From 42b813f83116e10e320a579682e900eeb1ce0cb3 Mon Sep 17 00:00:00 2001 From: Tomas Panik Date: Fri, 28 Nov 2025 12:03:56 +0100 Subject: [PATCH] Changelog is now obsolete so removing it --- deploy/helm/CHANGELOG.md | 1184 -------------------------------------- 1 file changed, 1184 deletions(-) delete mode 100644 deploy/helm/CHANGELOG.md diff --git a/deploy/helm/CHANGELOG.md b/deploy/helm/CHANGELOG.md deleted file mode 100644 index 79c72cf0b..000000000 --- a/deploy/helm/CHANGELOG.md +++ /dev/null @@ -1,1184 +0,0 @@ -# Changelog - -All notable changes to Helm charts will be documented in this file. - -The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/), -and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html). - -## Unreleased - -## [4.3.0-alpha.1] - 2024-11-11 - -### Added -- Adding OTEL Operator as subchart - -### Added - -- It is now possible to customize prefix of discovered metrics using `otel.metrics.autodiscovery.prefix` (default `k8s.`) - -## [4.2.0-alpha.5] - 2024-10-23 - -### Added - -- Added metrics for Etcd - - -## [4.1.0] - 2024-09-30 - -### Added - -- Improved compatibility with 3rd party tools for managing Helm charts. - - Added Helm chart README.md. - - Improved Helm chart documentation. -- Setting `cluster.uid` is now optional. If not provided it defaults to value of `cluster.name`. -- Initial support for OpenShift clusters. - - Set `openshift.enabled` to `true` during deployment to automatically generate necessary k8s resources. - - Settings `ebpfNetworkMonitoring.enabled` and `swoagent.enabled` must be set to `false` as they are not supported yet. - - This is an experimental feature and it may change in the future. - -### Changed - -- Settings `otel.metrics.k8s_instrumentation`, `otel.logs.k8s_instrumentation` and `otel.events.k8s_instrumentation` are deprecated and may be removed in a future version. -- Enabled automatic deployment of the OTEL Network Collector to the cluster. -- Improved build pipeline. -- Upgraded collector image to `0.11.5` which brings following changes: - - See Release notes for [0.11.5](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.11.5). - - Bumped 3rd party dependencies and Docker images. - -### Fixed - -- Configuring `otel.events.k8s_instrumentation.labels.excludePattern` or `otel.events.k8s_instrumentation.annotations.excludePattern` might cause the events collector to fail when manifest collection is enabled. -- Fixed the autoupdate job to run only on Linux Nodes. - -## [4.1.0-alpha.6] - 2024-09-23 - -### Added - -- Initial support for OpenShift clusters. - - Set `openshift.enabled` to `true` during deployment to automatically generate necessary k8s resources. - - Settings `ebpfNetworkMonitoring.enabled` and `swoagent.enabled` must be set to `false` as they are not supported yet. - -### Changed - -- Improved build pipeline. - -## [4.1.0-alpha.5] - 2024-09-20 - -### Changed - -- Improved Helm chart documentation. -- Additional improvements for supporting the optional `cluster.uid`. -- Improved build pipeline. - -## [4.1.0-alpha.4] - 2024-09-16 - -### Changed - -- Setting `cluster.uid` is now optional. If not provided it defaults to value of `cluster.name`. - -### Fixed - -- Fixed autoupdate job to run only on Linux Nodes. - -## [4.1.0-alpha.3] - 2024-09-13 - -### Fixed - -- Configuring `otel.events.k8s_instrumentation.labels.excludePattern` or `otel.events.k8s_instrumentation.annotations.excludePattern` might cause the events collector to fail when manifest collection is enabled. - -## [4.1.0-alpha.2] - 2024-09-09 - -### Changed -- Upgraded collector image to `0.11.5` which brings following changes: - - See Release notes for [0.11.5](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.11.5). - - Bumped 3rd party dependencies and Docker images. - -## [4.1.0-alpha.1] - 2024-09-03 - -### Changed - -- Enabled the automatic deployment of the OTEL Network Collector to the cluster. - -## [4.0.0] - 2024-08-28 - -### Highlights - -- Added collection of resource manifests. -- Improved filtering abilities. -- All container logs are collected by default. -- **Breaking**: Added validation of the OTEL endpoint provided in `values.yaml`. Deprecated endpoints are rejected now. -- **Breaking**: Removed deprecated `prometheus`, `prometheus-node-exporter` and `opencost` sub-charts. - -### Added - -- Observing changes to supported resources and sending their full manifests to SolarWinds Observability: - - The equivalent of `kubectl get -o=yaml --show-managed-fields`. - - The current list of watched resources: `pods`, `deployments`, `statefulsets`, `replicasets`, `daemonsets`, `jobs`, `cronjobs`, `nodes`, `services`, `persistentvolumes`, `persistentvolumeclaims`, `ingresses` and istio `virtualservices`. - - Enabled by default, can be enabled by setting `otel.manifests.enabled` to `true`. - - For now it runs in the event collector, so `otel.events.enabled` has to be set to `true` (default). - -### Changed - -- Changed the `otel.metrics.filter`, `otel.logs.filter` and `otel.events.filter` settings to be able to use OTTL syntax and access resource attributes: - - If a customer is using the [old filtering syntax](https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/v0.96.0/processor/filterprocessor#alternative-config-options), they behave like in 3.x.x and previous versions of the `k8s collector`. If a customer switches to using the [new syntax](https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/main/processor/filterprocessor#configuration), some of the metric/log-level attributes, like `k8s.deployment.name`, become resource attributes. -- Removed custom `swmetricstransform` processor (replaced with `filter`). -- Switched to using `container` parser in the `filelog` receiver for parsing container logs and their metadata. -- All container logs are collected by default. -- Breaking: Removed `prometheus`, `prometheus-node-exporter` and `opencost` sub-charts. -- Breaking: Added validation of the OTEL endpoint provided in `values.yaml`. In case a deprecated endpoint is detected, an error is reported during chart installation/upgrade. Only endpoints in format `otel.collector.xx-yy.cloud.solarwinds.com` are accepted. See [Data centers and endpoint URIs](https://documentation.solarwinds.com/en/success_center/observability/content/system_requirements/endpoints.htm) for the list of supported endpoints. -- Upgraded collector image to `0.11.4` which brings following changes: - - See Release notes for [0.11.4](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.11.4). - - Bumped 3rd party dependencies and Docker images. - - Upgraded OTEL Collector to v0.107.0. - - Removed deprecated `ballastextension`. The collector now uses environment variable `GOMEMLIMIT`. -- Upgraded SWO Agent image to `v2.9.3`. -- Upgraded `grpcurl` image to `v1.9.1`. -- Upgraded the `kube-state-metrics` subchart from `5.15.2` to [5.25.1](https://github.com/prometheus-community/helm-charts/releases/tag/kube-state-metrics-5.25.1) which