2.9 Release Notes

Version Compatibility Matrix

Component Versions
Nirmata Managed Kubernetes
1.16 1.15 1.14 1.13
GKE
1.15 1.14 1.13
EKS
1.14 1.13
AKS
1.16 1.15 1.14 1.13
Kyverno
1.1.3

What’s New in Nirmata 2.9

  • Multi-Catalog Support
  • Granular IAM (Per User RBAC)
  • Resource Management (Pre-configured environments)
  • Identity & Access Management
  • CRD Support
  • Cluster Secret Encryption
  • Resource Change Management - FailedApply state <!—Need to provide original HTML link—>
    • ModelUpdate - Add ResourceChange.failedReason
    • Update ResourceChange.failedReason whenever resource is failedApply.
    • Handle new coming change of resource that already has failedApply changes.
    • Support re-apply of failedApply changes.
    • Support cancel of FailedApply changes.
    • Handle reconciliation of resource that has failedApply changes.
    • Add ResourceChangeMonitor to auto-reapply resourceChange that failed for connection issue.
    • (UI) Add a new panel to show failed resourceChange.
    • (UI) Allow user to view the detail of failed resourceChange in the panel.
    • (UI) Allow user to re-apply or cancel for specific failed resourceChange or all.
    • (UI) Show related cluster-wide resource in application panel.
    • Support resource redeployment.
    • (UI) Show CRD/ClusterRole/ClusterRoleBinding in Application panel of Catalog.
  • Nirmata CLI (nctl) is now available with extended functionalities.
    • REST client - get single object.
    • Add JSON path support for output.
    • Add transformation to output.
    • nctl run from catalog.
    • nctl run from file.
    • Add the -o | --output flag.
    • Format timestamp.
    • nctl environment -e <name> delete app <name>.
    • nctl env set image deploy name -e <env> -r <run> [container=images,...].
    • nctl env apply app <name> -f file -n <name>.
    • Create utility for a standard result line.
    • nctl kubeconfig get/set.
    • nctl login

Additional Feature

  • Support multiple security groups for EKS cluster creation.
  • Allow changing of catalog and application from the running application.
  • Support Services of type ExternalName.
  • Assign teams to environments - Complete CRUD.
  • Create default policies for Kubernetes' best practices.
  • Support Autoscaling for AKS and EKS clusters.
  • Add AKS v1.15.4 and v1.15.5 kubernetes version.
  • Make Calico the default network plug-in to support network policies.
  • Allow changing service type from Nodeport/LoadBalancer to ClusterIP.
  • (UI) Show CRD/ClusterRole/ClusterRoleBinding in Application panel of Catalog
  • Remove actions from all readonly views under Cluster.
  • Add functionality for persistent favorite/all filter setting.
  • Display recently visited applications in catalog and environments [Default Policies Management].
  • Add column in ResourcePolicy table to show ResourcePolicyType.
  • Allow editing Policy-Clusters relation. [Default Policies Management].
  • Allow user to edit ONLY enable/disable for.
  • Allow to upload only one YAML file in Workload Policy creation wizard.
  • Provide a way to import all the kyverno best practices policies.
  • Support Kyverno v1.1.0, v1.1.1, and v1.1.2.
  • Default cluster policies for cloud providers should be for Kubernetes version 1.15.x.
  • Restore entire application from a backup.
  • Support new fields for Velero BackupConfigLocation Kubernetes 1.16.x support.
  • nadm - remove elasticsearch from the installation

