Operator 4.2.13

Other Changes

• The default WebLogic Monitoring Exporter version is now 2.2.2.
• Support has been significantly improved for customer use cases where a third-party organization or separate infrastructure team is responsible for managing the Kubernetes cluster and the team installing the operator has privilege only within a single namespace. The infrastructure team will be responsible for installing the CRD documents. The operator must be installed without including the webhook deployment. See Local namespace only with cluster role binding disabled for details and options.

Full Changelog: v4.2.12...v4.2.13

Operator 4.2.12

Bug Fixes

• Resolved an issue where the $DOMAIN_HOME/lib directory was not properly copied for auxiliary images

Full Changelog: v4.2.11...v4.2.12

Operator 4.2.11

Bug Fixes

• Resolved an issue where managed server rolling doesn't proceed because the readiness probe configuration has changed but where the new probe configuration isn't valid until after the instance is restarted.

Full Changelog: v4.2.10...v4.2.11

Operator 4.2.10

Bug Fixes

• Resolved an NPE related to FiberGate processing that could prevent the operator from initializing properly.

Full Changelog: v4.2.9...v4.2.10

Operator 4.2.9

Bug Fixes

• Resolved an issue where the operator did not recover from a prolonged outage of the Kubernetes control plane, such as during a Kubernetes version upgrade.

Full Changelog: v4.2.8...v4.2.9

Operator 4.2.8

Bug Fixes

• Resolved #5070 regarding the operator not properly honoring the pod and container security context defaults or configuration.

Full Changelog: v4.2.7...v4.2.8

Operator 4.2.7

Bug Fixes

• Resolved an issue where the operator during initialization may attempt to create pods for WebLogic Server instances that already have running pods. This behavior, while not affecting the already running pods, could confuse Kubernetes resource quota support causing Kubernetes to incorrectly report higher than expected CPU and memory usage.
• Resolved an issue where the operator pod would take longer than expected to shut down.

Full Changelog: v4.2.6...v4.2.7

Operator 4.2.6

Other Changes

• The default WebLogic Monitoring Exporter version is now 2.2.1.

Bug Fixes

• Resolved an issue related to the use of the domainBin and domainLibraries model directories when using model-in-image
• Resolved an issue related to correctly generating the volume and volume mounts for the introspector job when referencing a secret that has the dot character in the name of the secret.

Full Changelog: v4.2.5...v4.2.6

Operator 4.2.5

Bug Fixes

• Corrected a fix included in 4.2.4 for an issue in the introspection WLST and shell scripts to use jq to validate deltas when a model is changed to prevent failures in WLST when the model is very large

Full Changelog: v4.2.4...v4.2.5

Operator 4.2.4

Additional Features

• Customers can now configure startup probes using the startupProbe field similarly to existing support for configuring liveness and readiness probes

Other Changes

• The error reporting is improved when the use of topologySpreadConstraints causes pods to be unschedulable

Bug Fixes

• Resolved an issue in the quick-start sample YAML file defining the sample Domain and Cluster resources that resulted in the operator reporting that the Domain was invalid because the Cluster resource did not yet exist
• Resolved an issue in the sample waitForDomain.sh domain lifecycle script where a sed call was improperly written resulting in incomplete status information about clusters
• Resolved an issue in the introspection WLST and shell scripts to use jq to validate deltas when a model is changed to prevent failures in WLST when the model is very large
• Resolved an issue in the liveness probe and shutdown lifecycle scripts to more quickly exit the container after the server instance has stopped

Full Changelog: v4.2.3...v4.2.4