brought the following changes: - - Upgraded the `kube-state-metrics` image to [2.13.0](https://github.com/kubernetes/kube-state-metrics/releases/tag/v2.13.0). - - Added `restart_policy` to the `kube_pod_init_container_info` metric. - - Added new metric `kube_pod_container_status_last_terminated_timestamp`. - - For the list of all changes see the respective Git repositories. - -### Fixed - -- Adjusted setting `otel.https_proxy_url` to be applied also to SWO Agent and to disable SolarWinds Observability OTEL endpoint check init container. - -## [4.0.0-alpha.9] - 2024-08-23 - -### Added - -- Collect Istio `VirtualService` resource - -### Changed - -- Upgraded grpcurl image to `v1.9.1` -- Breaking: Removed `prometheus`, `prometheus-node-exporter` and `opencost` sub-charts. - -### Fixed - -- Added workaround for a bug introduced in OTEL Collector to v0.106.0. -- Adjusted setting `otel.https_proxy_url` to be applied also to SWO Agent and to disable SolarWinds Observability OTEL endpoint check init container. - -## [4.0.0-alpha.8] - 2024-08-15 - -### Changed - -- Upgraded the `kube-state-metrics` subchart from `5.15.2` to [5.25.1](https://github.com/prometheus-community/helm-charts/releases/tag/kube-state-metrics-5.25.1) which brought the following changes: - - Upgraded the `kube-state-metrics` image to [2.13.0](https://github.com/kubernetes/kube-state-metrics/releases/tag/v2.13.0). - - Added `restart_policy` to the `kube_pod_init_container_info` metric. - - Added new metric `kube_pod_container_status_last_terminated_timestamp`. - - For the list of all changes see the respective Git repositories. -- Upgraded collector image to `0.11.4` which brings following changes: - - See Release notes for [0.11.4](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.11.4). - - Upgraded OTEL Collector to v0.107.0. - - Removed deprecated `ballastextension`. The collector now uses environment variable `GOMEMLIMIT`. - - Bumped Go builder image to 1.22.6. -- Upgraded SWO Agent image to `v2.9.3` - -### Fixed - -- Fixed a typo in filter syntax warning - -## [3.4.2] - 2024-07-31 - -### Changed - -- Upgraded collector image to `0.10.3` which brings following changes: - - Bumped 3rd party dependencies and Docker images. - -## [4.0.0-alpha.7] - 2024-07-23 - -### Changed - -- Breaking: Add validation of the OTEL endpoint provided in `values.yaml`. In case a deprecated endpoint is detected, report an error during chart installation/update. -- Upgraded collector image to `0.11.3` which brings following changes: - - See Release notes for [0.11.3](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.11.3). - - Upgraded OTEL Collector to v0.105.0. -- Upgraded SWO Agent image to `v2.8.90` - -## [4.0.0-alpha.6] - 2024-07-22 - -### Changed - -- Upgraded collector image to `0.11.2` which brings following changes: - - See Release notes for [0.11.2](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.11.2). - - Bumped 3rd party dependencies and Docker images. - - Internal `k8sattributes` has been renamed to `swk8sattributes` - -## [3.4.1] - 2024-07-19 - -### Changed - -- Upgraded collector image to `0.10.2` which brings following changes: - - Bumped 3rd party dependencies and Docker images. -- Upgraded SWO Agent image to `v2.8.90` - -### Fixed - -- Additional improvements to cluster version detection - -## [4.0.0-alpha.5] - 2024-07-16 - -### Added - -- Added k8s.object.kind to manifest events. - -### Changed - -- Internal configuration of filterprocessor is using OTTL syntax now. - - -## [4.0.0-alpha.4] - 2024-07-11 - -### Added - -- Collect manifests also for Ingresses. - -### Changed - -- By default we are now gathering all logs. -- Fix collector version on metric `k8s.cluster.version`. - - -## [4.0.0-alpha.3] - 2024-06-25 - -### Added - -- Watching all supported resources and sending their full manifests to SWO. - - Disabled by default, can be enabled by setting `otel.manifests.enabled` to true. - - Runs in event collector, so require `otel.events.enabled` set to true. - -## [3.4.0] - 2024-06-28 - -### Changed - -- Upgraded collector image to `0.10.1` which brings following changes: - - See Release notes for [0.10.1](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.10.1). - - Bumped 3rd party dependencies and Docker images. - - Upgraded OTEL Collector to v0.103.0. -- Upgraded SWO Agent image to `v2.8.85` - -### Fixed - -- Only metric `k8s.cluster.version` metric is supposed to have attribute `sw.k8s.cluster.version` - -## [3.4.0-alpha.3] - 2024-06-27 - -### Changed - -- Reverted changes from 3.4.0-alpha.2. They are available in the 4.x branch. -- Upgraded collector image to `0.10.1` which brings following changes: - - See Release notes for [0.10.1](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.10.1). - - Bumped 3rd party dependencies and Docker images. - - Upgraded OTEL Collector to v0.103.0. -- Upgraded SWO Agent image to `v2.8.85` - -## [4.0.0-alpha.2] - 2024-05-30 - -## Fixed - -- Filtering journal logs stopped working in 3.4.0-alpha.2 - -## [4.0.0-alpha.1] - 2024-05-30 - -### Changed - -- As a followup to 3.4.0-alpha.2, the `otel.metrics.filter`, `otel.logs.filter` and `otel.events.filter` are now again backwards compatible. If a customer is using the [old filtering syntax](https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/v0.96.0/processor/filterprocessor#alternative-config-options), they behave like in 3.3.0 and previous versions. If a customer switches to using the [new syntax](https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/main/processor/filterprocessor#configuration), some of the attributes, like `k8s.deployment.name`, become resource attributes. -- Removed use of `swmetricstransform` (replaced with `filterprocessor`) - -## [3.4.0-alpha.2] - 2024-05-16 - -### Changed - -- Changed the `otel.metrics.filter`, `otel.logs.filter` and `otel.events.filter` settings to be able to access resource attributes like `k8s.deployment.name`, ... - - This is a breaking change if anyone was using them before to include only metrics/logs/events with a specific non-resource attribute. - -## [3.4.0-alpha.1] - 2024-05-09 - -### Fixed - -- Only `k8s.cluster.version` metric has attribute `sw.k8s.cluster.version` - -## [3.3.0] - 2024-04-26 - -### Added - -- Added instrumentation of workload attributes to collected logs (`k8s.deployment.name` etc.). Instrumentation of labels and annotations is disabled by default. -- Added option to configure `nodeSelector` and `affinity` for the SWO Agent. -- Added option to configure timeout for each attempt to send data to SWO OTEL endpoint. - - The default is `15s` (previously it was `5s`) to avoid unnecessary retries when the endpoint takes its time to respond. - -### Changed - -- Added environment variables `CLUSTER_UID`, `CLUSTER_NAME` and `MANIFEST_VERSION` to the SWO Agent StatefulSet. Future SWO Agent plugins may include them in their metrics. -- Upgraded