Releases: rchakode/kube-opex-analytics
Releases · rchakode/kube-opex-analytics
Kubernetes Opex Analytics 0.4.7
This is a hotfix release:
- Fixing issue #32 related to unsupported SHELL interpreter in Docker entrypoint script.
- Add documentation on option
KOA_ENABLE_PROMETHEUS_EXPORTER
which allows to enable or disable the Prometheus exporter of Kubernetes Opex Analytics.
Enjoy!
Kubernetes Opex Analytics 0.4.6
This is a hotfix release with the following changes:
- Enabling the Prometheus exporter is now optional: it now actually requires to set the startup configuration variable
KOA_ENABLE_PROMETHEUS_EXPORTER
totrue
. This change is motivated by issues #30 and #31 which, after investigations, seem to be related the uwgsi dipatcher needed to enable the Prometheus exporter.
Enjoy!
Kubernetes Opex Analytics 0.4.5
This release improve Docker image and Kubernetes deployment hardening
- Fixed Dockerfile to set file permissions later than some resources creation
- Restore securtyContext as default settings
- Restore default not-privileged fsGroup
Enjoy!
Kubernetes Opex Analytics 0.4.4-beta1
Changes related to Docker image build:
- disable volume in Dockerfile
- merge Docker image build steps in a single step
Kubernetes Opex Analytics 0.4.4
Single noticeable change:
- fixed Docker image action file
Kubernetes Opex Analytics 0.4.3
Changes in this release include:
- Add configuration option to specify cacert file for self-signed certifcate on Kubernetes API (#28)
- Enable endpoint to retrieve analytics datasets from third-party components.
- Enable data volume in Docker image.
- Automated Docker image build through Github Actions.
Enjoy!
Kubernetes Opex Analytics 0.4.2
This is a maintenance release with the following changes:
- Added compatibility with Kubernetes Security Context in order to run Kubernetes Opex Analytics's pods in unprivileged mode by default (i.e. programs inside the containers are executed under a
non-root
user account). This change fixes issue #15. However to avoid regression due to resources (e.g. files) created with former versions where containers were executed in privileged mode, it's still possible to launch the pods in privileged mode. In this latter case you need to edit the Helm values.yaml file and set the variablesecurityContext.enabled
accordingly. It must be set with a boolean value,true
to run pods in unprivileged mode andfalse
otherwise. - Misc improvements on uwsgi startup process.
Enjoy.
Kubernetes Opex Analytics 0.4.1
This mainly a hotfix release:
- Fixed issue #23 where certain memory capacity units were not properly parsed.
- Enhancement: Now RBAC token are reloaded before pulling Kubernetes API.
Enjoy.
Kubernetes Opex Analytics 0.4.0
Here is the list of changes:
- PR #14: Introduced a Prometheus Operator (along with ServiceMonitor), as well as a proper RBAC serviceaccount.
- PR #22 (related to enhancement #4): Introduced the possibility to export in JSON or CSV format, datasets related to hourly, daily and monthly usage analytics. Those datasets concern both CPU and memory usage, and should be exported separately.
- PR #21: Fixed date parsing for the frontend; this actually allows to also handle epoch timestamp.
- Fixed issue #20: Cleaning Bearer token in proxy mode, what lead to unauthorized access from Kubernetes API.
- Cleaning Helm deployment resources: Removed (useless) namespace binding on clusterrole and clusterrolebinding definitions.
Many thanks to @volatilemolotov and @ccamel for their wonderful contributions in this release.
Enjoy!
Kubernetes Opex Analytics 0.3.1
June 29, 2019. Releasing a quick hotfix version of Kubernetes Opex Analytics:
- Fixed issue #13 related to parsing of certains units of pods cpu metrics.
Enjoy.