# What's new?

Here you can find description of changes we've built into each release. While we try our best to make each upgrade automatic and as smooth as possible, there may be some cases where you might need to do something to ensure the application works smoothly. So please read through the release highlights!

## 4.2.4 (current version)
- Fix: Not able to install extensions by name
- Fix: Blank disabled `<Select>` element option (workspace menu)

## 4.2.3
- Filter out invalid contexts when added clusters to Lens
- Add notifications when helm actions fail
- Fix: Extension management page should be more resilient to install errors
- Fix: Loading spinner for custom resources no longer appears when the menu is minimized
- Fix: HPA's not sortable by age
- Fix: Reload pod logs if the values are missing

## 4.2.2
- Ability to install some extensions by their name only
- Fix: Extension Docs' spelling in a few places and improve protocol handler docs
- Fix: CRD display no longer crashes on older versions of kubernetes
- Fix: Current workspace properly tracks active cluster
- Fix: sorting resources by age temporally instead of lexically
- Fix: releases now reload when selecting new namespaces

## 4.2.1
- User is now notified if helm list fails
- Sorting order is now saved when switching views
- Fix: Node shells failing to open
- Fix: Tray icon is now reactive to changes
- Fix: Whole window is used for displaying workspace overview
- Fix: Workspace overview is now reactive to cluster changes
- Fix: Exported ClusterStore now enforces more invariants

## 4.2.0

- Add lens:// protocol handling with a routing mechanism
- Add common app routes to the protocol renderer router from the documentation
- New workspace overview
- New add cluster flow
- Persist Lens UI layout information between restarts.
- Notify about update after it has been downloaded
- Add persistent volumes info to storage class submenu
- Add Pod's image hash as overlay over image name
- Allow to define the path of the shell in app preferences
- Add horizontal scrolling to NamespaceSelect and NamespaceSelectFilter
- Autostart is now always in hidden mode
- Navigation menu in Preferences
- Add terminal clear shortcut for macOS
- Add the ability to hide metrics from the UI
- Add notification to user to add accessible namespaces when needed
- Change Cluster Settings button to be a menu like cluster icon menu
- Fix: Proper sorting resources by age column
- Fix: Events sorting with compact=true is broken
- Fix: Two charts refer to an arbitrary repository
- Fix: Group filtering not working on Custom Resources
- Fix: Font-size on `<code>`
- Fix: Update available notification was able to show twice
- Fix: Cluster-settings page back button navigation is broken
- Fix: Lens not clearing other KUBECONFIG env vars
- Fix: Workspace overview switching and enabled state not being stored storage
- Fix: Extension command palette loading
- Fix: Closing workspace menu after clicking on iframe
- Fix: extension global pages are never able to be visible
- Fix: recreate proxy kubeconfig if it is deleted
- Fix: Proxy should listen only on loopback device
- Fix: Block global path traversal in router
- Fix: Set initial cursor position for the editor to beginning
- Fix: Highlight sidebar's active section

## 4.1.5

- Proxy should listen only on loopback device
- Fix extension command palette loading
- Fix Lens not clearing other KUBECONFIG env vars

## 4.1.4

- Ignore clusters with invalid kubeconfig
- Render only secret name on pod details without access to secrets
- Pass Lens wslenvs to terminal session on Windows
- Prevent top-level re-rendering on cluster refresh
- Extract chart version ignoring numbers in chart name
- The select all checkbox should not select disabled items
- Fix: Pdb should have policy group
- Fix: kubectl rollout not exiting properly on Lens terminal

## 4.1.3

- Don't reset selected namespaces to defaults in case of "All namespaces" on page reload
- Fix loading all namespaces for users with limited cluster access
- Display environment variables coming from secret in pod details
- Fix deprecated helm chart filtering
- Fix RoleBindings Namespace and Bindings field not displaying the correct data
- Fix RoleBindingDetails not rendering the name of the role binding
- Fix auto update on quit with newer version

## 4.1.2

**Upgrade note:** Where have all my pods gone? Namespaced Kubernetes resources are now initially shown only for the "default" namespace. Use the namespaces selector to add more.

- Fix an issue where a cluster gets stuck on "Connecting ..." phase
- Fix an issue with auto-update

## 4.1.1

- Fix an issue where users with rights to a single namespace were seeing an empty dashboard
- Windows: use SHELL for terminal if set
- Keep highlighted table row during navigation in the details panel

## 4.1.0

**Upgrade note:** Where have all my pods gone? Namespaced Kubernetes resources are now initially shown only for the "default" namespace. Use the namespaces selector to add more.

