5.4.4 Release Notes

v5.4.4 Release Notes

Release Date: Mar 9 2022

Important

5.4.4-3 contains a critical SAML security update as well as NSX-T, provisioning wizard and windows domain join automation updates.

The morpheus-ui logging configuration file has changed from logback.groovy to logback.xml in v5.4.4 (/opt/morpheus/conf/logback.xml). The logback.groovy file from previous versions can be removed, and any updates to logback.groovy will not result in any logging configuration changes.

Deprecation Notice

The Venafi and AppDynamics integrations are deprecated in v5.4.4 and will be removed in v5.4.5. AppDynamic will return as a plugin at a later date.

Note: Items appended with 5.x.x are also included in that version

New Features

API & CLI

  • API docs site improvements made to speed slow load times
  • Added CRUD actions for creating credential sets matching UI functionality in Infrastructure > Trust > Credentials
  • Catalog items can now be associated with a secondary logo which is displayed when Morpheus dark mode theme is activated
  • Issuing DELETE requests to the service-plans API now fully deletes the Plan. Previously, the closest you could get to delete functionality was to update the Plan to be inactive.
  • Mute monitoring command is now mute rather than quarantine to match newer verbiage in Morpheus UI. Related commands have been updated to use “mute” verbiage as well
  • Power Schedules can now be incremented in single minute blocks, previously they were incremented in 15 minute blocks
  • Scale down AKS, MKS, GKE, and EKS Kubernetes clusters by removing workers through Morpheus API and CLI
  • Some API endpoints have moved to reflect the menu changes made in Morpheus UI. See Morpheus API docs for details on changes
  • GET response payloads from newer API endpoints no longer include success:true which makes the response more consistent with older endpoints

Bluecat

  • Bluecat DNS Integration added. Bluecat can now be added as a DNS Integration, in addition to the existing Bluecat IPAM integration.

Credentials

  • Adding Amazon and Azure Clouds now supports use of stored credential sets in addition to VMware Clouds which previously supported them
  • Improved handling for scenarios when an external credential store is offline and users are attempting to view or save credentials to it
  • When editing an existing credential, the location (stored internally or in an integrated external store) is now displayed

Kubernetes

  • MKS Kubernetes v1.22 layouts added for Digital Ocean, Google, SCVMM and vCloud Director Cloud types.
  • MKS Kubernetes v1.22 layouts added for Digital Ocean, Google, SCVMM, vCloud Director and Xen Cloud types.
  • Users can scale down MKS (Morpheus Kubernetes Service) clusters by deleting worker nodes when needed

Morpheus

  • Grails framework updated to Grails 5

Nutanix

  • Added support for UEFI boot option

Plugins

  • Improvements made to IPAM custom plugin development tools

Policies

  • Added “Network” scope to Max VM policies which will limit the number of VMs which can be connected to a specific network
  • Added new Policy type for Max Load Balancer Pools

Power Schedules

  • Power Schedules can now be set down to single minute granularity, previously they were incremented in 15 minute blocks

Security

  • CVE-2020-28052 Upgrade bouncycastle to 1.67
  • CVE-2021-29425 Update common-io to 2.7
  • CVE-2021-29425 Upgrade commons-io 2.7.0
  • CVE-2021-32769 : Upgrade micronaut to at least 2.5.9
  • CVE-2021-33813 Upgrade jdom to 2.0.6.1
  • CVE-2021-35517 Upgrade commons-compress to 1.21
  • CVE-2021-36373 Upgrade Ant to 1.9.16 either 1.10.11
  • CVE-2021-37714 Upgrade jsoup to 1.14.2
  • CVE-2021-41269 Upgrade cron-utils 9.1.6
  • CVE-2021-41767 Upgrade guacamole-common to 1.4.0
  • CVE-2021-42392 Upgrade H2 to 2:2.0.206
  • CVE-2021-42550 Upgrade logback-classic to 1.2.8
  • CVE-2022-21700 Upgrade micronaut-core to 1.4.0
  • Upgrade spring-security-web-5.1.13.RELEASE.jar to version 1.26
  • SAML: Critical SAML security update 5.4.4-3

UI

  • Adding the forceAccountId parameter to a URL, along with the appropriate Tenant ID, will now redirect the user to the login page for the correct Tenant if they don’t currently have a login session
  • Further tweaks made to dark mode theme to improve visibility and overall look of some elements

VMware

  • VMware Clouds can now be configured to inventory existing instances from a specific resource pool scope

Fixes

API & CLI

  • Fixed an issue related to paging large lists of objects, such as Instances, in Morpheus API and CLI
  • Fixed an issue that prevented Instance tags from being updated via Morpheus API or CLI
  • Fixed an issue that prevented provisioning using Azure Marketplace images via Morpheus API if the marketplaceOffer value in the config map contained hyphens or no spaces in the offer name
  • Users can no longer attempt to resize the root volume for Azure VMs through Morpheus API or CLI which is not supported and caused problems
  • networks API is updated to handle an array of Tenants for setting Tenant permissions on the Network

AVI

  • Fixed issue related to managing pools for AVI load balancers

Azure

  • Fixed an issue that could cause the Azure Availability Zone to come unset when changing other configurations on an Azure App Blueprint
  • UI error is now surfaced when Azure Marketplace terms haven’t been accepted and a failure occurs as a result. Previously it would just silently fail

BIND DNS

  • Improvements made to BIND DNS integration to smooth the initial integration creation experience

Blueprints

  • Fixed an issue that could cause configured resource pools on App Blueprints not to be saved correctly
  • Visibility settings for power schedules on App Blueprints are now honored properly. Previously even if the power schedule was hidden it would be shown as visible but locked
  • When the virtual image behind a Layout in an App Blueprint changes, storage controller information is now updated accordingly

