Note
This repository has been archived. For the latest updates and change log information, please refer to our developer documentation available at: developer.humanitec.com/updates/change-log
This document provides an overview of the changes we are making over time. Feel free to reach out to us in case of any specific questions.
-
New: Introduced a user details page for an overview of user information and RBAC settings. Access it by clicking on a user's name in the Organization Members page.
-
New: Added support for two new properties in the Terraform Driver:
- runner_pod_template: Enables users to configure the pod where the external runner job operates.
- manifests_output: Allows to specify a list of manifests, which the driver will output.
-
Fixed: Resolved issue where long email addresses caused the UI to break on the Organization Members page.
- Humanitec Operator: A new chart version
0.1.7
and app version0.11.0
has been released. The number of provisinoing attempts by the resource driver (3 by default) is now configurable, which addresses the issue of long deployments being marked as timeouts in certain instances.
-
New: Each workload screen in an active deployment in the UI now shows more information on the resource used to provision the workload.
-
New: The Pods status section in a workload view now explicitly highlights any pods running in a previous revision to better visualize the workload's state.
-
Improved: The
artefact
trigger in Humanitec Pipelines now enforces serial execution, so a batch will be kept open until the previous batch has finished running. No explicit concurrency-group setting is required. -
Improved: Terraform Driver now provides a more meaningful message when the service account specified in the external runner configuration does not exist in the target namespace.
-
Fixed: Minor RBAC issue that hid the option to create new applications in UI for users with the Manager role.
-
Humanitec CLI: Version 0.18.0 is now available. Changes since the last announcement:
0.17.0
adding a new--fail
flag forhumctl api
command to exit with a non-zero exit code for response >= 400.0.16.1
introducing the--strict
flag for thescore validate
command, ensuring it fails on warnings.0.15.0
adding the possibility to filter active resource usage by deployment.
-
Humanitec Terraform Provider: A new version
v1.1.0
has been released, adding Terraform resources to manage Workload Profiles and Workload Profile Chart Versions.
-
New: Introduced Deployment Pipelines, which allow to configure multi-step workflows and automate common processes and interactions around Applications and Environments. You can find more information in the updated developer documentation and the release announcement on our blog.
-
New: The Terraform Driver can be configured to run in a target Kubernetes cluster instead of the Humanitec infrastructure. Learn more here.
-
New: Added a rich set of examples for Score and Resource Management related capabilities (Pipelines to follow soon). Explore the full collection here.
-
Improved: Deleting a Resource Definition in UI is now also possible from its details view, not only the list of all resource definitions.
-
Fixed: Resources of type base-env are now consistently listed in the shared resources section of the UI (even in cases of long provisioning times).
-
Humanitec Agent: A new version
1.1.7
has been made available. This release includes a check on agent startup to validate the ids in the ORGS list as valid Humanitec ids. -
Humanitec Operator: A new version has been published. Chart version
0.1.6
and app version0.10.2
including cookies secrets storage optimization (including GCP cleanup). -
Humanitec CLI: Version 0.14.2 is now available. Changes since last announcement:
0.14.2
is now installing shell completions on osx by default when using brew for an improved UX.0.14.0
Introducing additional commands:humctl score init
simplifies your next score configuration with a score starter file.humctl score validate
to validate your score file locally.humctl score available-resource-types
to list available resource types to be used.humctl resources graph
to inspect your resource graph.
-
Humanitec Terraform Provider: A new version
v1.0.1
has been released, introducing example of a resource definition with secret_refs and private git. -
score-humanitec: A new version 0.10.2 has been released.
-
New: Introduced the Delete Active Resources API, allowing to detach an active resource. A detached resource remains intact but is no longer connected to the Platform Orchestrator.
-
Fixed: Switching frequently between Organizations no longer causes loading issues within an Application.
-
Improved: Tables with more than five entries now display a count at the bottom, indicating the total number of items.
-
Humanitec Agent: A new version
1.1.6
has been released:- Fixed a bug related to simultaneous liveness check routines being sent to the same tunnel.
- Updated the wstunnel library version for improved logging.
-
Humanitec CLI: Version 0.13.5 is now available, featuring support for resource accounts.
-
Humanitec Terraform Provider: As part of the
v1.0.0
release, all deprecated resources have been removed. If you haven’t migrated yet, you can pin the provider to the latestv0
version using:
terraform {
required_providers {
humanitec = {
source = "humanitec/humanitec"
version = "~> 0"
}
}
}
-
Improved: The "status" tab on the environments page now updates without requiring a hard refresh in certain instances when loading new deployments. Additionally, the "deployments" tab now displays the latest deployment.
-
Improved: Various minor UI enhancements have been implemented, including:
- Addition of "Created at" and "Created by" details to resource definitions.
- Standardization of timestamp formats across the UI.
- Improved visibility of container logs through timestamp highlighting.
-
Changed: Shared values and secrets are now validated to ensure they don't start with a digit. This prevents potential deployment failures when these values are referenced as part of a placeholder.
-
Humanitec Operator: A new version has been published. Chart version
0.1.4
and app version0.10.0
include:- Support for Azure workload identity.
- Improvements in resource status observability.
- Support for pod labels in the Operator’s chart.
-
score-humanitec: A new version 0.10.1 has been released:
-
Humanitec CLI: A new version 0.13.3 has been made available.
-
New: Introduced a "status" tab on the enviornment page, providing a view of the currently active deployment, including workloads and shared resources.
-
Improved: Status messaging in the UI. This addresses a series of issues related to inconsistent runtime statuses, ensuring a more reliable user experience.
-
Fixed: Matching criteria in UI no longer throw an error if a dot
.
is used in the Resource ID field. This allows to specify values such asmodules.workload-id.externals.my-bucket
orshared.app-dns
for example. -
Fixed: It is now possible to use the Terraform Driver with repositories hosted in Azure DevOps, resolving any previous compatibility issues.
- Humanitec CLI: A new version 0.13.2 has been made available.