collector image to `0.10.0` which brings following changes: - - See Release notes for [0.10.0](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.10.0). - - Bumped 3rd party dependencies and Docker images. - - Upgraded OTEL Collector to v0.98.0. - - ⚠️ Dropped support for several Windows versions that are out of support. The minimum requirement is now Windows 10 or Windows Server 2016. This is caused by the update of Go (and the OTEL Collector). -- Added validation schema for the provided Helm chart configuration. - - ⚠️ If an incorrect configuration is provided, the installation/update of the Helm release will end in error. The previous versions simply ignored the incorrect parts. -- Container logs from AWS EKS Fargate clusters are now sent to SWO as-is. `fluentbit.io/parser` and `fluentbit.io/exclude` annotations are ignored. This both fixes an issue with "empty" JSON logs sent to SWO and aligns the behavior with non-Fargate container logs. - This change is applied only to Pods that are started after the new `k8s collector` is deployed to the k8s cluster. -- Added validation of the OTEL endpoint provided in `values.yaml`. In case a deprecated endpoint is detected, report an warning during chart installation/update. - -### Fixed - -- Fixed Journal log collection on EKS (and other environments where journal logs are stored in `/var/log/journal`) - -## [3.3.0-alpha.6] - 2024-04-23 - -### Added - -- Added instrumentation of workload attributes on logs (`k8s.deployment.name` etc.). Instrumentation of labels and annotations is disabled by default. - -## [3.3.0-alpha.5] - 2024-04-19 - -### Added - -- Added option to configure `nodeSelector` and `affinity` for SWO Agent. - -### Changed - -- Added environment variables `CLUSTER_UID`, `CLUSTER_NAME` and `MANIFEST_VERSION` to the SWO Agent StatefulSet. Future SWO Agent plugins may include them in their metrics. -- Upgraded collector image to `0.10.0` which brings following changes: - - See Release notes for [0.10.0](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.10.0). - - Bumped 3rd party dependencies and Docker images. - - Upgraded OTEL Collector to v0.98.0. - - ⚠️ Dropped support for several Windows versions that are out of support. The minimum requirement is now Windows 10 or Windows Server 2016. This is caused by the update of Go (and the OTEL Collector). - -## [3.3.0-alpha.4] - 2024-04-10 - -### Changed - -- Added validation schema for the provided Helm chart configuration. - -## [3.3.0-alpha.3] - 2024-03-27 - -### Changed - -- Container logs from AWS EKS Fargate clusters are now sent to SWO as-is. `fluentbit.io/parser` and `fluentbit.io/exclude` annotations are ignored. This both fixes an issue with "empty" JSON logs sent to SWO and aligns the behavior with non-Fargate container logs. - This change is applied only to Pods that are started after the new `k8s collector` is deployed to the k8s cluster. - -## [3.3.0-alpha.2] - 2024-03-21 - -### Fixed - -- Fixed Journal log collection on EKS (and other environments where journal logs are stored in `/var/log/journal`) - -## [3.3.0-alpha.1] - 2024-03-13 - -### Added - -- Added option to configure `timeout`, which is time to wait per individual attempt to send data to SWO. - - default is `15s` (previously was `5s`) avoiding unnecessary retries when backend takes time to respond - -### Changed - -- Add validation of the OTEL endpoint provided in `values.yaml`. In case a deprecated endpoint is detected, report an warning during chart installation/update. - -## [3.2.0] - 2024-02-02 - -### Added - -- Added Linux ARM64 support -- Added Windows Server 2019 support -- Added option to deploy OpenCost collector - - Can be enabled by setting `openconst.enabled` to `true` in `values.yaml` -- Added option to pull images from ACR by setting `global.azure.images.` in `values.yaml` when running in AKS -- Added `otel.api_token` to allow setting API token for OTEL collector through `values.yaml` -- Added `otel.metrics.autodiscovery.prometheusEndpoints.podMonitors` configuration option. Define if you want to monitor applications that do not have Prometheus annotations. -- Started publishing custom Istio metrics, when available: `k8s.istio_request_bytes.delta`, `k8s.istio_response_bytes.delta`, `k8s.istio_requests.rate`, `k8s.istio_tcp_sent_bytes.rate`, `k8s.istio_tcp_received_bytes.rate`, `k8s.istio_requests.delta`, `k8s.istio_tcp_sent_bytes.delta`, `k8s.istio_tcp_received_bytes.delta`, `k8s.istio_request_bytes.rate`, `k8s.istio_response_bytes.rate` and `k8s.istio_request_duration_milliseconds.rate` - -### Changed - -- Upgraded OTEL collector image to `0.9.2` which brings following changes - - see Release notes for - [0.8.11](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.11), - [0.8.12](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.12), - [0.8.13](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.13), - [0.9.0](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.9.0), - [0.9.1](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.9.1), - [0.9.2](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.9.2) - - Switched from `scratch` base image to `distroless` - - Bumped 3rd party dependencies and Docker images - - OTEL upgraded to v0.91.0 -- Adjusted node-collector memory limits for better performance -- Removed sha256 from image references to enable multiarch support -- Removed `k8s.kube_daemonset_labels`, `k8s.kube_deployment_labels` and `k8s.kube_statefulset_labels` metrics (they were redundant anyway) - - There are no longer published by the updated `kube-state-metrics` -- Stopped sending histogram metrics to SWO -- Added PVC for SWO Agent, when enabled - -### Fixed - -- Fixed autoupdate job -- Fixed usage of `ebpfNetworkMonitoring.k8sCollector.relay.image.pullPolicy` -- Fixed restarts of the collector when automatic discovery and scraping of prometheus endpoints (`otel.metrics.autodiscovery.prometheusEndpoints.enabled`) was disabled or enabled on a Fargate environment - -## [3.2.0-alpha.17] - 2024-01-23 - -### Fixed - -- Fixed how `k8s.istio_request_duration_milliseconds.rate` metric is calculated - -## [3.2.0-alpha.16] - 2024-01-22 - -### Fixed - -- Fixed restarts of the collector when automatic discovery and scraping of prometheus endpoints (`otel.metrics.autodiscovery.prometheusEndpoints.enabled`) was enabled on a Fargate environment - -## [3.2.0-alpha.15] - 2024-01-17 - -### Fixed - -- Fixed restarts of the collector when automatic discovery and scraping of prometheus endpoints (`otel.metrics.autodiscovery.prometheusEndpoints.enabled`) was disabled - -## [3.2.0-alpha.14] - 2024-01-12 - -- Publish custom Istio metrics, when available: `k8s.istio_request_bytes.delta`, `k8s.istio_response_bytes.delta`, `k8s.istio_requests.rate`, `k8s.istio_tcp_sent_bytes.rate`, `k8s.istio_tcp_received_bytes.rate`, `k8s.istio_requests.delta`, `k8s.istio_tcp_sent_bytes.delta`, `k8s.istio_tcp_received_bytes.delta` -- Disable