Splunk Connect for Kubernetes 1.4.0 release

Splunk Connect for Kubernetes 1.4.0 release.

This release also includes various bug fixes and improvements from community contributions, Thank You very much for all the contributions!!

Release Highlights

• Collect pod labels from kubernetes.
• Annotation features (index routing, blacklisting, sourcetype overwrite)
• All images are upgraded to Redhat UBI images
• Subcharts folder restructured
• Support using environment variable for HEC host address
• CI improvements
• Bug fixes

Breaking Changes

• SCK 1.4.0 will not work with any previous image tags than https://github.com/splunk/splunk-connect-for-kubernetes/blob/release/1.4.0/PLUGIN_VERSIONS.sh
• For logging, namespace index routing is deprecated. Please use annotation feature to control routing https://github.com/splunk/splunk-connect-for-kubernetes/tree/release/1.4.0#managing-sck-log-ingestion-by-using-annotations

Splunk Connect for Kubernetes 1.3.0 release

Splunk Connect for Kubernetes 1.3.0 release
Splunk Connect for Kubernetes 1.3.0 adds support for Openshift 3.9, 4.1 and 4.2, IBM Kubernetes Service, Azure Kubernetes Service and ingesting logs from AWS Fargate and AWS ECS using AWS Firelens

This release also includes various bug fixes and improvements from community contributions, Thank You very much for all the contributions!!

Release highlights

• Add support for Splunk's ingest API(sdc.splunkbeta.com/reference/api/ingest/v1beta2)
• Specification of indexed fields to be added to logging and objects events
• Add support for ingesting Kubernetes Apiserver Audit logs
• Index routing for Kubernetes objects
• Allows specification of private docker registries
• Specification of custom filters
• Generic logformat support, added for CRI-O and containerd support can be used to specify other formats
• Add Support for Custom Metadata Tags
• Enable monitoring of logging fluentd daemonset using prometheus
• Added support for path and exclude_path for logging
• Add priorityClassName to helm charts to allow definition of pod priority class

Bug fixes and improvements

• Updated configuration for systemd plugin to use matches as filters is deprecated
• For multiline events, use the timestamp of the first event
• CI improvements(linting helm charts, run ci locally on CircleCI)

Shoutout to @mehstg, @vidarw, @kousik93, @jeffywu, @rhockenbury, @jarleborsheim, @gmichels, @tonyswu, @svrc-pivotal, @stonith, @nokesc, @dmizelle for their contributions and engagement with our product.

Sub Module Updates

Fluent Plugin Splunk HEC Release 1.2.0

• Add support for Splunk's ingest API(sdc.splunkbeta.com/reference/api/ingest/v1beta2)

Fluent Plugin Kubernetes Objects 1.1.2

• This release of Fluent Plugin for Kubernetes Objects updates an issue where watching would fail after an hour.

Fluent Plugin Kubernetes Metrics 1.1.2

• Fluent Plugin Kubernetes Metrics 1.1.2 includes robust error handling in case of missing metrics

Fluent Plugin Kubernetes Metrics Aggregator 1.1.1

• Fix for incorrect pod namespace
• Fixes for incorrect metric units
• Fix for missing error handling for http requests
• Fix for unit test to verify handling of request timeouts and make them fast

Splunk Connect for Kubernetes 1.2.0 release

Splunk Connect for Kubernetes 1.2.0 release

Splunk Connect for Kubernetes 1.2.0 adds extra support for Splunk app for Infrastructure and bug fixes and improvements from community contributions, Thank You very much for all the contributions!!

• Support for Splunk App for Infrastructure
• Configurable option to use existing secret for HEC token
• Configurable option for using variable docker logs path
• Bug Fixes and improvements
  - Configurable option for preventing metrics aggregator from running on master nodes
  - Added missing interval config for metrics daemonset
  - Added missing rbac create condition for metrics
  - Add appVersion to all the helm charts
  - Allow setting cluster name globally for all components
  - Make systemd components persist a position file for logging
  - Fix for checkpoint issues in the splunk-kubernetes-objects Helm chart for objects
  - Fix for concat plugin recursive deadlock issues for logging

Shoutout to @hexecute, @eye0fra, @VincentS, @Faqa, @mehstg, @ivanetchart, @sbkg0002, @DipeshMitthalal, @csmarchbanks, @peter-moffatt for their contributions and engagement with our product.

