CHANGELOG.md

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

NOTE: As semantic versioning states all 0.y.z releases can contain breaking changes in API (flags, grpc API, any backward compatibility). We use ⚠️ Breaking change: ⚠️ to mark changes that are not backward compatible (relates only to v0.y.z releases).

Unreleased

Added

• #213: [techdocs] Add TechDocs plugin, to access the documentation for your services within kobs.
• #215: [azure] Add Azure plugin, to monitor your Azure resources.
• #219: [azure] Add permissions for Azure plugin, so that access to resources and actions can be restricted based on resource groups.
• #220: [azure] Add auto formatting for the returned metrics of a container instance and fix the tooltip positioning in the metrics chart.
• #221: [azure] ⚠️ Breaking change: ⚠️ Add support for kubernetes services and refactor various places in the Azure plugin.
• #222: [azure] Simple cost management on subscription level
• #224: [harbor] Add support for multi-arch images.
• #226: [klogs] Add option to sort fields.
• #228: [azure] ⚠️ Breaking change: ⚠️ Add support for virtual machine scale sets and change required values to display metrics from Azure.

Fixed

• #227: [techdocs] Ignore comments in code blocks in table of contents.
• #234: [azure] Fix json tags in Azure permissions struct.

Changed

• #217: [azure] Use resource groups to get a list of container instances.
• #225: [core] ⚠️ Breaking change: ⚠️ Change options handling accross all plugins and re-add time property.
• #229: [opsgenie] Allow users to overwrite the selected time range in a dashboard for an Opsgenie panel via the new interval property.
• #230: [dashboards] Add special variables __timeStart and __timeEnd for dashboards.
• #231: [klogs] Highlight expanded row and do not use index as key. The same changes were also applied for the Elasticsearch plugin.
• #232: [core] Change options handling for various plugins.
• #233: [resources] Highlight expanded row for containers in Pod details.

v0.7.0 (2021-11-19)

Added

• #177: [istio] Add Istio plugin to get an overview of all applications in the service mesh.
• #182: [istio] Add top and tap commands and add details view for metrics.
• #183: [istio] Add panels and documentation and use the destination_app label instead of destination_workload for the Prometheus metrics.
• #184: [clickhouse] Add aggregations to visualize logs.
• #187: [clickhouse] Add options to download returned logs as .log or .csv file.
• #192: [clickhouse] Add options to download aggregation results as .csv file.
• #193: [elasticsearch] Adjust selected time range via logs chart and allow filtering via fields.
• #201: [sonarqube] Add SonarQube plugin to view projects and their measures within kobs.
• #202: [core] Add tooltip to refresh button to show selected time interval.
• #204: [grafana] Add Grafana plugin, to show dashboards from a Grafana instance and to embed Grafana panels into kobs dashboards.
• #205: [resources] Add support to download/upload files from/to a container.
• #208: Add documentation about how to Develop a Plugin.
• #209: [harbor] Add Harbor plugin, to show projects, repositories and artifacts including there vulnerabilities and build history from a Harbor instance.

Fixed

• #175: [prometheus] Fix tooltip when no unit is provided.
• #186: [jaeger] Fix tooltip position in traces chart.
• #189: [clickhouse] Fix download of .csv fiels.
• #191: [clickhouse] Fix returned logs, when user selected a custom order.
• #196: [elasticsearch] Fix number of documents.
• #199: Add volumes and volumeMounts properties to the Helm chart, to support the kubeconfig provider.

Changed

• #179: [clickhouse] Add sorting and show milliseconds in time column.
• #181: [core] ⚠️ Breaking change: ⚠️ Make the time selection across all plugins more intuitive. For that we removed the time property from the Options component and showing the formatted timestamp instead.
• #185: [clickhouse] Use pagination instead of Intersection Observer API to display logs.
• #188: [sql] Replace the GetQueryResults function with the implemention used in the Clickhouse plugin, to have a proper handling for float values.
• #190: [core] Unify list layout across plugin.
• #194: [elasticsearch] Use pagination instead of infinite scrolling to display logs.
• #195: [istio] Display upstream cluster instead of authority in the React UI and ignore query string in path.
• #197: [clickhouse] Change break down filter for aggregation, to allow more complex filters.
• #210: [klogs] ⚠️ Breaking change: ⚠️ Rename clickhouse plugin to klogs.

