Digital.ai Deploy 10.2.x Release Notes
Upgrade instructions
The Digital.ai Deploy upgrade process you use depends on the version from which you are upgrading, and the version to which you want to go.
For detailed instructions based on your upgrade scenario, refer to Upgrade Deploy.
Support Policy
See Digital.ai Support Policy.
Digital.ai Deploy 10.2.1
New features
In addition to bug fixes, the Digital.ai Deploy 10.2.1 includes the following new features:
- A well-rounded Central Configuration Management feature
- An option to Force Redeploy applications without having to do delta analysis or cleanup
- An enhanced Plugin Manager and Plugin Manager CLI to install, upgrade, and delete plugins with ease
- An enhanced user experience via Unified Navigation menus and widgets
- A new
globalContext
object to set attributes with global scope so that you can access them from any of the Deploy's planning contexts such aspre-plan
,deployed
,plan
andpost-plan
- Enhanced plugins:
- Kubernetes
- Apache Tomcat
- New plugins for:
- Azure Cloud Service
- Cloud Foundry CLI
- New APIs
Central Configuration Management
The Central Configuration Management feature—part delivered in Deploy 10.1—has been enhanced in Deploy 10.2.1.
The following noteworthy enhancements and changes are part of the Central Configuration Management feature in Deploy 10.2.1:
- Added a new
centralConfiguration/deploy-artifact-resolver.yaml
file - Moved the Akka DNS override configuration properties from
conf/xl-deploy.conf
tocentralConfiguration/deploy-server.yaml
- Moved Deploy database configuration properties (both repository and reporting database configuration properties) from
conf/xl-deploy.conf
tocentralConfiguration/deploy-repository.yaml
andcentralConfiguration/deploy-reporting.yaml
files respectively. - Moved Deploy server configuration properties from
conf/xl-deploy.conf
tocentralConfiguration/deploy-server.yaml
- Moved the task configuration properties from
conf/xl-deploy.conf
tocentralConfiguration/deploy-task.yaml
- Moved the satellite configuration properties from
conf/xl-deploy.conf
tocentralConfiguration/deploy-satellite.yaml
- Renamed the
deployit-defaults.properties
file astype-defaults.properties
- Moved some of the configuration properties from the
deployit.conf
file todeploy-client.yaml
anddeploy-server.yaml
files—one for configuring the client and the other for the server. - The
deploy-client-gui.yaml
file is no longer available and all its configuration properties have been subsumed by thedeploy-client.yaml
file - AES algorithm for encryption/decryption of sensitive configuration data such as passwords
- Multiple keys and key rotation for added security
The above list of changes is just for your understanding. The configuration files and properties are auto-migrated when you upgrade to or install Deploy 10.2.1 (or later). No user intervention whatsoever is needed when you upgrade to or install Deploy 10.2.1.
For more information, see Central Configuration Management
Force Redeploy applications
At times you may want to just redeploy an already deployed application by merging and overriding the content without doing the delta analysis or cleanup. Such situations arise when you want to simply destroy/uninstall the existing deployed (application) and install the application again.
Select the Force Redeploy property (check box) of the deployed application from the Deployment Properties dialog box and do the deployment in such situations.
Though this feature works with all the plugins, here's a list of plugins that were tested and qualified in Deploy 10.2.1. Other plugins would be included in the test coverage in due course.
-
File plugin
-
Command Plugin
-
Database plugin
-
Kubernetes plugin
-
AWS plugin
-
Terraform plugin
-
Webserver plugin
-
The Force Redeploy feature is not supported for plugins that are used to deploy WAR type deployables—Tomcat and JEE plugins, for example.
-
The deployment rollback feature is broken after you Force Redeploy an application. This would be fixed in one of the upcoming releases.
Plugin Manager and Plugin Manager CLI
Digital.ai Deploy 10.2.1 is equipped with an enhanced Plugin Manager and a new Plugin Manager CLI.
The enhanced Plugin Manager:
-
Lets you search and install plugins (Browse tab)
-
Shows an Upgrade button for plugins that are installed already and lets you upgrade the plugins whenever a new version is available (Browse tab)
-
Lets you upload a new plugin or a new version of an already installed plugin (Installed tab's Upload new plugin file)
For more information, see Plugin Manager
Deploy 10.2.1 is equipped with a new Plugin Manager CLI utility to delete plugins from both the filesystem and database. For more information, see Plugin Manager CLI
Term-based package retention policy
In addition to the count-based package retention policy, you can now define and rollout package retention policies based on a fixed term (in days). In other words, you can enforce a package retention policy that lets you purge packages that are older than a certain number of days. For more information, see Package retention policy documentation.
