Cluster Observability Operator release notes

You can now configure WebTLS for Prometheus and Alertmanager.

CVE-2024-6104

CVE-2024-24786

Previously, when you deleted the dashboard UI plugin, the consoles.operator.openshift.io resource still contained console-dashboards-plugin. This release resolves the issue. (COO-152)

Previously, the web console did not display the correct icon for Red Hat COO . This release resolves the issue. (COO-353)

Previously, when you installed the COO from the web console, the support section contained an invalid link. This release resolves the issue. (COO-354)

Previously, the cluster service version (CSV) for COO linked to an unofficial version of the documentation. This release resolves the issue. (COO-356)

Previously, metrics were not available to normal users when accessed in the Developer perspective of the web console, by clicking on Observe → Logs. This release resolves the issue. (COO-288)

Previously, the troubleshooting UI plugin used the wrong filter for network observability. This release resolves the issue. (COO-299)

Previously, the troubleshooting UI plugin generated an incorrect URL for pod label searches. This release resolves the issue. (COO-298)

Previously, there was an authorization vulnerability in the Distributed tracing UI plugin. This release resolves the issue and the Distributed tracing UI plugin has been hardened by using only multi-tenant TempoStack and TempoMonolithic instances going forward.

With this release, you can now use tolerations and node selectors with MonitoringStack components.

Previously, the logging UIPlugin was not in the Available state and the logging pod was not created, when installed on a specific version of OKD. This release resolves the issue. (COO-260)

With this release, the Cluster Observability Operator adds backend support for future OKD observability web console UI plugins and observability components.

With this release, the Cluster Observability Operator supports installing and managing observability-related plugins for the OKD web console user interface (UI). (COO-58)

Previously, if you tried to access the Prometheus web user interface (UI) at http://<prometheus_url>:9090/graph, the following error message would display: Error opening React index.html: open web/ui/static/react/index.html: no such file or directory. This release resolves the issue, and the Prometheus web UI now displays correctly. (COO-34)

CVE-2023-45142

Previously, certain cluster service version (CSV) annotations were not included in the metadata for COO. Because of these missing annotations, certain COO features and capabilities did not appear in the package manifest or in the OperatorHub user interface. This release adds the missing annotations, thereby resolving this issue. (COO-11)

Previously, automatic updates of the COO did not work, and a newer version of the Operator did not automatically replace the older version, even though the newer version was available in OperatorHub. This release resolves the issue. (COO-12)

Previously, Thanos Querier only listened for network traffic on port 9090 of 127.0.0.1 (localhost), which resulted in a 502 Bad Gateway error if you tried to reach the Thanos Querier service. With this release, the Thanos Querier configuration has been updated so that the component now listens on the default port (10902), thereby resolving the issue. As a result of this change, you can also now modify the port via server side apply (SSA) and add a proxy chain, if required. (COO-14)