- Change: list views default to a namespace (instead of listing resources from all namespaces)
- Command palette
- Generic logs view with Pod selector
- In-app survey extension
- Auto-update notifications and confirmation
- Possibility to add custom Helm repository through Lens
- Possibility to change visibility of common resource list columns
- Suspend / resume buttons for CronJobs
- Allow namespace to specified on role creation
- Allow for changing installation directory on Windows
- Dock tabs context menu
- Display node column in Pod list
- Unify age column output with kubectl
- Use dark colors in Dock regardless of active theme
- Improve Pod tolerations layout
- Lens metrics: scrape only lens-metrics namespace
- Lens metrics: Prometheus v2.19.3
- Update bundled kubectl to v1.18.15
- Improve how watch requests are handled
- Helm rollback window with more details
- Log more on start up
- Export PodDetailsList component to extension API
- Export Wizard components to extension API
- Export NamespaceSelect component to extension API

## 4.0.8

- Fix: extension cluster sub-menu/page periodic re-render
- Fix: app hang on boot if started from command line & oh-my-zsh prompts for auto-update

## 4.0.7

- Fix: typo in Prometheus Ingress metrics
- Fix: catch xterm.js fit error
- Fix: Windows tray icon click
- Fix: error on Kubernetes >= 1.20 on object edit
- Fix: multiline log wrapping
- Fix: prevent clusters from initializing multiple times
- Fix: show default workspace on first boot

## 4.0.6

- Don't open Lens at OS login by default
- Disable GPU acceleration by setting an env variable
- Catch HTTP Errors in case pod metrics resources do not exist or access is forbidden
- Check is persistent volume claims resource to allowed for user
- Share react-router and react-router-dom libraries to extensions
- Fix: long list cropping in sidebar
- Fix: k0s distribution detection
- Fix: Preserve line breaks when copying logs
- Fix: error on api watch on complex api versions

## 4.0.5

- Fix: add missing Kubernetes distro detectors
- Fix: improve how Workloads Overview is loaded
- Fix: race conditions on extension loader
- Fix: pod logs scrolling issues
- Fix: render node list before metrics are available
- Fix: kube-state-metrics v1.9.7
- Fix: CRD sidebar expand/collapse
- Fix: disable oh-my-zsh auto-update prompt when resolving shell environment
- Add kubectl 1.20 support to Lens Smart Terminal
- Optimise performance during cluster connect

## 4.0.4

- Fix errors on Kubernetes v1.20
- Update bundled kubectl to v1.17.15
- Fix: MacOS error on shutdown
- Fix: Kubernetes distribution detection
- Fix: error while displaying CRDs with column which type is an object

## 4.0.3

- Fix: install in-tree extensions before others
- Fix: bundle all dependencies in in-tree extensions
- Fix: display error dialog if extensions couldn't be loaded
- Fix: ensure only one app instance

## 4.0.2

We are aware some users are encountering issues and regressions from previous version. Many of these issues are something we have not seen as part of our automated or manual testing process. To make it worse, some of them are really difficult to reproduce. We want to ensure we are putting all our energy and effort trying to resolve these issues. We hope you are patient. Expect to see new patch releases still in the coming days! Fixes in this version:

- Fix: use correct apiversion for HPA details
- Fix: use correct apiversion fro CronJob details
- Fix: wrong values in node metrics
- Fix: Deployment scale button "minus"
- Fix: remove symlink on extension install and manual runtime uninstall
- Fix: logs autoscroll behaviour
- Performance fixes

## 4.0.1

- Extension install/uninstall fixes
- Fix status brick styles in pod-menu-extension
- MacOS: fix error on app start
- Performance fix: query all objects using single api call if admin and namespace list is not overridden
- Extension API fix: register a cluster page component properly to a route

## 4.0.0

- Extension API
- Improved pod logs
- Mechanism for users to specify accessible namespaces
- Tray icon
- Support networking.k8s.io/v1 for Ingress
- Add last-status information for container
- Add LoadBalancer information to Ingress view
- Add search by ip to Pod view
- Add Ready status column in the Deployment view
- Add +/- buttons in scale deployment popup screen
- Add stateful set scale slider
- Move tracker to an extension
- Ability to restart deployment
- Status bar visual fixes
- Update chart details when selecting another chart
- Use latest alpine version (3.12) for shell sessions
- Open last active cluster after switching workspaces
- Replace deprecated stable helm repository with bitnami
- Catch errors return error response when fetching chart or chart values fails
- Update EULA url
- Change add-cluster to single column layout
- Replace cluster warning event polling with watches
- Detect more Kubernetes distributions
- Performance fix when cluster has lots of namespaces
- Store more than largest kube api request amount in the event store
- Fix pod usage metrics on Kubernetes >=1.19
- Fix proxy upgrade socket timeouts
- Fix UI staleness after network issues
- Fix errors on app quit
- Fix kube-auth-proxy to accept only target cluster hostname
- Fix link to metrics stack resources