Unified Navigation—UX enhancements
One of the objectives of Deploy 10.2.1 is to provide better user experience. Digital.ai Deploy's navigation menus, icons, and widgets have been totally revamped to enhance the usability. For more information, see Get started with the Deploy user interface
New globalContext object for use in all planning contexts
You can now use the globalContext
object to set attributes globally and get those attributes in different planning contexts (such as pre-plan
, deployed
, plan
, and post-plan
) while executing Jython/Python scripts. For more information, see The DeploymentPlanningContext
Digital.ai Deploy Azure cloud service plugin
- Deployment of cloud service spec to the Azure cloud through the Digital.ai Deploy
- Digital.ai Deploy Azure cloud service plugin support is now extended to:
- Create a virtual network
- Create a storage account
- Create a public IP address
Digital.ai Deploy Cloud Foundry CLI plugin
Implemented multiple application deployment using single manifest file in the Cloud Foundry CLI integration plugin. For more information, see Cloud Foundry CLI plugin
Digital.ai Deploy Kubernetes plugin
The Kubernetes plugin is enhanced as follows:
- The
value
property provides details about the deployment endpoints - The secrets are deployed on Kubernetes cluster with the CI name instead of the file name
Digital.ai Deploy Apache Tomcat plugin
The default connection pooling library of Apache Tomcat plugin is updated to Apache Commons DBCP 2.x from DBCP 1.x.
API changes
- Added a new new API to get the overall number of deployments per deployment state—
GET /report/status-overview
. For more information, see GET /report/status-overview API documentation. - Added a new new API to get the number of deployments per application and environment—
GET /report/app-env-status
. For more information, see GET /report/app-env-status API documentation. - Added a new new API to get a summary of the tasks currently running and waiting in the queue—
GET /queue/summary
. For more information, see GET /queue/summary API documentation. - Added a new new API to list all the QueuedTasks waiting in the task queue—
GET /queue/tasks
. For more information, see GET /queue/tasks API documentation.
Bug fixes—10.2.1
- [ENG-2488] - The incorrect server-side error message that was used in place of the form-level validation error (if you leave the mandatory
Name
field empty when you create a configuration item) has been fixed. - [ENG-2618] - Fixed an issue that prevented users from creating deployment packages that consist of a single file with no content (0 bytes) on Deploy sites that use the Oracle database.
- [ENG-4509] - Fixed a security vulnerability to prevent directory traversal attacks.
- [ENG-5316] - aws.rds.DBInstanceSpec subnet group marked as required for AWS Plugin.
- [ENG-5351] - Fixed the issue of was.WmqActivationSpecification fails discovery when "connectionNameList" property has more than one host for WAS Plugin.
- [ENG-5621] - The maximum number of characters allowed for Deploy placeholders is 255 characters.
- [ENG-6609] - Fixed failed rollbacks of J2CResourceAdapterSpec deployments for WAS Plugin.
- [ENG-6748] - Fixed the issue on status check during Start and Stop Server control task runs for WLS Plugin.
- [ENG-6997] - Fixed the issue of was.SibQueueSpec failing during undeploy with cluster mode for WAS Plugin.
Digital.ai Deploy 10.2.2
Kubernetes Operator-based Installation
Note: Helm Charts-based installer is no longer available for Digital.ai Deploy 10.2 and later.
- Digital.ai Deploy 10.2.2 brings you a simple and sophisticated Kubernetes Operator-based installer to install Digital.ai Deploy on Amazon EKS, Azure AKS, and OpenShift on AWS clusters
- You can download the new Kubernetes Operator-based installer from the Deploy/Release Software Distribution site
- Once you download the installer Zip file, Digital.ai Deploy installation is as simple as adding a few configuration parameters and custom resource definitions to the installer files and running the installer
For more information about Kubernetes Operator-based installation, see Kubernetes Operator.
Bug fixes—10.2.2
- [ENG-2022] - Fix made to ensure multiple updates to the same CI will pick the latest key and token from the previous update for WAS Plugin.
- [ENG-6632] - Plugin Manager CLI Improvements.
Known issues
- [ENG-6866]—Users without Admin privileges are not able to view scheduled deployment/control tasks as soon as you turn the Deploy Maintenance Mode on.