-
New: Added support for Resource Classes in the UI. This provides a way for teams to specialize the type of a resource e.g.
external
orsensitive
S3 bucket. -
New: Introduced support for co-provisioning of resources in the UI. This allows for additional resources to be added to the Resource Graph without a Resource Reference required.
-
Improved: Enhanced the design and layout of the replica section in the UI, ensuring consistent status updates and improved error visibility.
-
Fixed: Resolved an issue where signing up with Google failed silently in certain instances. Users can now sign up without any disruptions.
-
Humanitec CLI: A new version 0.13.0 has been made available.
-
Humanitec Operator: A new chart version 0.1.3 (with app version 0.9.9) has been made available.
Happy holidays! Our product engineering team will be taking a break until January 2024. Stay tuned for our next update in the new year!
-
Improved: Placeholder highlighting in the UI now supports Resource Selectors. Resource Selectors extend the syntax of Resource Reference Placeholders by appending either a
>
or a<
symbol. -
Fixed: The organisation switcher in the UI is now accessible from the
/profile-settings
page. -
Improved: Tooltips have been refined to prevent cutoff or overflow issues in the UI.
-
Changed: Humanitec IDs have been updated:
- Minimum length requirement reduced to 2 characters.
- Consecutive dashes are no longer permitted.
-
Fixed: Updating the
port
input of a resource definition in the UI no longer causes unexpected errors. -
Fixed: Private resource dependencies are cleared from the UI upon deletion.
-
Fixed: The “Sign in with Google” button on the login page no longer flickers intermittently in the UI.
-
New: We've introduced a new UI for managing service users in Humanitec, enhancing the security and transparency of API token management. For detailed information, check out the release announcement on our blog.
-
Fixed: The issue causing the UI to freeze when specifying an image by name while creating a new container for a workload has been resolved.
-
Improved: UI timestamps now include the timezone, offering context for tracking work within the UI.
-
New: The Humanitec Terraform driver is now generally available. Following a successful experimental beta phase, we have addressed implementation gaps, resulting in a more robust and reliable user experience. Check out the feature announcement for more details.
-
Improved: Lowered the latency on all authorization-related API calls through a new authorization management system, ensuring faster and more responsive interactions across our platform.
-
Fixed: Issue with the container probe header editing that incorrectly created a new entry instead of replacing content in the existing one.
-
Fixed: Long inputs for shared values no longer overflow in the UI.
-
New: To make Humanitec's UI more intuitive and easier to learn, we have reworked the information architecture of all subpages in the Applications navigation item. You can find more details on the changes in the release announcement on our blog.
-
New: The UI for cloning deployments to an existing environment has been revamped, making it easier for developers to control the state of their target environment. Check out the release announcement on our blog for more information.
-
Fixed: Issue where re-deploying a past deployment would generate two new draft deltas instead of one.
-
New: Added support for the
grpc
type for liveness and readiness probes. -
New: Added support for loading skeletons in the UI to indicate that data is still being loaded.
-
Fixed: Setting a number in the path field of a liveness probe in the UI no longer results in a failed deployment.
-
Fixed: Several minor UX improvements have been implemented:
- Info icon tooltips now open on keyboard focus.
- Validation for workload names has been extended to exclude names starting with a number or a dash, as these result in failed deployments.
- Invalid email addresses can no longer be submitted when adding new members to an organization.
- An error is shown in the UI when a user attempts to sign up with an existing organization name.
-
New: The resource graph API endpoint allows developers to access the resource graph before deployment. This enables them to assess any changes to resources before initiating a new deployment.
-
Improved: After registering for a new Humanitec account, a loading page has been introduced to signal the organization creation process.
-
Improved: Several minor UX improvements have been implemented:
- The visibility of the checkmark icon has been improved in the dark theme.
- The name of a duplicated workload is now limited to a maximum of 50 characters, the same as for non-duplicated workloads.
- The phrasing on the cronjob schedule modal has been adjusted to no longer display "edit" when in view-only mode.
-
Improved: Network API errors are now displayed in the UI during the creation or modification of resource definitions.
-
Fixed: Users who lack the permission to create drafts can no longer see the "Create draft from diff" button on the compare deployments page.
-
Improved: A large number of Container Image versions no longer impairs the performance of the version selection component.
-
Fixed: The CPU settings of a workload are no longer hidden in the UI when creating a delta via the score-humanitec CLI.
-
Fixed: A loading issue related to surfacing routes specified for DNS resources in the UI.
-
Fixed: Inputs can now be reset to a blank string in terraform driver based resource definitions.
-
Fixed: Issues with incorrect formatting after updating JSON inputs in resource definitions.
-
Fixed: Issues with webhook triggers component that prevented users from resetting it to individual triggers after being set to All triggers.
-
Fixed: Values from input fields in read-only mode are now fully accessible and scrollable.
-
Fixed: The autocomplete list of users is no longer exceeding the width of the name input field in Application's People management screen.
-
Fixed: The environment status for a successfully deployed AKS cluster is no longer shown as
stopped
. -
Fixed: Selecting
cancel
on the filter for resource definitions in the UI now successfully clears any configured filters. -
Improved: A number of issues related to content overflowing in the UI:
- Reviewing long environment overrides values no longer requires horizontal scrolling.
- A large number of workloads no longer leads to the cloning modal overflowing.
- Long inputs in the configuration overlay for resource definitions no longer require horizontal scrolling to access the
edit
button.
-
New: Resource Definitions can now be filtered by their details (resource definition ID, resource type, driver type) as well as by matching criteria (environment type, environment ID, application ID, resource ID) on the resource management page.
-
Fixed: AWS SQS queue resources can now be added as shared or private resource dependencies to an application or workload.
-
Improved: The size of text area inputs has been increased on the configuration modals for creating and updating resource definitions to manage large (JSON) inputs more easily.
-
Fixed: Clearning the CPU limits on a workload no longer causes a failed deployment.
-
Fixed: Changes to ingress rules are now rendered correctly in the deployment diff.
-
Fixed: The creation of a data storage resource definition (e.g. Postgres, MySQL, MariaDB) no longer fails silently in the UI.