## 3.6.9
- Use Alpine 3.12 for node shell sessions
- Fix errors on app quit
- Fix kube-auth-proxy to accept only target cluster hostname

## 3.6.8
- Fix cluster connection issue when opening cluster settings for disconnected clusters
- Fetch available Helm repositories from Artifact HUB
- Check if user is cluster admin before opening cluster dashboard
- Fix issue when application is disconnecting too fast from pod shell
- Fix UI staleness after network issues

## 3.6.7
- Fix cluster dashboard opening when cluster is initially offline

## 3.6.6
- Fix labels' word boundary to cover only drawer badges
- Fix cluster dashboard opening not to start authentication proxy twice
- Fix: Refresh cluster connection status also when connection is disconnected

## 3.6.5
- Prevent drawer close when revealing secret value
- Fix app crash when CRD conditions were not present
- Add support for Stacklight prometheus metrics
- Terminal: set NO_PROXY env for localhost communication
- Fix CPU/Memory usage metrics when using prometheus operator
- Display last-applied-configuration annotation
- Fix Notifications not to block items not visually under them from being interacted with
- Fix side bar not to scroll after clicking on lower menu item
- Auto-select context if only one context is present in pasted Kubeconfig
- Fix background image of What's New page on white theme
- Reduce height on draggable-top and only render it on macos
- Download dir option is now consistent with other settings
- Allow to add the same cluster multiple times
- Convert bytes in memory chart properly
- Fix empty dashboard screen after cluster is removed and added multiple times in a row to application
- Dropdowns have pointer cursor now
- Proxy kubectl exec requests properly
- Pass always chart version information when dealing with helm commands
- Fix app crash when conditions are not yet present in CRD objects
- Fix kubeconfig generating for service account
- Update bundled Helm binary to version 3.3.4
- Fix clusters' kubeconfig paths that point to snap config dir to use current snap config path


## 3.6.4
- Fix: deleted namespace does not get auto unselected
- Get focus to dock tab (terminal & resource editor) content after resize
- Downloading kubectl binary does not block dashboard opening anymore
- Fix background image of What's New page on white theme

## 3.6.3
- Fix app crash on certain situations when opening ingress details
- Reduce app minimum size to support >= 800 x 600 resolution displays
- Fix app crash when service account has imagePullSecrets defined but the actual secret is missing
- Fix words in labels to be selectable either by hovering or double-clicking

**Known issues**

- Kubectl exec command does not work in terminal against clusters that are behind a load balancer and require Host header in request, for example Rancher clusters.

## 3.6.2
- Fix terminal connection opening

**Known issues**

- Kubectl exec command does not work in terminal against clusters that are behind a load balancer and require Host header in request, for example Rancher clusters.

## 3.6.1
- Inject Host header to k8s client requests
- Remove extra refreshEvents polling
- Fix windows installer when app directory removed manually

**Known issues**

- Kubectl exec command does not work in terminal against clusters that are behind a load balancer and require Host header in request, for example Rancher clusters.

## 3.6.0
- Allow user to configure directory where Kubectl binaries are downloaded
- Allow user to configure path to Kubectl binary, instead of using bundled Kubectl
- Allow user to select Kubeconfig from filesystem
- Show the path of the cluster's Kubeconfig in cluster settings
- Store reference to added Kubeconfig files
- Update logo
- Update Kubectl versions used with Lens
- Update Helm binary version
- Add support for PodDisruptionBudgets
- Add port-forwarding for containers in pod
- Add shortcut keys to menu items
- Improve light theme support
- Show GKE ingress IP
- Allow to remove clusters from right click
- Allow to trigger cronjobs
- Show devtools in menu
- Open last active cluster as default
- Log application logs also to log file
- Fix Dialog Esc keypress behavior
- Set new workspace name restrictions
- Fix cluster's apiUrl
- Fix: Cluster dashboard not rendered
- Fix app reload in cluster settings
- Fix proxy kubeconfig file permissions
- Move verbose log lines to silly level
- Add path to auth proxy url if present in cluster url
- Fix path validation message
- Fix: Refresh input values on cluster change
- Fix margins in cluster menu
- Restrict file permissions to only the user for pasted kubeconfigs
- Close Preferences and Cluster Setting on Esc keypress
- Fix: Update CRD api to use preferred version and implement v1 differences
- Fix: Allow to drag and drop cluster icons
- Fix: Wider version select box for Helm chart installation
- Fix: Reload only active dashboard view, not the whole app window
- Fix cluster icon margins
- Fix: Reconnect non-accessible clusters on reconnect
- Fix: Remove double copyright
- Fix: too narrow sidebar without clusters
- Fix app crash when iterating Events without 'kind' property defined
- Detect non-functional bundled kubectl
- Fix format duration rounding days error
- Handle unsupported resources properly after they've been created from editor
- Fix CRD api parsing
- Fix: allow to edit Endpoint resources
- Fix: handle status values that contains an object
- Fix: incorrect path to install/uninstall feature
- Fix: increase timeout when doing port-forward
- Fix: change manifests order for Metrics feature
- Fix: Master donut graph for memory usage only appears to show one master
- Fix: Error during creation of Kubernetes secret
- Fix: Show age of resource in seconds
- Fix: Node shell pods are pending
- Fix: Wrong created time in resource details