opencost metrics by default - -## [3.2.0-alpha.13] - 2024-01-08 - -- Make sure discoverd histogram metrics are not sent to SWO -- Publish custom Istio metrics, when available: `k8s.istio_request_bytes.rate`, `k8s.istio_response_bytes.rate` and `k8s.istio_request_duration_milliseconds.rate`. - -## [3.2.0-alpha.12] - 2024-01-04 - -- Upgraded OTEL collector image to `0.9.2` (see [Release notes](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.9.2)) which brings following changes - - Bumped 3rd party dependencies and Docker images - - OTEL upgraded to v0.91.0 -- Upgraded `otel/opentelemetry-ebpf-*` images to `v0.10.1` - -## [3.2.0-alpha.11] - 2023-12-21 - -- Revert deployments strategy to RollingUpdate - -## [3.2.0-alpha.10] - 2023-12-18 - -- Switched deployments into Recreate strategy - -## [3.2.0-alpha.9] - 2023-12-15 - -### Changed - -- Updating kube-state-metrics to 5.15.2 - - removing `k8s.kube_daemonset_labels`, `k8s.kube_deployment_labels` and `k8s.kube_statefulset_labels` metrics (they were redundant anyway) -- Upgraded OTEL collector image to `0.9.1` (see [Release notes](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.9.1)) - -## [3.2.0-alpha.8] - 2023-12-13 - -### Added - -- Added `otel.metrics.autodiscovery.prometheusEndpoints.podMonitors` configuration option. Define if you want to monitor applications that do not have prometheus annotations. -- Upgraded OTEL collector image to `0.9.0` (see [Release notes](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.9.0)) - -### Fixed - -- Fixed usage of `ebpfNetworkMonitoring.k8sCollector.relay.image.pullPolicy` - -## [3.2.0-alpha.7] - 2023-12-07 - -### Fixed - -- Fixed rendering on FluxCD, removing helm version constraint - -## [3.2.0-alpha.6] - 2023-12-05 - -### Added - -- Added Windows Server 2019 support - -## [3.2.0-alpha.5] - 2023-12-01 - -### Added - -- Images can be overriden by setting `global.azure.images.` in `values.yaml` -- Added `otel.api_token` to allow setting API token for OTEL collector through `values.yaml` - -## [3.2.0-alpha.4] - 2023-11-30 - -### Fixed -- Removing sha256 from image pulls as it does not allow multi arch images - -## [3.1.1] - 2023-11-30 - -### Fixed -- Fixed autoupdate job to use right image - -## [3.2.0-alpha.3] - 2023-11-29 - -- Added option to enable opencost metrics - -## [3.2.0-alpha.2] - 2023-11-27 - -- Fine tuning default node-collector values for better performance - -## [3.2.0-alpha.1] - 2023-11-27 - -### Added - -- Added ARM64 support - - Upgraded OTEL collector image to `0.8.12` (see [Release notes](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.12)) - - -## [3.1.0] - 2023-11-27 - -### Added - -- Added network monitoring for Linux nodes which is based on eBPF, so it does not require service mesh to be deployed. - - It is disabled by default, to enable it set `ebpfNetworkMonitoring.enabled: true` in `values.yaml` - - It is based on [opentelemetry-ebpf](https://github.com/open-telemetry/opentelemetry-ebpf) - - For scaling see `numIngestShards`, `numMatchingShards` and `numAggregationShards` in `values.yaml` - - See [exported_metrics.md](../../doc/exported_metrics.md) for list of metrics -- Automatic discovery and scraping of prometheus endpoints on pods. Driven by `otel.metrics.autodiscovery.prometheusEndpoints.enabled` option in `values.yaml`, by default enabled (Fargate not yet supported). - - In case `otel.metrics.autodiscovery.prometheusEndpoints.enabled` is set to `true` (which is by default) `extra_scrape_metrics` is ignored as there is high chance that those metrics will be collected two times. You can override this behavior by by setting `force_extra_scrape_metrics` to true. -- Added option to set `imagePullSecrets` in `values.yaml` -- Added option to configure `terminationGracePeriodSeconds` defaulting to 10 minutes, so that it is guaranteed that collector process whole pipeline -- Added option to configure `sending_queue` - - Added option to offload in metrics collector `sending_queue` to storage, reducing memory requirement for the collector -- Added option to configure `retry_on_failure` - - default for initial_interval is now `10s` (previously was `5s`) avoiding unnecessary retries when backend takes time to respond - -### Changed - -- Prometheus is no longer needed for kubernetes monitoring, therefore it is no longer deployed. By default, k8s collector does not scrape anything from Prometheus. - - If there was a Prometheus deployed by the previous version (using setting `prometheus.enabled``), it will be automatically removed. - - Setting `otel.metrics.prometheus.url` is still available, but is valid only in combination with `otel.metrics.extra_scrape_metrics`. - - If `otel.metrics.prometheus.url` is empty, `otel.metrics.prometheus_check` is ignored to not fail on missing Prometheus. - - Note that `otel.metrics.extra_scrape_metrics` is deprecated option and will be removed in future versions. - - Node metrics are now scaped from node-collector daemonset -- Decreased the default batch size for metrics, logs and events sent to OTEL endpoint to 512 to avoid too big messages -- Upgraded SWO Agent image to `v2.6.28` - -## [3.1.0-rc.6] - 2023-11-24 - -### Added -- Added option to configure liveness and readiness probes initial startup delay on metrics collector, defaulting to 10s - -## [3.1.0-rc.5] - 2023-11-24 - -### Fixed -- Fixed counterToRate conversion (when defined custom it broke cAdvisor processing) - -## [3.1.0-rc.4] - 2023-11-24 -- Fixed metrics in node-collector when logs are disabled -- Fine tuning default values of sending_queue for better performance - -## [3.1.0-rc.3] - 2023-11-23 - -### Fixed -- Fixed `k8s.node.name` from node collector - -## [3.1.0-rc.2] - 2023-11-23 - -### Added -- Added option to configure scaling of eBPF monitoring reducer -- Added option to configure eBPF reducer's `enableIdIdGeneration` which is unnecessary at the moment (default false). - -## [3.1.0-rc.1] - 2023-11-23 - -### Changed -- `sending_queue`.