v0.6.0 (2021-10-11)

Added

• #103: Add option to get user information from a request.
• #104: Add actions for Opsgenie plugin to acknowledge, snooze and close alerts.
• #105: Add Prometheus metrics for API requests.
• #112: Allow mapping values in Prometheus table panel.
• #113: Allow and improve customization of axis scaling.
• #116: Open details when clicking on Jaeger chart nodes.
• #117: Change Jaeger chart node color to red for erroneous traces.
• #119: Add Flux plugin to view and reconcile Flux resources.
• #122: Add ClickHouse plugin, to query show logs ingested by the kobsio/fluent-bit-clickhouse Fluent Bit plugin.
• #124: Add sql mode for ClickHouse to execute raw SQL queries.
• #126: Show request details when gettings logs from ClickHouse.
• #127: Allow ILIKE queries for ClickHouse logs, using the new =~ operator.
• #128: Allow users to specify dashboards within a Team or Application via the new inline property.
• #131: Add chart which shows the distribution of the logs lines in the selected time range for the ClickHouse plugin.
• #132: Support the download of log lines in their JSON representation in the ClickHouse and Elasticsearch plugin.
• #136: Allow custom order for the returned logs and add !~ and _exists_ operator for ClickHouse plugin.
• #138: Add option to stream (follow) logs via WebSockets.
• #149: Add SQL plugin to run queries against a configured SQL database instance. For now we are supporting the clickhouse, postgres and mysql driver.
• #151: Add actions for in log details view of the ClickHouse, so that users can filter based on the value of a field.
• #156: Add tags for applications.
• #159: Allow users to select a time range within the logs chart in the ClickHouse plugin.
• #160: Allow users to sort the returned logs within the documents table in the ClickHouse plugin.
• #161: Add support for materialized columns, to improve query performance for most frequently queried field.
• #162: Add support to visualize logs in the ClickHouse plugin.
• #170: Add Custom Resource Definition for Users.
• #171: ⚠️ Breaking change: ⚠️ Add authentication and authorization mechanism for plugins and resources. These settings are configured via the --api.auth.<SETTING> flags. Permissions are always granted on a team level. A users gets all permissions of the team where he is a member of.
• #172: Add authentication and authorization settings to Helm chart.

Fixed

• #102: Fix GitHub Action for creating a new Helm release.
• #109: Fix tooltip position in Prometheus charts.
• #110: Fix Dashboard tabs showing wrong variables.
• #111: Fix usage of memo in Dashboards and fix resources table for CRDs when a value is undefined.
• #114: Fix span order and additional fields for Jaeger plugin.
• #115: Fix maximum node size of the Jaeger chart and utilize useMemo to cache computation for it.
• #118: Fix null is not an object (evaluating 'e[Symbol.iterator]') error for Prometheus charts.
• #120: Fix reconcilation of Flux resources.
• #123: Fix fields handling in ClickHouse and Elasticsearch plugin.
• #125: Fix missing return statement in ClickHouse panel.
• #129: Fix handling of traces in the Jaeger plugin by using some function from the jaegertracing/jaeger-ui.
• #134: Fix time in log buckets of the ClickHouse plugin.
• #135: Fix read and write i/o timeouts for web terminal.
• #143: Fix a bug in the Jaeger plugin, where results were not refreshed after a user selected another service, operation, etc.
• #146: Fix logic for long running requests introduced in #144.
• #163: Remove legend from logs chart and use useDimensions hook.
• #168: Fix missing encoding of query parameter in ClickHouse plugin.