## 3.5.3
- Updated [EULA](https://k8slens.dev/licenses/eula.md)

## 3.5.2
- Fix application not opening properly in some cases by catching and logging error from shell sync.

## 3.5.1
- Fix kubernetes api requests to work with non-"namespaces" pathnames
- Fix: Handle invalid metrics responses properly
- Fix: Display namespace defined in kubeconfig always in the namespace selector
- Fix: Make sure that secret is defined before trying to decode
- Update Helm to 3.2.4
- Fix pasting unicode into terminal sometimes not working
- Fix: Open external links in web browser
- Fix kubectl binary version check

## 3.5.0
- Dynamic dashboard UI based on RBAC rules (hides non-accessible menus)
- Show object reference for all objects
- Unify scrollbars/paddings
- New logo
- Remove Helm release update checker
- Improve Helm release version detection
- Show owner reference on all resource details
- Fix: add arch node selector for hybrid clusters
- Fix pod shell command on Windows
- Fix app freeze after closing terminal on Windows
- Fix: use correct kubeconfig context on terminal when switching cluster
- Fix error when closing Lens on Windows
- Fix: deploy kube-state-metrics component only to amd64 nodes
- Translation correction: transit to transmit
- Remove Kontena reference from Lens logo
- Track telemetry pref changed event
- Integration tests using spectron

## 3.4.0

- Auto-detect Prometheus installation
- Allow to select Prometheus query style
- Show node events in node details
- Enable code folding in resource editor
- Improve dashboard reload
- Provide link to configMap from pod details
- Show system roles on Roles page
- Terminal dock tab improvements
- Fix port availability test
- Fix EndpointSubset.toString() to work without ports
- Return empty string if Helm release version is not detected
- Delay webview create on cluster page
- Fix no-drag css
- Fix node shell session regression
- Rebuild locales & fix translation bugs
- Show always Events title in resource details
- Fix missing spaces in container command
- Check also beta.kubernetes.io/os selector for windows pod shell
- Cache terminall shell env
- Cleanup cluster webview loading
- Update metrics feature components
- Update dashboard npm packages

## 3.3.1

- Do not timeout watch requests
- Fix pod shell error if no access to nodes
- Fix list sort by age
- Always refresh stores when object list is mounted
- Update @kubernetes/client-node to 0.11.1

## 3.3.0

- New section: endpoints
- Initial port-forward implementation for services
- Hide object-list applied filters by default
- Display emptyDir medium and size limit
- Show pod terminating status
- Fix default workspace remove
- Fix issues with crd plugins
- Fix use of bundled kubectl
- Clean up legacy references to Kontena
- Fix jobs sorting if condition is empty
- Electron 6.1.10

## 3.2.0

- Render colors in logs
- Add kubectl download mirror select to preferences
- Bundle helm3 binary
- Catch ipc errors on proxy exit
- SelfSubjectAccessReview use 'pods' resource
- Send Content-Type header on response for asset request
- Fix Helm chart version comparison
- Don't close namespace menu on select
- Change terminal fit-to-window icon
- Silence terminal websocket connection errors
- Always end watch stream if connection to kube-api ends
- Xterm v4.4.0

## 3.1.0

- Windows pod shell (powershell)
- Simplified internal architecture (improves watch & metrics stability)
- New icon
- Support `kubernetes.io/role` label for node roles
- Unlink binary download on error properly
- Electron v6.1.9

## 3.0.1

- Fix an issue with bundled kubectl

## 3.0.0

- Login / signup removed
- Prometheus fixes
- Helm v3.1.2
- Updated [EULA](https://lakendlabs.com/licenses/lens-eula.md)