Sub Module Updates
Fluent Plugin Kubernetes Metrics 1.1.1

• Fixes to support Splunk App for Infrastructure
• Added Contribution Guidelines
• Added Project issue templates

Fluent Plugin Splunk HEC Release 1.1.1

• Downgrade version of Fluent concat plugin to remedy noisy recursive deadlock issues
• Added Contribution Guidelines
• Added Project issue templates

Splunk Connect for Kubernetes 1.1.0 release

Splunk Connect for Kubernetes 1.1.0
Splunk Connect for Kubernetes 1.1.0 replaces Heapster as the method for collecting metrics from Kubernetes. Splunk built Fluentd plugins will now query, aggregate and send Kubernetes metrics to Splunk.

• [Feature] Kubernetes namespace to Splunk Index routing. Enable routing of events and metrics from multiple namespaces to multiple indexes.
• [Feature] Ability to tag up events and metrics with custom cluster_name metadata
• Improvement to the projects CI and testing framework to enable quicker project development and ability to test and accept community contributions.

The update to 1.1.0 also includes a refresh of the sub-components that make up SCK.

Fluentd Plugin Splunk HEC v1.1.0

• Updated 3rd party libraries to latest versions
• Improved projects Testing and CI automation
• Reduced dependency on Ruby 2.4 to support Debian
• Added time_key parameter configurable
• Added configurable parameters for Net HTTP Persistent

Fluentd Plugin Kubernetes Metrics v1.1.0
The release of Fluentd Plugin Kubernetes metrics is a fluentd input plugin which collects metrics from Kubernetes from multiple api endpoints and forwards to fluentd.
The plugin collects metrics from:

• The kubelet summary api
• The kubelet stats api
• The cAdvisor metrics api

Fluentd Plugin Kubernetes Metrics Aggregator v1.1.0
The release of Fluentd Plugin Kubernetes Metrics Aggregator is a fluentd input plugin which collects cluster, namespace and node level metrics from Kubernetes

Fluentd Plugin Kubernetes Objects v1.1.0
The release of Fluentd Plugin for Kubernetes Objects updates the projects 3rd party libraries.

V1.0.1

This is a minor bug fixing release.

Please see https://github.com/splunk/splunk-connect-for-kubernetes/milestone/1?closed=1 for the list of fixed bugs.

This release updated fluent-plugin-splunk-hec to v1.0.1, please see https://github.com/splunk/fluent-plugin-splunk-hec/milestone/1?closed=1 for the changes of that plugin.

v1.0.0

Regenerated manifests.

v1.0.0-beta.1

Updates:

• refactored configurations for log sources
• multiline support
• custom sourcetype support
• refactored configurations for objects to support multiple apiGroups.

This is released under SPLUNK PRE-RELEASE SOFTWARE LICENSE AGREEMENT.

v1.0.0-alpha.4

Updates:

• metrics names now have "kube." prefix
• "kube:container:" prefix of sourcetype of container logs has been removed
• "pod", "namespace", "container_name" and "container_id" have been extracted and indexed as index-time fields for container logs

This is released under SPLUNK PRE-RELEASE SOFTWARE LICENSE AGREEMENT.

v1.0.0-alpha.3

Fixed a couple issues with RBAC permissions for splunk-kubernetes-objects chart.

Released under SPLUNK PRE-RELEASE SOFTWARE LICENSE AGREEMENT.

v1.0.0-alpha.2

This is a bug fixing for pre-release v1.0.0-alpha.1. Mainly fixed bugs in the logging fluentd config files.

Released under SPLUNK PRE-RELEASE SOFTWARE LICENSE AGREEMENT.