`queue_size` changed from `1000` to `200`, decreasing amount of memory it can take -- Upgraded OTEL collector image to `0.8.10` (see [Release notes](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.10)) -- Upgraded SWO Agent image to `v2.6.28` - -### Added -- Added option to configure terminationGracePeriodSeconds defaulting to 10 minutes, so that it is guaranteed that collector process whole pipeline -- Added option to offload sending_queue to storage, reducing memory requirement for the collector -- Added option to configure sending_queue -- Added option to configure retry_on_failure - - default for initial_interval is now `10s` (previously was `5s`) avoiding unnecessary retries when backend takes time to respond - -### Fixed -- ebpf monitoring: Added necessary init containers making sure that all components start in the right order - -## [3.1.0-alpha.2] - 2023-11-16 - -### Added - -- Added option to set `imagePullSecrets` - -## [3.1.0-alpha.1] - 2023-11-10 - -### Added - -- Automatic discovery and scraping of prometheus endpoints on pods. Driven by `otel.metrics.autodiscovery.prometheusEndpoints.enabled` option in `values.yaml`, by default enabled (Fargate not yet supported). - - In case `otel.metrics.autodiscovery.prometheusEndpoints.enabled` is set to `true` (which is by default) `extra_scrape_metrics` is ignored as there is high chance that those metrics will be collected two times. You can override this behavior by by setting `force_extra_scrape_metrics` to true. -- Upgraded OTEL collector image to `0.8.9` (see [Release notes](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.9)) - -## [3.0.0-alpha.6] - 2023-11-07 - -- correctly filtering out `ebpf_net` metrics by default (which is internal eBPF telemetry) - -## [3.0.0-alpha.5] - 2023-11-06 - -### Fixed - -- Fixes metrics collector, when `otel.metrics.extra_scrape_metrics` and `prometheus.enabled` are set -- node-collector no longer mounts `hostPort`. This was solved by separating kernel-collector into separate daemonset - -## [3.0.0-alpha.4] - 2023-11-03 - -### Changed - -- Decreased the default batch size for metrics, logs and events sent to OTEL endpoint to 512 to avoid too big messages - -## [3.0.0-alpha.3] - 2023-11-03 - -### Fixed - -- If `otel.metrics.prometheus.url` is empty, it ignores `otel.metrics.prometheus_check` to not fail on missing Prometheus - -## [3.0.0-alpha.2] - 2023-11-03 - -### Fixed - -- Fixing installation/upgrade on Helm 3.9.0 and bellow - -## [3.0.0-alpha.1] - 2023-11-02 - -### Changed - -- Upgraded OTEL collector image to `0.8.8` (see [Release notes](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.8)) -- Prometheus is no longer needed for kubernetes monitoring, therefore is no longer deployed. By default metrics collector does not scrape anything from Prometheus. - - `otel.metrics.prometheus.url` option is still available, but is used only if `otel.metrics.extra_scrape_metrics` is used. - - Note that `otel.metrics.extra_scrape_metrics` is deprecated option and will be removed in future versions. - -### Added - -- Added network monitoring for Linux nodes which is based on eBPF, so does not require service mesh to be deployed. - - it is disabled by default, to enable it set `ebpfNetworkMonitoring.enabled: true` in `values.yaml` - - it is based on [opentelemetry-ebpf](https://github.com/open-telemetry/opentelemetry-ebpf) - - see [exported_metrics.md](../../doc/exported_metrics.md) for list of metrics - -## [2.8.0] - 2023-10-30 - -### Added - -- Added monitoring of logs for Windows nodes -- Added services metrics (Scrape `kube_service_*` and `kube_endpoint_*` metrics) - -### Changed - -- Changing log message attributes to respect OTEL log format -- Upgraded OTEL collector image to `0.8.6` (see [Release notes](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.6)) -- Updated labels so that resources can be identified more easily - - `app.kubernetes.io/name` changed to container application name (e.q. `swo-k8s-collector` for SWO k8s collector, `swo-agent` for SWO agent) - - `app.kubernetes.io/part-of` always set to `swo-k8s-collector` -- Update logs daemon requests/limits -- Removed attributes `net.host.name`, `net.host.port`, `http.scheme`, `prometheus`, `prometheus_replica` and `endpoint` from exported metrics - -### Fixed - -- Fixed nodeselector for `kube-state-metrics` so that it is deployed on linux nodes only -- Detection of Node name for Fargate Nodes's metrics -- Adding autofix from corrupted storage - -## [2.8.0-alpha.8] - 2023-10-25 - -* Removed memory request and ballast for logs daemonset - -## [2.8.0-alpha.7] - 2023-10-24 - -* Updated image to `0.8.6` (see [Release notes](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.6)) -* Scraping labels/annotations from Services -* Removed useless metrics `kube_service_annotations`, `kube_service_labels`, `kube_endpoint_annotations`, `kube_endpoint_labels` - -## [2.8.0-alpha.6] - 2023-10-11 - -* Updated labels so that resources can be identified more easily - * `app.kubernetes.io/name` changed to container application name (e.q. `swo-k8s-collector` for SWO k8s collector, `swo-agent` for SWO agent) - * `app.kubernetes.io/part-of` always set to `swo-k8s-collector` - -## [2.8.0-alpha.5] - 2023-10-11 - -### Fixed - -- Adding autofix from corrupted storage - -## [2.8.0-alpha.4] - 2023-10-09 - -### Fixed - -- Detection of Node name for Fargate Nodes's metrics - -### Added - -- Scrape kube_service_* and kube_endpoint_* metrics - -### Removed - -- Removed attributes `net.host.name`, `net.host.port`, `http.scheme`, `prometheus`, `prometheus_replica` and `endpoint` from exported metrics - -### Changed - -- Updated docker image to 0.8.5 (see [Release notes](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.5)) - -## [2.8.0-alpha.3] - 2023-10-06 - -### Changed - -- Changing log message attributes to respect OTEL log format - -### Fixed - -- Fixing nodeselector for kube-state-metrics so that it is deployed on linux nodes only - -## [2.8.0-alpha.2] - 2023-10-04 - -### Added - -- Add monitoring windows node logs - -## [2.8.0-alpha.1] - 2023-09-11 - -### Added - -- Added windows container for logs monitoring - -## [2.7.0] - 2023-09-04 - -### Added - -- Added new Helm settings `aws_fargate.enabled` and `aws_fargate.logs.*` that allow the k8s collector Helm chart to setup AWS EKS Fargate logging ConfigMap - - Setting `prometheus.forceNamespace` can be used to force deployment of the bundled Prometheus to a specific non-Fargate namespace - -### Changed - -- Upgraded OTEL collector image to [0.8.2](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.2) which brings following changes - - Bump library/golang from 1.20.7-bullseye to 1.21.0-bullseye and update some 3rd party dependencies - - OTEL upgraded to v0.81.0 - - Updating `k8sattributes` to instrument attribute indicating that object exists -- Metrics no longer send `k8s.node.name` resource attribute if node does not exists in Kubernetes (for example in case of Fargate nodes) -- Adjusted Events collection to not produce resource attributes for entities that do not exists in Kubernetes -- DaemonSet for Log collection now restricts where it runs: - - Fargate nodes are excluded - - Only linux nodes with amd64 architecture are included - -### Fixed -- Fixed Journal log collection on EKS (and other environment where journal logs are stored in `/var/log/journal`) - -## [2.7.0-alpha.8] - 2023-08-31 - -### Changed -- Upgraded OTEL collector image to [0.8.2](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.2) which brings following changes - - Bump library/golang from 1.20.7-bullseye to 1.21.0-bullseye and update some 3rd party dependencies - -## [2.7.0-alpha.7] - 2023-08-30 - -### Fixed -- Usage metrics for nodes - -## [2.7.0-alpha.6] - 2023-08-28 - -### Changed -- Metrics will no longer send `k8s.node.name` resource attribute if node does not exists in Kubernetes (for example