Skip to content

Latest commit

 

History

History
128 lines (89 loc) · 8.71 KB

2020-10-23.md

File metadata and controls

128 lines (89 loc) · 8.71 KB

This Week in Enhancements - 2020-10-23

Hello, everyone!

Here you will find an overview of activity in the OpenShift enhancements repository in the last week. You can find previous editions on github.

Single-node Deployments And Your Operator

As mentioned 2 weeks ago, there are now several enhancements discussing variations of single-node deployments. (It can't be a "cluster" if there's only one node, right?) Both the single-node-developer and single-node production approaches build on the cluster profile work started for IBM Cloud, which may be necessary but not sufficient to provide reliable operation of OpenShift without multiple control plane nodes. The two sets of requirements have some overlap, but the clusters will end up being used under different conditions and with different expectations, so we don't currently expect them to be identical. We're looking for input from all teams on the enhancements, to understand the impact of making operators run on a single-node. Please take some time today or early next week to leave comments on the enhancements.

Merged Enhancements

<PR ID>: (activity this week / total activity) summary

There were 5 Merged pull requests:

  • 419: (0/88) update: Enhancement for adding a manifest annotation for object removal (jottofar)

    Currently CVO managed object removal is handled by jobs. Jobs have many immutable components, so they aren't a great match for CVO managed manifests. This enhancement replaces that approach with a new manifest annotation requesting the CVO delete the in-cluster object instead of creating/updating it. This will provide a more straightforward way for engineering to remove content.

  • 432: (0/29) ingress: Add Default Apps Domain enhancement (dustman9000)

    This enhancement allows for the configuration of a default apps domain, which can be used to specify an alternative default domain for user-created routes. This alternative domain, if specified, overrides the default cluster domain for the purpose of determining the default host for a newly created route.

  • 437: (12/27) storage: Add OpenStack Cinder CSI driver support (Fedosin)

    In the past storage drivers have been provided either in-tree or via external provisioners. With the move to advocating CSI as the way to deliver storage drivers, OpenShift needs to move the in-tree drivers (and others that our customers need) to use CSI. With the focus on operators in Red Hat OCP 4.x these CSI drivers should be installed in the form of an Operator.

  • 441: (15/110) console: dynamic plugins for console (spadgett)

    OpenShift Console currently has static plugins that allow teams to contribute features to the UI such as CNV and OCS. These plugins live in a packages directory inside the openshift/console repo and extend console through a well-defined plugin API. Static plugins are built with console and included in the console image. There are currently over a half-dozen plugins.

    In addition to static plugins, we'd like to introduce dynamic plugins. Dynamic plugins are contributed by operators and loaded at runtime. Eventually we want to migrate static plugins to dynamic plugins to decouple them from the console and allow plugins to be updated more frequently.

Merged Minor Updates

  • 506: (0/0) general: Add Network Edge members to approvers (Miciah)

Closed Enhancements

<PR ID>: (activity this week / total activity) summary

There was 1 Closed pull request:

  • 72: (0/14) installer: Bootstrap Ignition Config: provisional state (iamemilio)

New Enhancements

<PR ID>: (activity this week / total activity) summary

There were 5 New pull requests:

  • 503: (4/4) windows-containers: WMCO upgrade enhancement update (sebsoto)

    This enhancement allows a cluster administrator to upgrade a deployed Windows Machine Config Operator (WMCO), and associated Windows node components, to a newer version.

  • 504: (42/42) general: single-node production deployments (dhellmann)

    Add a new single-node-production-edge cluster profile for production use in "edge" deployments on servers that are not considered to be resource-constrained.

    A cluster deployed using this profile will differ from the default self-managed-highly-available cluster profile in several significant ways:

    • The single node serves as both the cluster’s control plane and as a worker node.
    • Many operators will be configured to reduce the footprint of their operands, such as by running fewer replicas.
    • In-place upgrades will not be supported by the first iteration of this cluster profile.

    One example of this use case is seen in telecommunication service providers implementation of a Radio Access Network (RAN). This use case is discussed in more detail below.

  • 507: (50/50) authentication: new/auth: allow users to manage their tokens (stlaz)

    While users are quite capable of retrieving access tokens to be able to work with OpenShift services, their capabilities to see the tokens they created and to handle them is a bit lacking. This enhancement describes how the self-management of users' access tokens can be improved.

  • 508: (0/0) console: [WIP] Add Customize Developer Catalog Categories enhancement (jerolimov)

    The OpenShift Console provides a "Developer Catalog" which enables the user to create apps and services based on Operator backed services and templates. The entries are provided by operators, helm repositories, event sources, etc.

    Cluster admins can provide additional templates by installing operators or adding helm repositories. But the catalog filter navigation (All Items, Languages, Databases, Middleware, ...) is currently hard coded. We want enable the cluster admins to add, modify, delete (or hide) this catalog filter navigation.

Updates to Existing Enhancements

  • 505: (0/0) general: Add mention of operator CR to operator dev doc (marun)

Active Enhancements

<PR ID>: (activity this week / total activity) summary

There were 12 Active pull requests:

  • 500: (72/72) operator-sdk: Automated downstream Operator SDK into the OpenShift ecosystem. (jmrodri)
  • 463: (46/401) machine-api: Describing steps to support out-of-tree providers (Danil-Grigorev)
  • 501: (20/20) cluster-logging: LOG-928: Creating elasticsearch hot warm phase architecture proposal (ewolinetz)
  • 496: (16/88) metering: Add slim install metering enhancement (timflannagan1)
  • 482: (16/74) general: Add single-node-developer Cluster Profile (rkukura)
  • 429: (14/94) kube-apiserver: Enforce label scheme (mmirecki)
  • 467: (8/23) machine-config: Add MCO Flattened Ignition proposal (hardys)
  • 448: (8/97) cluster-logging: Proposal to enable JSON data processing (lukas-vlcek)
  • 369: (8/32) scheduling: Add new pod priority class (lilic)
  • 146: (3/195) installer: openstack: Add Baremetal Compute Nodes RFE (pierreprinetti)
  • 493: (2/42) storage: Add vsphere csi operator enhancement (gnufied)
  • 411: (1/28) installer: run the Assisted Installer on-premise as opposed to utilizing a cloud service (mhrivnak)

  • Doug Hellmann
  • Russell Bryant