Digital.ai Deploy 10.2.3
Kubernetes Operator-based Installation
With Deploy 10.2.3, the Kubernetes Operator offers the following enhancements:
- Kubernetes Operator supports migration of Helm-based installation to Operator-based installation. You can upgrade Deploy 10.0/10.1 versions installed using Helm Charts to Deploy 10.2 on AWS EKS and OpenShift on AWS clusters.
- You can configure OIDC authentication using Keycloak on your Kubernetes Operator-based installation. For more information, see Using Keycloak on Kubernetes Operator-based Deployment.
- You can download the XL-CLI binaries from the Software Distribution site.
Bug fixes
- [ENG-6600] - Fixed the issue where the Terraform plugin does not function when Stitch is enabled.
- [ENG-6934] - Rollback is not functional when "Force redeploy" Deployment property is enabled. Instead undeploy / delete manually based on the requirement.
- [FI-496] - Fixed Deploy 9.7.10 and 10.0.0 performance issues.
- [FI-659] - Fixed the issue with large DARs that caused the Deploy application to go down.
- [FI-753] - Fixed the Digital.ai Deploy Connection Lost Error.
Known issues
- [ENG-6622]—
MissingDirectoryReferenceException
message is displayed when you upgrade from Deploy 10.1 to Deploy 10.2. - [ENG-7005]—Currently there is an issue when you upgrade from Deploy 9.8.1 to Deploy 10.2.1 with Oracle 19c. You can get a hotfix for the issue from the support team. The fix is will be included with the 10.2.3 release.
Digital.ai Deploy 10.2.4
Bug fixes
- [FI-744] - Fixed the issue where XL-deploy failed when configured to start the JMX.
- [FI-746] - Fixed the RabbitMQ max message size exceeded error.
- [FI-798] - Fixed the issue where the upgrade to 10.2.2 Directory reference updates cannot be applied.
- [FI-800] - Fixed the placeholders issue where the empty folders are deleted.
- [FI-829] - Fixed the "Cannot be cast error" when upgrading XLD from 9.5.3 to 10.2.2.
Digital.ai Deploy 10.2.5
Bug fixes
- [ENG-7777] - Fixed the issues with worker's central-config client connecting to the master via LB.
- [FI-787] - Updated the
task.step.execution-threads
property inxl-deploy.conf
file. - [FI-801] - Fixed the issue where upgrade to Deploy 10.2.1 does not copy the security settings for OIDC.
- [FI-831] - Fixed the Worker unable to connect with the Master issue.
Digital.ai Deploy 10.2.6
Bug fixes
- [ENG-7904] - Fixed the issue with stored Cross-Site Scripting (XSS) in the Maintenance banner.
- [ENG-7980] - Fixed the broken Jython patch for issues 2618 and 2894.
Digital.ai Deploy 10.2.7
Bug fixes
- [FI-830] - Fixed the AWS CloudFormation file validation issue.
- [FI-851] - Fixed the memory consumption issue caused when upgrading to Deploy 10.0.5.
Digital.ai Deploy 10.2.8
In addition to bug fixes, the Digital.ai Deploy 10.2.9 includes the following new features:
- Enhanced Kubernetes Operator-based installer
- Enhanced Cluster Auto-downing feature
Kubernetes Operator-based Installation
The Kubernetes Operator-based installer offers the following enhancements:
- Kubernetes Operator now supports installation of Deploy on Google Kubernetes Engine (GKE),Kubernetes On-premise cluster and OpenShift VMWare vSphere clusters.
- You can also upgrade to Deploy 10.2 from Deploy 10.1 (installed using Helm Charts).
Cluster Auto-downing
The Akka framework is upgraded to support the Split Brain Resolver strategy:
For more about Split Brain Resolver, see https://doc.akka.io/docs/akka/current/split-brain-resolver.html
Bug fixes
- [ENG-8336] - Fixed a bug in Plugin synchronization on Windows where the plugin manager is unable to delete old Plugin file on the file system.
- [FI-868] - Fixed the OIDC session time out upon inactivity issue. Fixed the issue with user session timeout handling where user had to re-login despite OIDC session being active. Deploy will now use the already active OIDC session without forcing user to re-login.
- [FI-869] - Fixed the issue with importing
file.Folder CI
that has accented characters. - [FI-872] - Fixed the XL CLI user list fetching issue.
- [FI-880] - Fixed the issue with scheduled tasks where user with proper permissions is not able to see them.