in case of Fargate nodes) - -## [2.7.0-alpha.5] - 2023-08-22 - -### Changed -- Adjusted bundled prometheus to not run on Fargate nodes by default -- Allowed use of `prometheus.forceNamespace` option of bundled prometheus, to force namespace where prometheus is deployed - -## [2.7.0-alpha.4] - 2023-08-17 - -### Changed -- Adjusted Log group name used for Fargate logs -- Adjusted Events collection to not produce resource attributes for entities that do not exists in Kubernetes -- Upgraded OTEL collector image to [0.8.1](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.1) which brings following changes - - Bump library/golang from 1.20.6-bullseye to 1.20.7-bullseye and update some 3rd party dependencies - - Updating `k8sattributes` to instrument attribute indicating that object exists - -## [2.7.0-alpha.3] - 2023-08-17 - -### Added -- There are new Helm settings `aws_fargate.enabled` and `aws_fargate.logs.enabled` that allow the k8s collector Helm chart to setup AWS EKS Fargate logging ConfigMap - -### Changed -- Log collection DaemonSet now restrict where it runs: - - Fargate nodes are excluded - - Only linux nodes with amd64 architecture are included - -### Fixed -- Fixed Journal log collection on EKS (and other environment where journal logs are stored in `/var/log/journal`) - -## [2.7.0-alpha.2] - 2023-07-18 - -### Changed -- Upgraded OTEL collector image to [0.8.0](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.8.0) which brings following changes - - OTEL upgraded to v0.81.0 - -## [2.6.0] - 2023-07-17 - -### Changed -- `k8s.pod.spec.cpu.limit` is calculated from kube-state-metrics (Kubernetes describe) and not from container runtime metrics. This should make the metric more reliable. -- We don't modify original prometheus metrics anymore. - -### Fixed -- Fixed filter on kube_namespace_status_phase, only values with 1 are sent - -## [2.6.0-alpha.2] - 2023-07-11 - -### Fixed -- Fixed filter on kube_namespace_status_phase, only values with 1 are sent - -## [2.6.0-alpha.1] - 2023-06-08 - -### Changed -- We don't modify original prometheus metrics anymore. - -## [2.5.0] - 2023-06-12 - -### Added -- Added PV and PVC metrics - - Updating docker image to `0.7.0` (which is capable of instrumenting PV and PVC with labels/annotations) - -### Changed -- Updating docker images `solarwinds/swo-agent`, `busybox`, `fullstorydev/grpcurl` and `alpine/k8s` to latest available versions - -### Fixed -- Fixed CPU/Memory usage Pod metrics on latest cAdvisor/containerd (not relying on Pod level datapoints, but doing SUM of container datapoints) -- Fixed node level network metrics for environments where pod level network metrics are not available (for example Docker runtime) - -## [2.5.0-alpha.6] - 2023-06-08 - -### Changed -- Updating docker images `solarwinds/swo-agent`, `busybox`, `fullstorydev/grpcurl` and `alpine/k8s` to latest available versions - -## [2.5.0-alpha.5] - 2023-06-07 -### Fixed -- Fixed node level network metrics for environments where pod level network metrics are not available (for example Docker runtime) - -## [2.5.0-alpha.4] - 2023-06-05 -### Fixed -- Fixed CPU/Memory usage Pod metrics on latest cAdvisor/containerd (not relying on Pod level datapoints, but doing SUM of container datapoints) - -## [2.5.0-alpha.3] - 2023-06-02 -### Added -- Added `k8s.kube_pod_spec_volumes_persistentvolumeclaims_info` metric to connect Pod and PVC - -## [2.5.0-alpha.2] - 2023-05-31 - -### Changed -- `access_mode` is now published as resource attribute -- `kubelet_*` metrics are published to SWO with prefix `k8s` (to be consistent with other kubernetes related metrics) - -## [2.5.0-alpha.1] - 2023-05-25 -### Added -- Added PV and PVC metrics - -### Changed -- Updating docker image to `0.7.0` (which is capable of instrumenting PV and PVC with labels/annotations) - -## [2.4.1] - 2023-05-25 - -### Changed -- Updating docker image to `0.6.0` (which includes some security fixes and add forwardconnector OTEL component) - -### Fixed -- Fixed filter on kube-state-metrics so that only specific metrics are sent - -## [2.4.0] - 2023-05-16 - -### Added -- Added new container metrics `k8s.container.fs.iops`, `k8s.container.fs.throughput`, `k8s.container.network.bytes_received` and `k8s.container.network.bytes_transmitted` -- Added scraping of `kube_pod_init_container_*` metrics -- Merics `k8s.container.spec.cpu.limit`, `k8s.container.spec.cpu.requests`, `k8s.container.spec.memory.requests`, `k8s.container.spec.memory.limit` and `k8s.container.status` now include datapoints for both init and non-init containers -- `kube-state-metrics` is now bundled with the Helm chart so that its metrics are predictable -- FIPS compliance - -### Changed -- Upgraded OTEL collector image to [0.5.2](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.5.2) which brings following changes - - FIPS support - - Updated build dependencies (security fixes) - -### Removed -- Removed metrics `k8s.cluster.memory.utilization` and `k8s.cluster.cpu.utilization` - they are replaced by composite metrics calculated by the SWO platform - -### Fixed -- Fixed Autoupdate - - Adjusted permissions to be able to update ClusterRoles for future increments - - The update is now atomic, so in case it fails, it will rollback (it will not leave Helm release in Failed state) -- Metric `k8s.kube_pod_status_phase` should not send values with 0 anymore - -## [2.4.0-alpha.6] - 2023-05-04 - -### Fixed -- `k8s.kube_pod_status_phase` should not send values with 0 anymore - -## [2.4.0-alpha.5] - 2023-05-02 - -### Added -- FIPS compliance - -### Fixed -- Updated docker image to [0.5.1](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.5.1) which contains security fixes -- Fixed Autoupdate - - Adjusted permissions to be able to update clusterroles for future increments - - The update is now atomic, so in case it fails it will rollback (it will not leave Helm release in Failed state). - -## [2.4.0-alpha.4] - 2023-04-25 - -### Changed -- Updated metrics `k8s.container.fs.iops`, `k8s.container.fs.throughput` to be correctly bind by swo processing pipeline - - -## [2.4.0-alpha.3] - 2023-04-25 - -### Changed -- Updated metrics `k8s.container.fs.iops`, `k8s.container.fs.throughput` to be correctly bind by swo processing pipeline - - -## [2.4.0-alpha.2] - 2023-04-24 - -### Added -- kube-state-metrics is now bundled with the Helm chart so that its metrics are predictable - -### Changed -- `k8s.cluster.memory.utilization` and `k8s.cluster.cpu.utilization` are no longer calculated. They are replaced by composite metric calculated by the platform -- `k8s.container.spec.cpu.limit`, `k8s.container.spec.cpu.requests`, `k8s.container.spec.memory.requests`, `k8s.container.spec.memory.limit` and `k8s.container.status` now includes datapoints for both init and non-init containers - -## [2.4.0-alpha.1] - 2023-04-19 - -### Added -- Added new container metrics `k8s.container.fs.iops`, `k8s.container.fs.throughput`, `k8s.container.network.bytes_received`, `k8s.container.network.bytes_transmitted` -- Added scraping of `kube_pod_init_container_*` metrics - -## [2.3.0] - 2023-04-13 - -### Added -- Added automatic extraction of Kubernetes labels and annotations from resources (Pods, Namespaces, Deployment, StatefulSet, ReplicaSet, DaemonSet, Job, CronJob, Node) and sent using resource attributes with metrics end events. -- A new option to deploy `prometheus` as part of the k8s collector chart installation, controlled by setting `prometheus.enabled: true` in `values.yaml`. -- New StatefulSet with light weight SWO Agent optionally deployed by default -- Added syslog attributes for log entry so that logs are properly displayed by LogViewer (`syslog.facility`, `syslog.version`, `syslog.procid`, `syslog.msgid`) - - Added resource level attributes: `host.hostname` contains name of the pod (represented as System in LogViewer), `service.name` contains name of the container (represented as Program in LogViewer). -- New metrics are scraped from Prometheus: `k8s.kube_replicaset_spec_replicas`, `k8s.kube_replicaset_status_ready_replicas`, `k8s.kube_replicaset_status_replicas` -- Added metrics `k8s.cluster.version` which extract version from `kubernetes_build_info`. Metric `kubernetes_build_info` is no longer published - -### Fixed -- Enabled `honor_labels` option to keep scraped labels unchanged -- Fixed `k8s.job.condition` resource attribute to handle Failed state -- Fixed calculation of `k8s.pod.spec.memory.limit` on newer container runtime (no longer use `container_spec_memory_limit_bytes`, but `kube_pod_container_resource_limits`) -- Fix grouping conditions for `container_network_*` and `container_fs_*` metrics to not rely on container attribute - -## [2.3.0-alpha.7] - 2023-04-12 -- Added automatic extraction of Kubernetes labels and annotations from events. - -## [2.3.0-alpha.6] - 2023-04-06 - -### Added -- Added automatic extraction of Kubernetes labels and annotations from additional resources (Deployment, StatefulSet, ReplicaSet, DaemonSet, Job, CronJob, Node) and sent using resource attributes with metric - -## [2.3.0-alpha.5] - 2023-04-06 - -### Changed -- Enabled honor_labels option to keep scraped data over server-side labels - -### Fixed -- Fixed calculation of `k8s.pod.spec.memory.limit` on newer container runtime (no longer use `container_spec_memory_limit_bytes`, but `kube_pod_container_resource_limits`) - -## [2.3.0-alpha.4] - 2023-03-29 - -### Added -- New StatefulSet with light weight SWO Agent optionaly deployed by default -- Added syslog attributes for log entry: `syslog.facility`, `syslog.version`, `syslog.procid`, `syslog.msgid`. -- Added resource level attributes: `host.hostname` contains name of the pod, `service.name` contains name of the container. - -## [2.3.0-alpha.3] - 2023-03-24 - -### Changed - -- Fixed k8s.job.condition resource attribute to handle Failed state - -### Added - -- New replicaset metrics `k8s.kube_replicaset_spec_replicas`, `k8s.kube_replicaset_status_ready_replicas`, `k8s.kube_replicaset_status_replicas` - -## [2.3.0-alpha.2] - 2023-03-22 - -### Added - -- A new option to deploy `prometheus` as part of the k8s collector chart installation, controlled by setting `prometheus.enabled: true` in `values.yaml`. -- Added automatic extraction of Kubernetes labels and annotations from resources (Pods, Namespaces) and sent using resource attributes with metric - -## [2.3.0-alpha.1] - 2023-03-21 - -### Changed - -- Fix grouping conditions for `container_network_*` and `container_fs_*` metrics to not rely on container attribute -- Added metrics k8s.cluster.version which extract version from kubernetes_build_info. Metric kubernetes_build_info is not published - -## [2.2.0] - 2023-03-23 - -### Added - -- Collect metrics about kube jobs. -- Attribute k8s.pod.name to events -- Added internal ip attribute for node [#168](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/168). -- Added metric k8s.kubernetes_build_info for calculation of sw.k8s.cluster.version -- InitContainers where Prometheus and OTEL endpoints are checked -- k8s.namespace.name moved from attributes to resource.attributes -- Filtering out [internal metrics generated by Prometheus scraper](https://prometheus.io/docs/concepts/jobs_instances/#automatically-generated-labels-and-time-series) (`scrape_duration_seconds`, `scrape_samples_post_metric_relabeling`, `scrape_samples_scraped`, `scrape_series_added`, `up`). -- introduced k8s.job.condition resource attribute for job which can be Pending, Complete or Failed -- introduced k8s.container.spec.cpu.limit metric for CPU quota -- Added possibility to deploy `PodMonitor` resources so that OTEL collector telemetry is scraped by Prometheus Operator (see [Prometheus Operator design](https://prometheus-operator.dev/docs/operator/design/)) -- Added k8s.container.cpu.usage.seconds.rate metric -- Adding `container.id` and `container.runtime` attributes to `k8s.kube_pod_container_info` metric for unique container identification [#182](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/182). -- Added optional autoupdate support (set by `autoupdate.enabled` in `values.yaml`) [#196](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/196). - -### Changed - -- Fix grouping conditions for container*network*_ and container*fs*_ metrics to not relly on container attribute -- Added metrics k8s.cluster.version which extract version from kubernetes_build_info. Metric kubernetes_build_info is not published -- Filtering out datapoints for internal k8s containers (with name "POD", usually using image "pause") -- Upgraded OTEL collector image to [0.4.0](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.4.0) which brings following changes - - OTEL upgraded to 0.73.0 - - Updated build dependencies - -## [2.2.0-beta.1] - 2023-03-16 - -### Added - -- Adding `container.id` and `container.runtime` attributes to `k8s.kube_pod_container_info` metric for unique container identification [#182](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/182). -- Added optional autoupdate support (set by `autoupdate.enabled` in `values.yaml`) [#196](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/196). - -### Changed - -- Upgraded OTEL collector image to [0.4.0](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.4.0) which brings following changes - - OTEL upgraded to 0.73.0 - - Updated build dependencies - -## [2.2.0-alpha.4] - 2023-03-14 - -### Added - -- Added possibility to deploy `PodMonitor` resources so that OTEL collector telemetry is scraped by Prometheus Operator (see [Prometheus Operator design](https://prometheus-operator.dev/docs/operator/design/)) -- Added k8s.container.cpu.usage.seconds.rate metric - -### Changed - -- k8s.job.condition, state name Pending changed to Active -- Filtering out datapoints for internal k8s containers (with name "POD", usually using image "pause") - -## [2.2.0-alpha.3] - 