## 2.7.0

- Workspaces
- Helm 3 support
- Improved cluster menu
- Snap packaging
- Add setting to allow untrusted certs for external http traffic
- Minor tweaks & bug fixes

## 2.6.4

- Minor bug fixes

## 2.6.3

- Fix kubectl download issue
- Fix terminal missing HTTPS_PROXY environment variable
- Minor bug fixes

## 2.6.2

- Minor bug fixes

## 2.6.1

- Kubernetes watch API reconnect fix
- Minor bug fixes

## 2.6.0

- More clusters supported; Improvements to cluster authentication
- User Interface for CRDs (Custom Resource Definitions)
- Cluster notifications; Display warning events counter on cluster switcher view
- Support for Microsoft Azure AKS + AAD code flow
- Minor bug fixes

## 2.5.1

- Fix cluster add problem on fresh installs

## 2.5.0

- Light theme
- Per cluster HTTP proxy setting
- Load system certificate authorities on MacOS & Windows
- Reorder clusters by dragging
- Improved in-application documentation
- Minor bug fixes

## 2.4.1

- Minor bug fixes.

## 2.4.0

- Allow to configure Prometheus address per cluster
- Allow to configure terminal working directory per cluster
- Improved new user experience: invitation code is not required anymore
- New cluster settings UI
- Fix OIDC with custom CA
- Use configured HTTP proxy for kubectl downloads
- Fix missing icons and fonts for users working offline or behind firewalls
- Minor bug fixes

## 2.3.2

- Minor bug fixes

## 2.3.1

- Minor cluster connection fixes

## 2.3.0

- Massive performance improvements
- Allow to customize cluster icons
- UI for Pod Security Policies
- Support username/password auth type in kubeconfig
- Minor bug fixes

## 2.2.2

- Minor bug fixes

## 2.2.1

- UI performance improvements
- Respect insecure-skip-tls-verification kubeconfig option
- Network timeout tweaks

## 2.2.0

- Allow to configure HTTPS proxy via preferences
- Do not send authorization headers if kubeconfig has client certificate
- Minor UI fixes

## 2.1.4

- OIDC authentication fixes
- Change the local port range from 9000-9900 to 49152-65535
- Minor UI bug fixes
- Show error details when add cluster fails
- Respect namespace defined in kubeconfig
- Notify about new kube contexts in local kubeconfig

## 2.1.1

- Minor kubeconfig auth-provider fixes.

## 2.1.0

- Don't auto-import kubeconfig
- Allow to import contexts from the default kubeconfig
- Show whats-new page if user running new version of app
- UI performance improvements & minor fixes
- Improved error messages when cluster cannot be accessed
- Improved kubeconfig validation
- Sync environment variables from login shell
- Use node affinity selectors to match OS for metrics pods
- Terminal: zsh fixes
- Terminal: override terminal initscript set KUBECONFIG with Lens provided one
- Handle network issues better
- Menu: show "About Lens" also on Linux & Windows
- Notify if cannot open port on boot
- Improve free port finding
- Sort clusters by name

## 2.0.9

- Wait shell init files are written into the disk
- Use always temp file(s) when applying resources
- Bundle server binaries
- Show errorbox on fatal boot errors
- Let app start, if already logged in, when no networking available

## 2.0.8

- Remove clusters with malformed kubeconfig when initializing clusters
- Show & accept EULA before login
- Download the correct kubectl for 32bit and check kubectl md5sums
- 32bit windows support

## 2.0.7

- Really disable invites when no more left. :)

## 2.0.6

- Remove shell outputs before shell process is started
- Catch kubeconfig load errors better
- Fix app initialization & login timeout cases
- Add Report an Issue to Window menu
- Target linux nodes only in metrics pods

## 2.0.5

- Minor bug fixes.

## 2.0.4

- Enable user invitations in menu
- Better handling for possible errors in kubeconfig authontication
- Kill backend processes on application exit
- Update dashboar UI components to v1.10.0
- Introduce "User Mode" feature for clusters
- Run login shells in embedded terminals
- Fix cluster settings page scroll issue

## 2.0.3

- Enable persistence for metrics collection only if cluster has default storage class available
- Fix cluster online checking

## 2.0.2

- AppImage Linux application packaging
- Ensure correct version of `kubectl` on terminal shell
- Better error handling for manually added cluster configrations

## 2.0.1

- Add information to request invitation

## 2.0.0

Initial release of the Lens desktop application. Basic functionality with auto-import of users local kubeconfig for cluster access.