-
Fixed: The usage of a placeholder (e.g.
platform-${context.env.id}
) for a Virtual Host defined inside a AMQP resource definition is no longer denied by the UI. -
Improved: Placeholders defined inside a Redis resource definition are now surfaced in the UI after the definiton is created or updated.
-
New: Workload
Labels
now support the usage of placeholders. -
Improved: The matching criteria of a resource definition are no longer reordered on click of
delete
. -
Fixed: The configuration of an s3 bucket with the terraform driver is no longer prevented by an error thrown on empty optional input fields.
-
Fixed: The creation date of Humanitec users no longer shows an incorrect date (0001-01-01T00:00:00Z).
-
New: Added a "Deployed values" tab to past deployments which lists all Shared app values and secrets used in the deployment.
-
Improved: More prominent placeholder styling including short descriptions for each placeholder name.
-
Improved: Search filters are no longer case sensitive.
-
Fixed: A syntax error in the cURL request used to connect a new CI pipeline to Humanitec.
-
Fixed: Container images with unknown versions are no longer hidden in the UI.
-
Fixed: Resolved concurrency issues related to adding multiple Shared secrets in parallel with the Humanitec Terraform Provider.
-
New: To increase the transparency of changes in Shared values and secrets and enable more functionality, like reverting to past values, we have introduced Versioning of Shared Values and overrides. Versioning of Shared values and secrets means that every change generates a new version of an app value or environment override. We have been collecting those versions for over six months, and you can access them either through UI or API. In addition to operations on individual values, API enables actions on the entire Value Sets.
-
New: Added support for defining Shared values and secrets to the Humanitec Terraform Provider.
-
Improved: The deployment diff UI has been revamped, making it easier for developers to compare app states and troubleshoot issues. Check out the release announcement on our blog.
-
Improved: Added support for viewing error messages of past failed deployments.
-
Improved: It is now possible to view the active resources of Humanitec's default resource definitions in the "Usage" tab of the Resource Definition Details screen.
-
Improved: The deployment diff now shows changes for newly created applications that haven't been deployed yet.
-
Improved: The design of private and shared resources of running deployments. A direct link to the connected Resource Definition has been added for Organization Administrators.
-
Improved: Driver inputs are now validated against Driver Definition Schemas to catch misconfiguration errors in resource definitions.
-
Fixed: Selecting an asterisk
*
as a matching criterium via the API is now correctly interpreted as "all values". -
Fixed: Deployment diff is now rendered correctly for past deployments with empty properties.
-
Fixed: Setting a Workload Variable's value to an empty string no longer renders as a deleted key in the deployment diff view.
-
Fixed: Input validation of matching criteria on the Resource Definition Details screen. Invalid values can no longer be submitted accidentally.
-
Fixed: Editing of Liveness and Readiness container probe headers. Values are now updated as expected.
-
Fixed: Configuring Memory settings on Windows and Chrome no longer causes unexpected behaviour in the UI.
-
Fixed: Deleting an active resource of a deleted app on the Resource Definition Details screen no longer throws an error.
-
Fixed: Invalid manifests returned by a driver no longer result in an unexpected error. An error stating that the generated manifest is invalid is now shown.
-
Fixed: Creating a template driver that outputs a manifest with an invalid location value no longer silently fails. An error stating that the location in the specific resource definition is invalid is now shown.
-
Improved: The active resources displayed in the
Usage
tab on the Resource Definition Details screen now include a link to the environment they are matched to. -
Improved: Resource provisioning errors now include information about the affected resource definition ID and resource type.
-
Improved: When attempting to define matching criteria that have already been defined for a resource definition of the same type, an error message pointing to the latter is shown.
-
Improved: Added support for artefact versions, environment types and applications to the Humanitec Terraform Provider.
-
Improved: Added support for HTTP/2 & 3 to reduce API latency.
-
Fixed: An interaction bug related to cloning deployments.
-
Fixed: Defining commands in the container overrides section of a workload.
-
Fixed: Users will be redirected to the login page after their session times out.
-
New: Release of a new resource management screen. This UI update includes a redesign of existing features as well as support for new functionality to increase observability and control for platform teams. Check out the feature announcement on our blog to learn more.
-
New: Extended MariaDB and MySQL driver inputs with optional parameter
privilege list
that will be applied to the generated user. -
Fixed: It is now possible to use browser's native Back button to return to the App Overview screen.
-
Fixed: Trailing whitespace characters are no longer removed from Shared App Values and Secrets inputs.
-
Fixed: Creating a delta via the API and navigating to it by updating the draft id in the URL no longer causes the UI to be unresponsive.
-
Fixed: Long deployment comments no longer overflow on the deployment diff dialog box.
-
Fixed: To avoid disruption of ongoing deployments it is no longer possible to trigger a new deployment while a deployment is in progress.
-
Changed: Cloning a whole environment now matches the process for cloning individual workloads. In both cases a new delta is created in the target environment.
-
New: Added support for context placeholder autocomplete in Resource Definitions.
-
Improved: Humanitecs GitHub Actions workflow now allows to skip the building step and use an existing container image instead (Issue #38).
-
Fixed: An interaction bug related to (un-)archiving container image versions.
-
Fixed: A validation issue on the Resource Definition creation modal.
-
New: The Artefact Version API now supports PR refs. Next to branch- and tag refs artefact versions built on a PR branch can be pushed by users.
-
Fixed: Issue preventing users from discarding a draft of a cloned deployment.
-
Fixed: Service account name on Workload Details Screen is no longer reset to
default
after re-submitting the currently defined name. -
Improved: The color palette of the Humanitec UI has been updated to match better the Web Content Accessibility Guidelines.
-
Improved: More intuitive and easier to read design for table components across the application:
- Shared App Values & Secrets on App Settings Screen,
- App Values & Secrets Overrides on Environment Settings Screen,
- Variables, Files, and Volume Mounts on Workload Details Screen.
-
Improved: Responsive design of Application Details Screen.