2023-03-09 - -### Added - -- k8s.namespace.name moved from attributes to resource.attributes -- Filtering out [internal metrics generated by Prometheus scraper](https://prometheus.io/docs/concepts/jobs_instances/#automatically-generated-labels-and-time-series) (`scrape_duration_seconds`, `scrape_samples_post_metric_relabeling`, `scrape_samples_scraped`, `scrape_series_added`, `up`). -- introduced k8s.job.condition resource attribute for job which can be Pending, Complete or Failed -- introduced k8s.container.spec.cpu.limit metric for CPU quota - -## [2.2.0-alpha.2] - 2023-03-02 - -### Added - -- InitContainers where Prometheus and OTEL endpoints are checked -- Added metric k8s.kubernetes_build_info for calculation of sw.k8s.cluster.version - -## [2.2.0-alpha.1] - 2023-02-21 - -### Added - -- Collect metrics about kube jobs. -- Attribute k8s.pod.name to events -- Added internal ip attribute for node [#168](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/168). - -## [2.1.0] - 2023-02-16 - -### Added - -- Added telemetry port to kubernetes "ports" [#129](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/129) -- Added `tolerations` config of Logs collection DaemonSet (with default to run on tainted nodes) [#141](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/141) -- if telemetry is enabled (true by default) OTEL collectors will contain prometheus annotations so that telemetry is discovered by Prometheus [#152](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/152). -- configuration of `file_storage` extension is now available in `values.yaml`. [#157](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/157) - - default `timeout` is now set to `5s` - - Log collector suffix clusterId into folder which it mounts for checkpoints (e.q.`/var/lib/swo/checkpoints/`). This avoid unpredictable errors in scenario when previous monitoring was not deleted. [#161](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/161) -- new events transform pipeline which sets **sw.namespace** attribute to **sw.events.inframon.k8s** [#155](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/155). - [#145](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/145). -- Exposed some configuration of [filelogreciever](https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/main/receiver/filelogreceiver) in `values.yaml` [#146](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/146). - -### Changed - -- Upgraded OTEL collector image to [0.3.0](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.3.0) which brings following changes - - OTEL upgraded to 0.69.0 - - Added `filestorage` so it can be in processors - -### Fixed - -- Properly annotate configmap checksums [#151](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/151) -- Template now use `https_proxy_url` from the right place [#151(https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/151) -- Added optimizations to Log collector preventing Out of Memory situations [#137](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/137) - - Added `filelog.storage` to persist checkpoints in persistent storage, not in memory. - - Setting `max_concurrent_files` to 10 (from default 1024), this is main memory optimization, reducing amount of concurrent log scans - - Increased default memory limit to `700Mi` (which should be enough for large logs) - - Having by default `150Mi` difference between OTEL memory limit and Kubernetes memory limit, so that OTEL has enough buffer (this prevents OOMing) - -## [2.1.0-beta.2] - 2023-02-15 - -### Changed - -- Remove generated attribute k8s.deployment.name -- Log collector suffix clusterId into folder which it mounts for checkpoints (e.q.`/var/lib/swo/checkpoints/`). This avoid unpredictable errors in scenario when previous monitoring was not deleted. [#161](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/161) - -## [2.1.0-beta.1] - 2023-02-09 - -### Added - -- if telemetry is enabled (true by default) OTEL collectors will contain prometheus annotations so that telemetry is discovered by Prometheus [#152](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/152). -- configuration of `file_storage` extension is now available in `values.yaml`. [#157](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/157) - - default `timeout` is now set to `5s` -- new events transform pipeline which sets **sw.namespace** attribute to **sw.events.inframon.k8s** [#155](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/155). - -### Fixed - -- Properly annotate configmap checksums [#151](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/151) -- Template now use `https_proxy_url` from the right place [#151(https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/151) - -## [2.1.0-alpha.3] - 2023-01-31 - -### Added - -- Add attribute k8s.deployment.name to exported logs - [#145](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/145). - -## [2.1.0-alpha.2] - 2023-01-30 - -### Added - -- Exposed some configuration of [filelogreciever](https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/main/receiver/filelogreceiver) in `values.yaml` [#146](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/146). - -### Changed - -- Changed default value `start_at` property of [filelogreciever](https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/main/receiver/filelogreceiver) to `end` [#146](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/146). - -## [2.1.0-alpha.1] - 2023-01-25 - -### Added - -- Added telemetry port to kubernetes "ports" [#129](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/129) -- Added `tolerations` config of Logs collection DaemonSet (with default to run on tainted nodes) [#141](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/141) - -### Changed - -- Upgraded OTEL collector image to [0.3.0](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/releases/tag/0.3.0) which brings following changes - - OTEL upgraded to 0.69.0 - - Added `filestorage` so it can be in processors - -### Fixed - -- Added optimizations to Log collector preventing Out of Memory situations [#137](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/137) - - Added `filelog.storage` to persist checkpoints in persistent storage, not in memory. - - Setting `max_concurrent_files` to 10 (from default 1024), this is main memory optimization, reducing amount of concurrent log scans - - Increased default memory limit to `700Mi` (which should be enough for large logs) - - Having by default `150Mi` difference between OTEL memory limit and Kubernetes memory limit, so that OTEL has enough buffer (this prevents OOMing) - -## [2.0.2] - 2023-01-18 - -### Added - -- Initial Helm release. -- Create Error reason as mapping from combination of Event Reason and Type fields [#115](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/115) -- Add support for HTTPS proxies [#117](https://github.com/solarwinds/swi-k8s-opentelemetry-collector/pull/117)