Changed

• #106: ⚠️ Breaking change: ⚠️ Change Prometheus sparkline chart to allow the usage of labels.
• #107: Add new option for Prometheus chart legend and change formatting of values.
• #108: Improve tooltip position in all nivo charts.
• #121: ⚠️ Breaking change: ⚠️ Allow multiple queries in the panel options for the Elasticsearch plugin.
• #130: ⚠️ Breaking change: ⚠️ Allow multiple queries in the panel options for the Jaeger plugin.
• #133: Improve querie performance to get logs from ClickHouse.
• #137: Change log view for the ClickHouse and Elasticsearch plugin.
• #139: Update Go and JavaScript dependencies.
• #140: Fill the chart for the distribution of the log lines with zero value.
• #141: Add node_modules to .dockerignore.
• #144: Avoid timeouts for long running requests in the ClickHouse plugin.
• #147: Improve query performance for ClickHouse plugin and allow custom values for the maximum amount of documents, which should be returned (see #133).
• #148: Improve reliability of kobs, by do not checking the database connection for a configured ClickHouse instance.
• #150: ⚠️ Breaking change: ⚠️ The ClickHouse plugin can now only be used together with the kobsio/fluent-bit-clickhouse output plugin for Fluent Bit. For raw SQL queries against a ClickHouse instance the SQL plugin added in #149 can be used.
• #152: Improve performance for large dashboards and open Application page in gallery view.
• #155: Allow users to get all Applications from all namespaces, by allowing an empty namespace list.
• #157: Imporve query performance for ClickHouse plugin.
• #158: Improve Pormetheus HTTP metrics and create a metric to find the most used fields in queries.
• #164: Improve chart handling across plugins.
• #167: Improve styling across plugins.
• #169: Rework home page to include plugins, applications and teams.
• #173: Move CRD interfaces to core plugin, so that we can use them within the plugin.

v0.5.0 (2021-08-03)

Added

• #81: Add markdown plugin, which can be used to render a markdown formatted text in a dashboard panel.
• #83: Extend Kubernetes resource with Teams, Applications and Dashboards via annotations.
• #84: Add actions for resource, so that they can be modified or deleted within kobs.
• #87: Rework Kiali plugin to show the topology chart from Kiali for a list of namespaces.
• #89: Rework Opsgenie plugin to show alerts and incidents from Opsgenie.
• #91: Add force delete option for Kubernetes resources.
• #92: Preparation to build a own version of kobs using the kobsio/app template.
• #93: Show status of Kubernetes resource in the table of the resources plugin.
• #97: Add support for Kiali metrics.
• #98: Add terminal support for Kubernetes Pods.
• #100: Add support for Ephemeral Containers.

Fixed

• #94: Fix variable handling for dashboards.
• #99: Fix WebSocket connections for the Kubernetes terminal in environments with an idle timeout for all connections.

Changed

• #82: Improve error handling for our API.
• #85: Improve overview page for Pods, by displaying all Containers in an expandable table and by including the current resource usage of all Containers.
• #86: Improve overview page for Nodes, by displaying the resource metrics for the CPU, Memory and Pods.
• #88: Improve handling of actions for Kubernetes resources.
• #95: It is now possible to get Kubernetes resources for all namespaces by not selecting a namespace from the select box on the resources page.
• #96: Add RSS plugin to show the latest status updates of third party services.
• #101: Show logs in the terminal.

v0.4.0 (2021-07-14)

Added

• #74: Add new Custom Resource Definition for Dashboards as a replacement for Templates.
• #75: Add placeholder for Dashboards, which allows users to pass custom values to Dashboards.
• #76: Add support for variables and time ranges in Dashboards.

Changed

• #71: ⚠️ Breaking change: ⚠️ Remove protobuf as requirement and rework project structure for better plugin support.
• #73: ⚠️ Breaking change: ⚠️ Add a new version for the Applications and Teams Custom Resource Definitions (v1beta1).
• #77: ⚠️ Breaking change: ⚠️ Rework Prometheus plugin.
• #78: ⚠️ Breaking change: ⚠️ Rework Elasticsearch plugin.
• #79: ⚠️ Breaking change: ⚠️ Rework Jaeger plugin.
• #80: Adjust documentation, demo, Helm chart and Kustomize files for the new CRDs and plugins.

v0.3.0 (2021-06-03)

Added

• #45: Add value mappings for sparkline charts in the Prometheus plugin.
• #49: Add new chart type table for Prometheus plugin, which allows a user to render the results of multiple Prometheus queries in ab table.
• #51: Add new command-line flag to forbid access for resources.
• #52: Add option to enter a single trace id in the Jaeger plugin.
• #56: Add actions for Elasticsearch plugin to include/exclude and toggle values in the logs view.
• #58: Add plugin support for Teams. It is now possible to define plugins within a Team CR, which are then added to the teams page in the React UI.
• #59: Add support for Templates via the new Templates CRD. Templates allows a user to reuse plugin definitions accross Applications, Teams and Kubernetes resources.
• #60: Add support for additional Pod annotations and labels in the Helm chart via the new podAnnotations and podLabels values.
• #63: Add Kiali plugin (in the current version the Kiali plugin only supports the graph feature from Kiali).
• #66: Add edge metrics for Kiali plugin.