-
Fixed: Issue preventing users from updating CPU and Memory settings on Workload Details Screen in the UI.
-
Fixed: Workload variables with multiple Placeholders render correctly in the UI.
-
Fixed:
Create new app
input field validation no longer expands beyond the base element. -
Improved: Reworded "mode" error message on Files component on Workload Details Screen.
-
Fixed: Auto-focus on input fields no longer steals focus from the user.
-
Fixed: Modals are no longer hidden on Firefox in the case of overflow.
-
Fixed: The search filter in the images tab on the Organization Settings Page allows to filter images by name.
-
Improved: It is no longer possible to select an archived image version for a conatiner.
-
New: Released Pausing & Resuming of Environments feature which allows you to scale down all of the Workloads of type Deployment to 0 replicas with one click of a button or single API call. Resuming the environment will scale the Workloads back to the number of replicas that were set at the time of pausing. This can help you save costs on unused infrastructure. You can find more information in our documentation.
-
Improved: There is no longer a limit of the number of Applications users can create via the UI.
-
Fixed: The Profile Settings page now correctly lists users' roles on all Organisations they are a member of.
-
Fixed: Resolved an issue where Placeholder autocompletion would not work properly with inputs typed manually and not selected from the dropdown.
-
New: Added autocompletion of placeholders in
labels
,variables
andfiles
properties of a Workload in Deployment Sets. -
Improved: Performance of the
All deployment versions
functionality in the App Details Screen for Applications with a large number of past deployments. -
Fixed: Image version selector no longer requests all versions even if
show archived
option wasn't checked. It caused performance issues for images with a large number of archived versions. -
Fixed: Long Deploy comments no longer affect accessibility and alignment of action buttons in active deployment cards on the App Details Screen.
-
Fixed: An issue related to failing Deployment Diff of a past deployment after adding a Shared Resource.
-
New: Introduced a new API and updated UI which allow for image management at scale. Users can now identify, filter, archive and update images and their versions with ease. More detailed information on new capabilities can be found in the feature announcement article on Humanitec's blog.
-
Fixed: Provisioning a resource with an invalid placeholder now returns the expected error code with a more descriptive message.
-
Fixed: The replicas input field on the Workload Details Page of an active deployment is no longer displayed as enabled for users without the deployer role on the environment type.
-
Fixed: An issue related to the organization name being displayed as empty after users sign up to Humanitec.
-
Fixed: The confirm modal which is displayed when deleting matching criteria of a resource definition that has active resources associated with it is no longer showing duplicate content.
-
Improved: The files editor which is utilised for the configuration of webhooks and files now contains a "show whitespace characters" option to make users aware of potential formatting issues.
-
Improved: YAML input for container command and argument overrides can now be copy-pasted.
-
Improved: The UI for container overrides on the Workload Details Page has been reworked. Command and argument overrides can now be defined as YAML.
-
New: Added support for archiving image versions on the Organization Settings Page. This can be done by selecting the three dot menu of an image and clicking on
archive
next to a given version. Archived images remain accessible for use. An image can beunarchived
anytime.
-
Fixed: Corrected the runtime status for container pods stuck in phase
pending
. The Workload Details Page of an active deployment will now show a warning withNo container status available, pod is pending
. -
Fixed: An issue related log messages being cut off on the Workload Details Page of an active deployment.
-
Improved: Container images are no longer managed via a Humanitec specific ID. Instead, the image name (e.g.
registry.humanitec.io/public/sample-app:1.3.0
) is utilised as an identifier. The UI and documentation for the following features have been updated as a result of this: -
Fixed: An interaction bug related to deletion of resource definitions that have active resources associated with them. The confirmation modal now closes once the deletion has been confirmed.
-
Fixed: Long Workload variables are no longer cut off in view-only mode. The full value can be viewed via a tooltip on hover.
-
Fixed: Long Deploy comments no longer overflow the active and past deployment cards on the App Details Screen.
-
Fixed: Runtime errors triggered by an invalid service account are now exposed on the Workload Details Screen of running deployments.
-
Changed: Added a border as a visual indication for selected app deployments on the App Details Screen.
-
Improved: The deployment status is now displayed for past deployments on the App Details Screen.
-
Improved: Entities are no longer deleted without a warning. The deletion has to be confirmed by the user.
-
Improved: Users attemping to downgrade their organization level or app level role, have to confirm the role change in the UI.
-
Fixed: An interaction bug related to the "Active" toggle of the Automations component on the Environment Settings Screen.
-
New: A loading bar is displayed on page load.
-
Changed: Users have to type in the name of an application to confirm its deletion.
-
Improved: Pending invitations to a Humanitec organization display the time until expiration.
-
New: Removing a resource definition that has active resources depending on it will mark it as
Pending deletion
in the UI until all the resources associated with it are re-deployed.
-
Changed: Implemented a dedicated view-, add-, and edit-mode for several components to give users more control when making changes in the UI. See the App Members section as an example:
-
Changed: The colour contrast of the dark- and light theme according to the Web Content Accessibility Guidelines (WCAG) 2.1.
-
Fixed: Error messages no longer persist between the configuration overlays in the Accounts section of the Organization Settings Page.
-
Improved: The user experience for working with cronjob schedules in the UI. Cron expressions...
-
Fixed: On the Resource Management Page, deleting a matching criteria that has active resources associated with it no longer silently fails in the UI. The action has to be confirmed by the user.
-
Fixed: An interaction bug related to the population of secret values in the shared values and secrets overrides component.
-
Fixed: The handling of placeholder values when configuring a resource definition on the Resource Management Page. As usual, a literal
${
can be inserted by escaping the{
with a backslash.
-
Changed: Namespaces are now managed via Humanitec Resources and are therefore no longer visible on the Environment Settings Screen. A resource definition of type
k8s-namespace
can be created using thehumanitec/static
orhumanitec/template
driver. Learn more here. -
Fixed: An interaction bug related to automatic environment level overrides for certain app default values and secrets.
