in HA.

Orchestrator#1444 - Removed access to the net use managers feature.

Orchestrator#1211 - Removed BasicLogger to preserve application stability.

Orchestrator#858 - All startup and stopping scripts in the scripts directory have been removed.

Orchestrator#815 - Added max_supported_server_version, min_supported_server_version, and is_beta to all plugins' metadata.yaml files.

Orchestrator#759 - Replaced the name "slaves" with "workers".

Orchestrator#485 - Improvements to the way the Work Order page is displayed in the Dashboard.

Orchestrator#210 - Git integration now supports repositories hosted on any platform, not just github.com.

Orchestrator#202 - An error message is now shown when attempting to update the value of a configuration parameter with a string that contains too many characters.

When stopping workers from the UI, the workers are no longer restarted. However, when killing workers from the UI, they are restarted regardless of whether the process manager is enabled or disabled.

If you have workflows using the AsperaSharesManagement plugin, you need to save those workflows and launch them again in 4.1.0.

If Orchestrator is running behind a firewall, you must allowlist the rubygems.org domain to enable access to required Ruby gems.

You may need to save certain workflows again after upgrading if version 4.1.0 includes updated plugins that introduce new fields not present before the upgrade.

Fixed issues

Connection to Active Directory failed:
Active Directory has been disabled. Adjust your settings, click on Enable Directory Service, and save again to retest.

Orchestrator#1965 - Administrators are able to deactivate other administrator accounts.

Orchestrator#1849 - Automatic-refreshing of data on orchestrator pages is paused when the window is minimized.

Orchestrator#1167 - After applying patch level 3, resources disappear and it is not possible to create or edit them.

Orchestrator#1131 - Security enhancement related to user authentication.

Orchestrator#1112 - If you are declaring custom attributes for a node and change from a password connection to a key connection, after saving these changes all custom attributes are lost.

Orchestrator#663 - Autoscale workers can help take on work orders when all other work orders are busy.

Orchestrator#954 - Security improvement to address certain vulnerable libraries.

Orchestrator#951 - Issues with importing a workflow to Orchestrator using the API.

Orchestrator#282 - Fixed the import_plugin endpoint.

Orchestrator#257 - Orchestrator allows to start a Work order even if the workflow is not published.

Orchestrator#203 - The default logon_timeout value has been updated to 3600 seconds.

Orchestrator#166 - Issues with Orchestrator and Git integration.

Documentation updates

Included a new section for Zero-downtime patching in a High-Availability cluster.

Added a new section for the Git integration.

Included a new section for Migrating to a supported OS.

Known issues and limitations

Orchestrator#4568 - When reopening the filter options in the Inline Journals page. The From and To fields show the default dd/mm/yyyy values instead.

If you import a node monitor from one instance (for example, instance 1) that monitors a specific node into another instance (for example, instance 2), and the corresponding node does not exist in the target instance, polling the node monitor results in an error.

Exporting a monitor from previous Orchestrator versions into 4.1.0 is not currently supported.

Orchestrator#3939 - Issues may occur when storing input string variables that contain a colon (:). For example, if a user enters a string such as IP:Port in the Map Inputs tab, after saving and reopening the step, the string may appear as IP"Port". However, the input variable will still display the correct value in the launched work order details. Workaround: Pass these input strings as parameters and then map the input variable to those parameters.

Orchestrator#3905 - In an high availability (HA) setup configured with an external load balancer, when workers across multiple nodes are running work orders, the load balancer can only detect workers from a single node instead of recognizing all active node workers.

Orchestrator#3880 - If you encounter the "Execution Error in workstep 6 preparation for activation: undefined method 'timeout' for nil" error while processing a work order, go to Engine > Processes and "kill" all workers (they will restart automatically). After doing this, work orders will work as expected.

rpm -Uvh ibm-aspera-orchestrator-*.rpm --replacefiles 

Orchestrator#157 - Capsules are loaded during the Orchestrator server startup, but updates made after startup are not reloaded automatically into Puma. In contrast, engine and worker processes continuously reload updated capsule files as part of their normal operation.

Supported plugins

The following table contains all supported plugins for the current version release of Orchestrator.

System requirements

Operating systems

Linux 64-bit

Database

Web server

Runtime environments

Browsers

Product support

For online support, go to the IBM Aspera Support site at https://www.ibm.com/mysupport/. To open a support case, log in with your IBMid or set up a new IBMid account.