The container image mirror settings that did nothing are gone

The deployment chart offered a containerImages block — an upstream registry, a username, an allowlist of repositories, a cache directory and size — for an installation to serve container images to the rest of the fleet through its own /v2 endpoint. The code behind it was built and tested, but no portal ever switched it on. So an operator who filled in the block got a configuration that looked wired and did nothing: no mirror, and no error saying so. Even the check that was meant to refuse a half-configured mirror could never run, because it lived inside the part that was never switched on.

The fleet's own registry, cr.meshweaver.cloud, is a separate registry service, not this code. With that settled, the unused mirror is removed, together with the containerImages chart settings and the ContainerImages__… configuration keys they rendered.

What you need to do: nothing, unless one of your own values files still sets containerImages — then delete that block and any ContainerImages-Password Key Vault mapping that went with it. They had no effect before and have none now. Where an installation pulls its platform images from is unchanged: the chart default is still the upstream registry, and an installation that pulls from cr.meshweaver.cloud says so with selfUpdate.registry and portal.imagePullSecret, as described in A Container Registry in Memex.

Reconnecting…
The connection to the server was interrupted. Trying to restore it…
Trying again…
The connection could not be restored. Reloading the page…
The server was updated. Reloading the page to pick up the latest version.