Fixes

  • Invalid running count shown for stateful sets.
  • Validation Error while deploying Add-ons YAML from Nirmata UI.
  • Email subject spelling error (Nirmata Monitoring).
  • Elasticsearch backup / restore.
  • Failed to change service type (ClusterIP -> NodePort).
  • When a node is deleted from the node cluster panel, number of desired instances are not updated.
  • Creation of AWS hostgroup shows pending create only.
  • Unable to change default network policies for custom clusters.
  • Tenant does not login user; user service shows tenant mismatch.
  • Pod count shows wrong number of total pods
  • Changes to secrets in an running application are not applied in the cluster.
  • Devops user with view access to particular cluster then should not allow to add velero add-on.
  • Create a button for draining the nodes.
  • Unable to create or edit config map in UI.
  • Cannot deploy Helm charts that include CRDs.
  • Resize cluster dialog takes a long time to load (in a new session).
  • Sometimes direct connect managed cluster fail to create.
  • After upgrading direct connect managed cluster in devtest2 and production, it goes to unknown state.
  • Deploy application after upgrade cluster was unknown state.
  • Default environments are not visible on UI (but able to see 4 namespaces on cluster resource panel).
  • An unknown error occurred while retrieving more activity.(in devtest2).
  • There are no versions availableon UI for Aks cluster upgrade.
  • Former node port services retains node port.
  • React pages does not load sometimes due to undefined state.
  • Showing only used CPU but there in no total CPU for the particular cluster.
  • Discovered cluster isInitialized flag not set to true causing cluster initial flow repeated every time cluster connectedState changed.
  • Nirmata managed cluster creation on AWS EC2 fails CRD import get stuck (hangs).
  • For a particular cluster, seven environments are deployed, by default, but the cluster UI shows only five Environments count.
  • While creating velero-addOn on UI, there are no cloud providers available.
  • While adding policy it is showing No results found in category field.
  • While downloading Inventory report, an extra page is downloaded and which is not required.
  • Optimize loading of environment and applications cards.
  • While deleting some applications, it does not ask for confirmation (confirmation text box).
  • While creating/adding application, it should only allow .yaml/.yml file formats but here it is allowing to import any file formats.
  • While deploying applications sometimes it throws a validation error for some yaml files which is attached below.
  • Application label does not display on UI.
  • For Devops user, it is showing incorrect cluster count in the inventory report.
  • Post patch 2.7.4, large environments are better, but still laggy.
  • Fix Catalog UX.
  • Unsupported kubernetes version v1.11 should remove from EKS cluster creation UI.
  • For devops user recent activity/activity is not displaying in Dashboard UI.
  • ‘Devops’ user with ns-creator permissions has inappropriate access to Nirmata functions.
  • Nirmata application UI not updating events - looks very flaky.
  • nadm fails to scale Nirmata services.
  • Cluster in the prod account is constantly generating state change events.
  • Should not show the Enable encryption option for v1.12.3 cluster version.
  • Sometimes getting java heap space issue while creating AKS cluster in PE.
  • Default add-on catalog categories are not showing up in cluster add-on menu.
  • disallow-default-ns policy is not working properly.
  • Infinite Loading state after a catalog is deleted.
  • API key generation does not work.
  • Nirmata keeps losing context for kubernetes control-plane containers, causing instability.
  • User cannot download kubeconfig file if access at an environment level is granted.
  • User’s see cluster information in user settings page even though user has no access to any of the cluster and has access to one of the environments of the cluster.
  • Catalog list load fails while assigning catalog permissions to a team and works on second try.
  • Helm Repo imports are hard coding nirmata as the default namespace Devops user able to create namespace if he has only view access to particular cluster.
  • Unable to upgrade managed cluster.
  • Scale-up is not working for aws cluster in PE.
  • Jira credentials with certificate do not work.
  • Failed to deploy velero addon in Nirmata PE.
  • Cannot create simulator devices.
  • When a host-group or cluster is created, the cluster/host-group status shows only after refresh.
  • For devops user I gave view access to cluster , inside a cluster if i click on settings Encryption option is blank.
  • The accept self-signed certs option does not work in cluster policy.
  • Remove EKS 1.12 cluster version.

Known Issues

  • While adding jira failed to retrieve issue type id.
  • Failed to create EKS cluster in Nirmata PE.(unknown error)