-
Fixed: Clicking into the
value
field of a secret variable in the shared values and secrets overrides section no longer causes the value to temporarily disappear in the UI.
-
Fixed: Persistent volumes added as shared resources to an Application are no longer missing from the deployment manifest.
-
Fixed: Values defined as part of app default values & secrets can be empty.
-
Fixed: Variable placeholders no longer break if a resource dependancy has the same ID as the Workload.
-
Fixed: Payload entries of webhooks are now preserved in the UI.
-
Fixed: Users that hold the Developer or Owner role on an Application can delete environments of types they have Deployer rights for.
-
Fixed: Selecting "Manage versions" on an image (inside the Images tab on the Organization Settigs Page) that is missing a tag will no longer throw an error in the UI.
-
Fixed: Container states are no longer shared between Workloads when navigating via the breadcrumb menu.
-
Improved: Enhanced error messaging for Azure authentication issues with AKS clusters.
-
Improved: Changes made in the UI after a session expires are no longer lost with no warning. Instead the user will be directed to the login page.
-
New: Added support for Resource References, which allow to reference the outputs of other resources from within Resource Definitions.
-
New: Images can now be added by image name. This allows to create container objects in Workloads that pull from registries not registered with Humanitec as described here.
-
Fixed: An interaction bug related to copying the ingress path on running deployments.
-
Improved: Error messages for driver connection failures on cloudsql postgres database instances are now being fully surfaced in the UI.
-
Fixed: The UI no longer prevents users from updating the credentials of an existing resource definition on the Resource Management Page.
-
Fixed: For existing resource definitions utilizing the
humanitec/template
driver, updating one field will no longer clear the adjacent input field. -
Fixed: The validation of
sub-directories
specified as part of volume mounts on the Workload Details Page. -
Fixed: Changes made to the service account of a Workload are no longer discarded if no container has been added yet.
-
Fixed: Added support for pinning Workloads via the
pin to top
button on Firefox. -
Fixed: A number of small interaction bugs with duplicate or persisting error messages.
-
Fixed: Workloads based on the default-cronjob profile are no longer marked as
failed
in a running deployment when there is no failure. -
Fixed: When editing an existing resource definition on the Resource Management Page:
- Driver input values are reflected correctely in the UI.
- The UI no longer forces an update of secret values.
-
Fixed: When inviting a new user to an Organization that is already a member of a different Organization the UI no longer incorrectly displays an error message.
-
Fixed: Drafts based on deployment clones no longer persist in the UI after being deployed.
-
Fixed: The
Service account
field on the Workload Details Page can now be cleared. -
Fixed: Workloads can be pinned via the
pin to top
button on the App Details Page. -
Fixed: App- and workload level environment variables can have empty values.
-
Improved: The design of several resource icons on the Resource Management Page.
-
Fixed: A deployment issue with
Postgres
resources. -
Fixed: An interaction bug related to editing existing
Volume Mounts
on the Workload Details Page. -
Fixed: Creating a draft from a deployment diff no longer times out in the UI.
-
Improved: Service and container ports configured on the Workload Details Page must be in the range 1 - 65535.
-
Fixed: The authentication flow for active Humanitec users accepting invites to a new Organization.
-
Fixed: A temporary authentication issue related to new users accepting invites to an Organization.
-
Fixed: An interaction bug related to discarding deployment clones in the UI.
-
Fixed: Error messages on the "Create new environment" overlay and the "Account details" form are no longer hidden in the UI.
-
Improved: Nested values are now supported by the resource
Details
overlay on the App Details- and Workload Details Page of running deployments. -
Improved: Error message handling for users attempting to log in without a registered GitHub or Google account.
-
Improved: Existing
Volume Mounts
are now validated in the UI when being updated.
-
New: Added support for Service Ports on the Workload Details Page.
-
Fixed: Corrected form validation on the "Accept Invite" page.
-
Fixed: Switching Organization on the App Details or Workload Details Page no longer causes a "We can't find the app you are looking for" error (403).
-
Fixed: Defining
Volume Mounts
for sharedPersistent Volume
resources no longer causes a deployment to fail. -
Fixed: Several UI issues, including:
- The user-dropdown on the "Manage deployers for x environment type" overlay as well as the image-dropdown on the "Create new container" overlay are no longer cut off.
- Removed redundant "More" link on dropdown menus.
- A number of interaction bugs with discarding deployment clones in the UI.
-
Changed: Initial input validation is triggered
on submit
instead ofon blur
in the majority of cases.
-
Improved: Custom Workload Profiles based on supported features are dynamically displayed and editable in the UI.
-
New: Added support for Labels & Annotations on the Workload Details Page.
-
New: Added support for CronJobs on the Workload Details Page.
-
Fixed: On the App Details Page, Workload error states are now being indicated on the Workload card.
-
Fixed: The edit overlay of Kubernetes cluster resources with the
humanitec/k8s-cluster-eks
driver no longer displays incorrect form parameters. -
Fixed: An Environment Type can now be deleted in the UI if an environment with the same ID exists in the active Organization.
-
Fixed: The "Deploy changes" confirmation overlay no longer reloads after being submitted via keyboard enter.
-
Fixed: Validation of the "Resource ID" matching criterion on the Resource Management Page.
-
Fixed: Several UI issues, including:
- Corrected routing of "Back to" button on the Organization Settings Page.
- Updated "Learn more" link for Workload level error messages.
- Removed duplicate Ingress resource type from Resource Management Page.
- Improved handling of special characters in the Deployment Diff.
-
Fixed: When deleting a shared
Persistent Volume
resource, associatedVolume Mounts
are automatically removed from the Workload as well. -
Fixed: Environment variables no longer throw an error for empty values.
-
Fixed: A Safari browser compatability issue.
-
Fixed: Creating a new webhook no longer throws an error.
-
Fixed: It is no longer possible to switch the driver when editing an existing resource definition.
-
Fixed: A number of smaller interaction bugs with setting resource requirements for CPU and Memory.
-
Fixed: The help modal is fully accessible on the log in page again.