Fixed

• #43: Fix hosts and gateways list for VirtualService in the Helm chart.
• #44: Add default logo for teams, which is shown when a team doesn't provide a logo and improve metrics lookup for Prometheus plugin.
• #50: Fix determination of the root span in the Jaeger plugin.
• #54: Fix fields handling in Elasticsearch plugin.

Changed

• #46: Support multiple types for the legend in a Prometheus chart and use a custom component to render the legend.
• #47: Display the legend at the Prometheus page as table and use color of selected metric in chart.
• #53: Improve Jaeger plugin, by allow filtering of services and operations and adding several actions for traces.
• #55: Allow a user to add a tag from a span as filter in the Jaeger plugin.
• #57: Visualize the offset of spans in the Jaeger plugin.
• #61: Improve caching logic, by generating the teams and topology graph only when it is requested and not via an additional goroutine.
• #62: Show the name of a variable within the select box in the Prometheus dashboards.
• #64: Recreate Pods when ConfigMap in Helm chart is changed.
• #67: ⚠️ Breaking change: ⚠️ Adjust Pod and Service labels, which can now be set via the pod.labels, pod.annotations, service.labels and service.annotations values.

v0.2.0 (2021-04-23)

Added

• #29: Add a new dependencies section to the Application CR. These dependencies are used to show a topology graph for all Applications.
• #31: Add plugin support for Kubernetes resources.
• #32: Add support for container logs via the Kubernetes API.
• #34: Add a new Custom Resource Definition for Teams. Teams can be used to define the ownership for Applications and other Kubernetes resources. ⚠️ Breaking change: ⚠️ We are now using the apiextensions.k8s.io/v1 API for the Custom Resource Definitions of kobs.
• #39: Add Opsgenie plugin to view alerts within an Application.
• #40: Add metric name suggestions for Prometheus plugin.
• #41: Add overview and Pods tab for resource details.
• #42: Add VirtualService specification for Istio to the Helm chart.

Fixed

• #33: Fix the topology graph, which crashes, when an Application has a dependency to an Application, which doesn't exists.

Changed

• #30: Support multiple versions of a CRD and allow the specification of namespaces in the Applications resources list.
• #35: Add new field displayName for plugins, to overwrite the name which is displayed for a plugin tab in the frontend.

v0.1.0 (2021-04-04)

Added

• #4: Add Custom Resource Definition for Applications.
• #6: Add Prometheus as datasource for Application metrics.
• #8: Add new page to directly query a configured Prometheus datasource.
• #10: Add Elasticsearch as datasource for Application logs.
• #12: ⚠️ Breaking change: ⚠️ Add plugin system and readd Prometheus and Elasticsearch as plugins.
• #13: Add Jaeger plugin to show traces for an Application and to compare traces.
• #16: Add support for multiple queries in the Prometheus plugin page.
• #18: Add metrics and logs for the gRPC server.
• #19: Use multiple colors in the Jaeger plugin. Each service in a trace has a unique color now, which is used for the charts.
• #21: Add preview for Applications via plugins.
• #22: Add Helm chart.
• #23: Add Kustomize files and demo.
• #24: Add documentation.
• #26: Add support for all Kubernetes resources.

Fixed

• #1: Fix mobile layout for the cluster and namespace filter by using a Toolbar instead of FlexItems.
• #9: Fix time parsing for the datasource options.
• #14: Fix loading of Jaeger services, when a user opend the Jaeger plugin, where the service query parameter was already present.
• #15: Fix resources tab of an Application, where resources were loaded multiple times.

Changed

• #7: Share datasource options between components and allow sharing of URLs.
• #11: ⚠️ Breaking change: ⚠️ Refactor cluster and application handling.
• #17: Use location to load applications, which allows user to share their applications view.
• #20: Rework usage of icons, links handling and drawer layout.
• #25: Change the URL for the Helm repository to helm.kobs.io.