Updating a configmap does not automatically result in the pods running with the latest version, the pod needs to be restarted to mount the latest version of the configmap/secret. Helm does automate this via the checksum annotations on the pods, but that only works if the helm template is re-rendered.
Building a new image does not always update the deployment with the new image hash. Openshift translates the image stream tag to a specific hash when the deployment config is applied. This should be updated via image triggers, but doesn't reliably work when multiple image triggers are set. If in question, verify and apply the deployment again, which should result in the latest image version being used.
If a node has disk-pressure, free up some space, then potentially restart k3s to immediately update the status.