-
Changed: Workloads are added via an overlay which allows users to change the underlying Workload Profile if needed.
-
Improved: A containers maximum and minimum resource requirements for CPU and Memory can now be cleared in the UI. This enables full support for all QoS (Quality of Service) classes in Kubernetes: BestEffort, Burstable and Guaranteed.
-
Fixed: An interaction bug related to selecting "all images except" on the environment automation component.
-
Fixed: An interaction bug related to duplicating Workloads.
-
Fixed: Several UX issues in relation to selecting and deleting draft deployments.
-
Fixed: The scaling of replicas on the Workload Details Page of a running deployment is no longer limited to two. In addition to that can a Workload temporarily be disabled by scaling its replicas down to 0.
-
Fixed: Validation for volume mount path now correctly requires a rooted path with at least 1 directory.
-
Improved: Accessibility according to the Web Content Accessibility Guidelines (WCAG) 2.1:
- Improved app content for screen readers.
- Improved keyboard accessibility of submenus and toggles.
-
New: Added support for updating Resource Account credentials.
-
New: Added a "Status" column to the Organization Members list. After an invitation is sent, the user's status is pending. When the recipient accepts the invitation, their user status switches to active.
-
New: The user role is now visible on the Profile Settings Page.
-
Fixed: Reselecting the currently active Organization in the Organization Menu no longer makes a hard refresh of the App Overview Page necessary.
-
Improved: Users are notified if their currently active environment on the App Details Page has been deleted and are automatically directed to the next environment.
-
Improved: Keyboard accessibility according to the Web Content Accessibility Guidelines (WCAG) 2.1:
- (Un-)pinning of Workloads.
- Editing of Files on the Workload Details Page.
-
Changed: Resources are no longer managed as "Dynamic Resources" and "Static Resources" in the Organization Settings. The implementation of a new Resources Management screen instead allows to manage the provisioning and allocation of resources in one place.
-
New: Next to Environment Types, Resource Definitions can now be matched to Environment Names, Apps as well as Resource IDs in the UI.
-
Changed: Organization Settings and Resources Management were moved from the User Menu to the Organization Menu.
-
Changed: Accounts are now managed from a separate tab in the Organization Settings. Users can dynamically reference existing accounts when configuring resources from the Resources Management screen.
-
New: It is now possible to define headers for "HTTP link" Liveness and Readiness probes on the Workload Details Page.
-
Improved: If a shared DNS resource is deleted, associated Ingress paths are automatically removed from the according Workloads as well.
-
Improved: Accessibility according to the Web Content Accessibility Guidelines (WCAG) 2.1:
- Enhanced keyboard accessibility throughout the application.
- Improved website content for screen readers.
-
Fixed: The UI no longer forces updates of secrets when editing an existing Resource Definition.
-
Fixed: Several interaction bugs when (de-)activating environment automation rules.
-
Fixed: The "Manage versions" overlay for Images in the Organization Settings no longer throws an error for images without an image tag.
-
Fixed: Several UI issues, including:
- The user dropdown on the "Manage deployers" overlay no longer opens by default and blocks the view.
- The "Active" toggle on the Webhooks component is no longer clickable by users in Developer role.
- Error states are no longer shared between account configuration overlays.
- Switching between drafts no longer causes issues on the UI.
-
Improved: Numerous accessibility improvements for keyboard only users.
-
Improved: Better icons for resources.
-
Fixed: Corrected validation for namespace and container names.
-
Fixed: Resources not duplicated when workload duplicated.
-
Fixed: Placeholders for shared resources not expanded in Config Map.
-
Fixed: Better handling of long app and environment names.
-
Fixed: Organization names overflow organization-switcher dropdown.
-
Fixed: Menus at the bottom of lists are truncated as they go off the screen.
-
Fixed: Some resource and workload creation actions were not possible for users with some roles.
-
Fixed: Number of small interaction bugs with managing roles for users.
-
Fixed: Clone functionality in some cases is missing view of detailed changes.
-
Fixed: Discard clone causes a page refresh.
-
New: Added support for custom Workload Profiles in the UI. Next to the default profile
default-module
users now have the option to choose a different profile according to their use case. Container images are added in a second step on the Workload Details Page.
-
New: Resources can now be provisioned on App level. On the Application Details Page, users have the option to add shared resources that can be referenced within the Workloads of their App.
-
New: Added support for partial clones. Instead of cloning the entire deployment by default, users can now choose to include / not include certain Workloads in their clone.
-
New: Added a
...
settings icon to App cards with shortcuts to "View", "Settings" and "Delete". -
New: Users using ad-blockers are now warned that some features will not be available to them in the Humanitec UI. The reason for this is that Humanitec uses a system called Optimizely for UI feature flags. If a user has ad blocking services enabled that block calls to Optimizely, it has to be manually removed from the blacklist in the ad-blocking extension in order to access all Humanitec features.
-
Improved: Container images are now sorted by date.
-
Improved: Several UI components have been transformed into link elements that allows users to open them in a separate browser tab.
-
Changed: The design of several UI components:
- Increased color contrasts in the light theme.
- Increased the font size of label elements.
- Adjusted the design of tooltips.
- Intensified the border color of Workloads in the "Has changes" state.
-
Changed: The following wordings:
External resources
has been renamed toResource dependencies
on the Workload Details Page.- The
Close
button has been replaced with aBack to ...
button.
-
Fixed: Draft deployments are now accessible during running deployments.
-
Fixed: It is no longer possible to create Workloads in deleted Environments.
-
Fixed: Ports defined for liveness and readiness probes in the container configuration section are now displayed on the Workload Details Page of running deployments.
-
Fixed: The Deploy button was renamed to Re-Deploy for active deployments.
-
Fixed: Draft deployments can now be accessed while a deployment is in progress.
-
Fixed: The environment status no longer defaults to "Stopped" after a successful deployment.
-
Fixed: Several UI issues, including:
- Entering multiple Placeholder Paths into the Value field of the Variables component no longer breaks the UI.