Catalog

  • Fixed an issue that caused provisioning failures in catalog items if the Layout was set via Inputs in certain ways

Clusters

  • Clouds with “private” visibility and assigned to a Subtenant are now selectable as provisioning targets in the Cluster wizard from the Primary Tenant matching the behavior in Instance and App wizards
  • Improved validation in the Add Cluster wizard to ensure an IP address is entered when a network with static IP is selected

Datastores

  • Fixed an issue that could cause default datastores not to be honored for certain networks or clouds

Domains

Fixed issue with automated Windows Domain joins 5.4.4-2 5.4.4-3

Huawei Cloud

  • Fixed an issue that could prevent existing projects from being selected when integrating a new Huawei Cloud

Kubernetes

  • Fixed issue with adding External Kubernetes Cluster in AWS requiring plan selection
  • Improved static IP address handling for Kubernetes clusters in the Add Cluster wizard
  • Relabeled title of the modal for adding workers to EKS clusters to reduce confusion

MaaS

  • Fixed an issue that could prevent proper stopping and starting of MaaS machines from the Infrastructure menu

MicrosoftDNS

  • MicrosoftDNS entries are now synced correctly when using an intermediate jump server

Morpheus Worker

  • Fixed issue with image uploads using morpheus worker hitting Socket Buffer limit

NSX

  • Fixed enabling dhcp on existing NSX-T segments 5.4.4-3
  • Fixed NSX-T distributed firewall rule source and destination loading issue 5.4.4-3
  • Fixed NSX-T LB pool creation error 5.4.4-3
  • Fixed dchp range validation on NSX-T segment creation 5.4.4-3
  • Fixed subtenant NSX-T Network selection issue 5.4.4-3

OpenStack

  • Errors are no longer thrown when restoring from an OpenStack backup which has moved from its original storage space
  • Improved OpenStack API detection for scenarios when an OpenStack environment has services on multiple domains and subdomains

Option Lists

  • Fixed an issue that caused keys rather than values to be returned when Option Lists were presented as Typeahead fields in Inputs

Oracle Cloud

  • Fixed an issue that could cause Oracle Cloud Instance clone to fail

Policies

  • Subtenant administrators can now set Policies which are scoped to Clouds shared with the Tenant from the Primary Tenant
  • When a Policy is scoped to multiple Tenants, the full list of Tenants can be viewed from the Policies list page by clicking on the info (i) button
  • When scoping a Policy to a Tenant, previously-selected Clouds or Networks on the Policy are no longer cleared after the Tenant is set unless the Tenant does not have access to the Cloud or Network

Provisioning

  • Fixed permission issue with disk when used does not have access to associated Virtual Image record 5.4.4-3
  • Fixed networks being reloaded when layout is changed in wizard 5.4.4-3

Reports

  • OpenStack Instance now show the correct CPU counts on Instance Inventory Summary Reports

Roles

  • Access to create and manage Snapshots no longer requires “Full” access to Infrastructure: Compute and “Read” access to Backups. Users with “Read” access to Infrastructure: Compute and “None” access to Backups are now able to manage Snapshots
  • Removing Roles from users with API tokens generated no longer throws errors

Rubrik

  • Fixed an issue that could cause 500 errors to be thrown when Rubrik backups were selected from an Instance backup tab

SCVMM

  • Fixed an issue that could cause Linux consoles not to work properly for SCVMM Instances

Security

  • Changes made to login session handling to improve application security
  • SAML: Critical SAML security update 5.4.4-3

Security Scans

Fixed permission issue perventing users with security scan role permission from accessing security scans 5.4.4-3

ServiceNow

  • Fixed an issue that could cause provisioning from a ServiceNow integration to fail when naming Policies were in effect

Terraform

  • Fixed an issue caused by applying Terraform state changes when Morpheus naming policies were in place
  • Fixed data loading issue when clicking “i” button on tf resources
  • Fixed issue with Terraform App provisioning status not completing after Approval policy is approved and resources are created.
  • Fixed issue with applying available updates to terraform modules.
  • Fixed issue with passing options in the morpheus-ui terraform command line
  • Fixed issue with tf provisioning on cloud with existing key/value cloud profiles (not terraform cloud profiles)
  • Fixed issue with wrong app to cloud association potentially assigned when multiple clouds of same type are available in the target group

UI

  • “Location” heading renamed to “Addresses” on the Inventory (Instance Detail) page for provisioned Catalog Items
  • A warning message is now surfaced in the UI to let the User know they cannot delete a Spec Template when it is tied to a Layout. Previously the delete action would silently fail which could cause confusion
  • The History tab on an Instance detail page is no longer empty if the User does not have Monitoring: Logs permissions

UpCloud

  • Fixed an issue that caused provisioning to UpCloud to fail under some circumstances

VDI Pools

  • Fixed an issue that could cause VDI sessions not to display properly for SCVMM-based VDI pools

VMware

  • Fixed an issue that caused VMware Clouds to become stuck and unable to be deleted
  • Fixed an issue that prevented provisioning to VMware Clouds shared with a Subtenant and which had just one cluster-type data store
  • Improvements made in syncing process for Morpheus Wiki content with VMware notes fields

Workflows

  • Fixed an issue that caused Input values not to be pre-populated when executing one-off Operational Workflows from the Instance detail page under certain conditions

Appliance & Agent Updates

Appliance

  • morpheus-ui logging configuration file changed from logback.groovy to logback.xml.
2 Likes

This is a huge release that also takes care of some much needed updates on underlying frameworks. Java 11 +, Grails 5+, Spring Framework, and much more. It is possible some regressions may sneak in in this release mainly due to this necessary upgrade. We apologize in advance if anything comes up and are working feverishly to resolve things quickly.

1 Like