- [FI-888] - Fixed the issue with 'repository.keystore.password' post the Deploy setup.
- [FI-896] - Fixed the issue with cluster AutoDowning where XLD fails to start in hot-standby mode.
- [FI-900] - Fixed the issue with
http.context.root
configuration when accessing Reports. - [FI-907] - Fixed the upgrader cluster configuration issue.
- [FI-910] - Fixed the issue where the accented characters were not displayed in Dictionary upon saving.
Digital.ai Deploy 10.2.9
Note: We recommend that you do not upgrade to this version due to a known issue with Oracle database where the application fails to authorize the user in case of non-admin users. Instead, we recommend that you install the next maintenance version that includes the fix.
Field incidents
- [FI-847] - Fixed the issue where invalid archive is not copied when
Replace Placeholders
is enabled. - [FI-853] - Fixed the latency issue that caused some delay in loading the Deploy application logo.
- [FI-856] - Fixed the issue where the non-admin users cannot view the old archived tasks that were visible before the upgrade.
- [FI-884] - As part of making lock plugin more resource friendly, a new option -
lockRetryAsync
is introduced. When set totrue
(the default value), the retry releases the currently executing thread to the pool instead of blocking it. This will improve the overall system responsiveness. When explicitly set tofalse
, the lock plugin blocks the currently executing thread. - [FI-897] - Updated warnings for unresolved placeholders when updating the deployment.
- [FI-934] - Fixed the crowd integration issue.
- [FI-936] - Fixed the issue that limited maximum number of expressions in a list to 1000.
- [FI-938] - Fixed the deployment issue that occurred when Deploy is started as a service.
- [FI-951] - Fixed the latency issue that caused some delay in loading the Deploy application logo.
- [FI-955] - Fixed the Task archive issue with mysql database.
- [FI-963] - Digital.ai Deploy and Release are not susceptible to
log4j
vulnerability. Though Digital.ai Deploy and Release were bundled withlog4j-to-slf4j
andlog4j-api
files, these JAR files cannot be exploited. While neither these files nor Digital.ai Deploy and Release are susceptible to attack due to thislog4j
vulnerability, some threat detection systems may flag these files. To avoid any such false positive threat indications in future, the latestlog4j
library files (interface and supporting third party libraries) were bundled with both Digital.ai Release and Deploy. For more information, see Log4j Vulnerability to Zero-Day Exploit and Digital.ai Release and Deploy.
Digital.ai Deploy 10.2.10
Field incidents
- [FI-954] - Fixed a UI issue that prevented automatic user session timeouts past the threshold idle time.
Digital.ai Deploy 10.2.11
Bug Fixes and Field Incidents—10.2.11
- [ENG-9080] — Fixed an issue where the cluster data was not properly persisted in the database.
- [FI-953] — Fixed an issue that created a deployment task and indefinitely queued it when wrong hostname was entered for the
wls.Domain
type. - [FI-982] — Fixed an issue that displayed an error in the Register Deployed section even after deployment was cancelled.
- [FI-1006] — Fixed an issue that prevented the display of a particular activity in the monitoring and report pages.
- [FI-1011] — Fixed an issue that prevented the
Run.sh
command from starting the satellite with wrapper Java options.
Digital.ai Deploy 10.2.13
Bug Fixes and Field Incidents—10.2.13
Note: There is no Deploy version 10.2.12.
- D-20597 - Fixed an issue with the Digital.ai Deploy Oracle Service Bus (OSB) plugin that prevented users from being able to deploy packages with placeholders.
- ENG-7906 - Fixed an issue due to which only users with Report-View permission were able to download the generated report.
- D-20598 - Fixed the INTERACTIVE_SUDO connection type vulnerability.
Digital.ai Deploy 10.2.15
Bug Fixes and Field Incidents—10.2.15
Note: There is no Deploy version 10.2.14.
- D-18895 - Fixed an issue with the user session expiry for a deleted user, who has the permission or access to an active session. Now, a deleted user will be denied access to an active session.
- D-20452 - Fixed an issue with Deploy that is set up in HA mode, which shows a Null pointer exception while making changes in Instance customization.
- D-20906 - Fixed an issue with the upgrade of minor releases where Central Configuration files were not copied when using the
previousInstallation
flag.
Digital.ai Deploy 10.2.16
Bug Fixes and Field Incidents—10.2.16
- D-19883 - Fixed an issue that removed the Digital.ai logo up on adding a custom logo.