- Placeholder Paths in the Value field of the Variables component now support dash characters.
- The "More Environments" dropdown on the App Details Page no longer overflows for large amount of environments.
-
Improved: Keyboard accessibility according to the Web Content Accessibility Guidelines (WCAG) 2.1.
Nothing to report for todays product update as our team has been taking some well earned time off in the past 2 weeks. The next update will be published on August 13th.
-
New: Workload cards can now be pinned to the top of the App Details Page. This allows users to set focus on most relevant workloads in the UI.
-
New: Dynamic resources can now be connected to Humanitec via Azure accounts.
-
Changed: The deploy button has been moved into the draft- and deployment list items on the left hand side of the App Details Page.
-
Improved: Humanitec's accessibility according to the Web Content Accessibility Guidelines (WCAG) 2.1:
- Improved keyboard accessibility with focus on links, table-, and list components.
- Improved navigation flows throughout the application.
-
Fixed: It is now possible to remove previously specified overrides for the container command in the container configuration of the Workload Details Page.
-
Fixed: Environment status (Running/Stopped/Failed) and status icon are now in sync on the Environment Settings Page.
-
Fixed: User names are being shown again on active and past deployments on the App Details Page.
-
New: Added an image build versions overview for each container image registered with Humanitec. This allows to review image versions and filter by commit ID, branch, or tag.
-
New: Added information on resource details to the Workload Details Page of running deployments.
-
Fixed: The environment runtime status is no longer mirrored on all environment tabs of an Application.
-
Fixed: Deleting a volume resource no longer leaves invalid volume mounts in the Workload but instead removes these automatically.
-
Fixed: Deleting a DNS resource now automatically removes all associated routing rules defined in the Ingress section of the Workload.
-
Several UI fixes, including:
- Fixed: The draft version dropdown on the App Details Page no longer overflows for large amount of draft versions.
- Improved: Added an automatic copy interaction to Ingress routes shown on the Workload Details Page of running deployments.
- Improved: Enhanced error message handling in several parts of the application.
-
Changed: Significantly simplified how to connect CI pipelines (like Bitbucket Pipelines, GitHub Actions, and Circle CI) from the Images tab in Organization settings. The CI pipelines just need to notify Humanitec about new image builds. Check out the docs for more information.
-
Fixed: Changes made in a Draft are no longer replicated to running and past deployments in the UI.
-
Fixed: When deleting an App, the Drafts associated with it will no longer show up in subsequently created Apps of the same name.
-
Fixed: Ingress manifests are no longer missing from exported manifests of running and past deployments with allocated Ingress resources.
-
Several UI fixes, including:
- Fixed: When changing a Variable key in a Draft, the old value will now be displayed on the deployment Diff.
- Improved: When adding a file to a Draft, its path is displayed in addition to its mode on the deployment Diff.
- Fixed: Overflowing deployment Diffs and Workload errors are now scrollable.
- Improved: Active deployments that stopped running are marked as "Stopped".
-
New: Added support for path-based routing with Ingress on the Workload Details Page. This feature replaces the Public URL section.
-
Changed: Design and layout of App Settings. Added a danger zone and removed tabs from the UI to allow for a clearer overview and structure of the screen.
-
Improved: Extended error message handling via tooltips in the Variables component for invalid placeholder paths.
-
Fixed: It is now possible to add variables to a workload with multiple containers.
-
Fixed: Configuration of Webhooks.
- Closing the triggers-dropdown no longer removes all selected triggers.
- Existing Webhooks can be disabled/enabled.
-
New: Added support for third-party container registries.
-
Fixed: It is now possible to cancel the creation of a workload on the application details page.
-
Fixed: The image dropdown on the Workload Details Page no longer overflows.
-
Improved: Defined handling of special characters such as "", '<' or '>' in the variables component on the Workload Details Page.
-
Changed: The full image name is now shown for automatic deployment comments.
-
Improvement: We reworked the behaviour of the Variables component on the Workload Details Page. Inconsistencies in terms of its navigation and handling were removed to ensure a reliable user experience for entering, editing and saving data.
- Improved keyboard accessibility.
- Easier handling of Placeholder Paths.
- Corrected behaviour for multi-line input.
-
Fixed: It is no longer possible to cause a failed deployment due to a file being mounted into the root of a container.
-
Fixed: Many UI inconsistencies.
- New: Added support for Persistent Volumes.
Mount paths are specied on the workload level.
We will shift the product update to April 9th due to the short break over Easter.
-
New: Forms for adding resource definitions now present fields in a more intuitive order and include help text.
-
Improvement: Draft can now be discarded even if it is the only draft. (A new blank draft is created in its place.)
-
Fixed: Behaviour of clones in the UI is corrected. Clone includes comment from deployment it was created from and is properly represented in the draft. Code version can be updated in the draft applied to clone.
-
Fixed: Submitting variable value now possible when suggestions drop-down is shown.
-
Fixed: Live container statuses sometimes not shown after deployment.
-
New: Pod status is now communicated on the Workload tile in the Active Deployment mode.
-
Fixed: It is now possible to replace a deleted resource definition with a new one of the same type with overlapping matching criteria while the original one is still in pending deletion state.
-
New: Errors at the Environment level are now surfaced in UI.
-
New: Aiven added as an Account Type & documentation added for connecting Aiven managed databases.
-
Improved: Environments and Deployments pane in Workload Details screen expanded and Environment Tabs restructured.
-
Fixed: Clicking outside the Dynamic/Static Resource Definition dialog does not cause all data to be automatically lost.
-
Fixed: Form for specifying static DNS did not handle TLS certificates correctly.
-
Fixed: Many UI inconsistencies.
-
New: Workloads can now depend on multiple resources of the same type.
-
New: Added support for Aiven (Database-as-a-Service provider).
-
Changed: We changed the default deployment behaviour so that workloads no longer wait for a successful readiness probe before terminating the old pods. This means that the state of the namespace will match the requested deployment, but will mean that pods might be in a failed state. These will be reported in the UI. It will also mean deployments complete much faster. We are working on Blue/Green Deployments to provide a reliable way of ensuring environment wide consistency.
-
Improved: Both, our GitHub Action as well as our CircleCI Orb, now support the specification of additional docker arguments (e.g.,
build-args
). -
Improved: Updated documentation of environment variables & secrets and container configuration.
-
Fixed: Failing application deletion in the case of deployment errors.
-
Fixed: Last environment in an application can no longer be deleted.
Here is our first product update 2021. We started with some important new features into the new year.
-
New: We created an Environment Settings Screen that bundles all environment-specific settings. We moved the Automation Rules and environment-specific Shared values and secrets overrides into this new screen. We now also allow users to specify a namespace in Kubernetes for each environment. This has been supprted by the API but not by the UI.
-
New: We are now allowing users to delete environments directly from the environment dropdown menu on the App Details Screen.
-
New: Out-of-the-box support for Amazon S3 buckets as Dynamic Resources.
-
Improved: Support multi-line values in Shared values and secrets.
-
Improved: Listing of dynamic resources.
-
Fixed: Adding overrides for shared variables cause console error Cannot read property 'id' of undefined.
-
Fixed: Missing UI elements after creating a new environment.
We enjoyed some time off between the years and we hope you did too. Stay tuned for the first product update 2021 which we will publish on January 22nd.
-
New: Add configuration as a file that will be accessible from your container (UI&API).
-
New: Specify Service account that a pod should adopt via UI.
-
New: Create Webhooks Deployment (started/finished) and Environment (created/deleted) events. (UI&API).
-
Improved: Show deployers for all deployments.
-
Improved: Show who created a resource definition.
-
Improved: User experience for re-deploying a historical deployment set.
-
Fixed: It’s not possible to remove command overwrites and args.
-
Fixed: The overview of deploy changes does not show removed env variables.
-
Fixed: Container images dropdown for adding a workload is partially hidden.
-
Fixed: Adding substitute variables for s3bucket.
-
Fixed: Small UI issues on Firefox.
-
Fixed: Blinking smiley behavior :-)
-
Improved: Environments are ordered by creation date instead of alphabet.
-
Improved: Users can choose to delete an environment type and see associated environments.
-
Improved: A lot of great styling updates!
-
Fixed: After adding a managed workload (Redis) the workload dropdown selection is not cleared.
-
Fixed: Selecting placeholder from prompt popup causes currently entered text to be ovewritten.
-
Fixed: After creating a draft from a diff the module detail page is not updated.
-
Fixed: Envtype list is cut off on static resources page.
-
Fixed: After deploying the user can directly see the currently running version instead of the draft.
-
Fixed: Variables are not displayed in the diff if it is removed.
-
Fixed: S3 bucket should only be available for adding to module when a S3 resource definition exists.
-
Fixed: user reported issues with user pagination and user invite via Github.
-
Improved: Based on user feedback, changed Modules to Workloads across application.
-
New: A "My Apps" screen with all applications that a user has an access to as well as their respective environments. The user can choose to open the required application or an environment directly from the My Apps screen.
-
New: Redirect to "My Apps" after deleting an application.
-
New: Introduced an error screen for javascript errors and 404s with a redirect to "My Apps".
-
Fixed: Automated deployments do not show up in UI until they are finished.
-
Improved: Scroll bars across application.
-
New: Deleting Resource Definitions via the UI. If the resource definition has any active resources associated with it, the user is informed about those resources and their suggested replacements. If the user proceeds with the deletion, the resource definition will be marked with pending deletion until all active resources are redeployed.
-
Improved: Input component for shared value overrides.
-
Improved: Snippet for Bitbucket integration.
-
Fixed: Error handling in a case of a blank screen.
-
Fixed: Allowing users to select “None” for static resources that do not need a driver.
-
Fixed: Real-time updates for automated deployments.
-
New: UI and API for adding S3 Bucket resource definition and defining dependency within your modules.
-
New: Welcome emails to newly registered users.
-
Improved: Container runtime status added to application view - users can see at a glance if any of their modules are failing.
-
Improved: Expandable deployment history view. Users can filter based on deployment comment and find their historical deployments easier.
-
Improved: CI integration guide - now presented as a wizzard.
-
New: Export manifests for your chosen deployment set. Manifests can be downloaded as .yaml files and stored in a repository.
-
New: UI support for connecting your own GKE or EKS cluster via Static Resources.
-
Improved: Possibility to add and confirm contact email through profile settings if it is not retrieved during registration.
- New: UI for Runtime status
- View the runtime status for deployed applications and easily debug errors when they occur
- Users are informed if any of the replicas are failing and the affected deployment and modules are marked with a warning
- To simplify debugging, container errors are reported for each replica separately and presented to the user together with the container log
- New: API for deleting resources
- Improved: Live container logs
- Users do not have to refresh to see the latest logs
- Fixed UI issues and logs formatting
- Fixed: Wrong DNS being shown for modules (user reported bug)
- Improved: Container logs
- New: Humanitec CircleCI Orb for simple integration between CircleCI and Humanitec
UI improvements:
Documentation update:
- Release of the API documentation (currently in beta) available here incl. a first tutorial how to use the Humanitec API
General new features:
General new features:
- Allow users to connect dynamic resources of type Postgres with Google Cloud Platform resource account
UI improvements:
New API endpoints:
- API endpoint for deployment automation - allowing users to create, update and delete rules for automated deployments; Humanitec deploy new images automatically based on the defined rules
General new features:
Onboarding support:
General new features:
- Option to create a diff between two deployment sets within one environment
- Allow for patching of environments: apply a diff to a deployment set in any environment
Onboarding support:
- Added sample images that allow for a quicker onboarding (e.g., option to deploy first application without connecting own CI pipelines)
General new features:
- Option to clone deployments to another environment
- Added Elasticsearch and MariaDB as managed images
- Allow to switch between drafts and deployments directly from the module configuration view
New features only available via the API:
- Deploy to your own GCP account (API documentation will be pusblished soon on our docs site)