Skip to content

Latest commit

 

History

History
158 lines (114 loc) · 9.61 KB

2020-10-09.md

File metadata and controls

158 lines (114 loc) · 9.61 KB
Raw

This Week in Enhancements - 2020-10-9

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.

Merged Enhancements

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

There were 8 Merged pull requests:

  • 454: (0/49) [network] User Space Pod Interface and API Library (Billy99)

    Some applications require higher bandwidth and lower latency than can be provided by kernel networking. One solution is to bypass the kernel network stack and perform the networking in user space. An example of this is writing an application that does its own packet processing using the DPDK Libraries.

    This enhancement describes the changes needed to enable running DPDK based applications in a pod on an OCP deployment.

  • 466: (1/12) [console] Internationalization for console (rebeccaalpert)

    Our goal is to add internationalization to OpenShift's front-end (with some constraints) in order to improve the user experience for a wider range of customers worldwide. OpenShift does not currently have this functionality. We have created a POC using the react-i18next framework, which allows for the marking and translation of strings in the application, and moment.js for localizing dates and times. We have also added a plugin, i18next-pseudo, that generates pseudotranslations, and have connected with the Red Hat globalization team for early-stage testing and to learn more about the translation process used for other products at Red Hat. We want to implement localization more widely in the console.

  • 469: (1/16) [console] console user settings (christianvogt)

    The OpenShift Console has many features targeting different personas; from the Administrator to the Developer. There are limited features in Console today that save settings in order to enhance the user experience upon returning to any particular page. These settings are stored in browser local storage and therefore are not accessible in another browser or device. Many features that could benefit from customization do not provide customization today due to lack of persisted user settings.

    By introducing persisted user settings, many new and existing features will be designed with user settings in mind. Providing a richer experience which the user can tailor to their wants.

  • 478: (0/4) [tools] add a tool to report on activity in this repository (dhellmann)

    This PR introduced the tool that generates most of this weekly newsletter.

Minor Merged Updates

  • 476: (0/1) [this-week] Add weekly newsletters to the repository (russellb)
  • 490: (0/0) [tools] reformat summaries in newsletter (dhellmann)
  • 491: (0/0) [this-week] Add newsletter for 2020-10-02 (russellb)
  • 494: (0/0) [network] Update standalone kube-proxy ports (danwinship)

Closed Enhancements

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

There were 2 Closed pull requests:

  • 182: (0/11) [console] Web Terminal (l0rd)
  • 312: (0/30) [authentication] hardening openshift-apiserver default policy resources on upgrade clusters (vareti)

New Enhancements

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

There were 4 New pull requests:

  • 492: (37/37) [rhcos] add rhcos-inject enhancement (crawford)

    Throughout it's existence, OpenShift 4 has put a heavy influence on installation flows that make use of installer-provisioned infrastructure. This has largely been successful for predictable environments, such as AWS, Azure, or GCP, but it has (unsurprisingly) proven to make deployments into less predictable environments more difficult. These less predictable environments introduce a lot of variability into areas like network configuration, disk layout, and the life cycle of the machines and that makes it difficult or impossible for OpenShift to start from a foundation of shared assumption. To bridge this gap, admins need a way to inject a certain amount of customization before OpenShift installation can begin.

  • 493: (38/38) [storage] Add vsphere csi operator enhancement (gnufied)

    This enhancement proposes deployment of vSphere CSI driver on Openshift as a default component.

  • 495: (12/12) [baremetal] [WIP] Updates to SLO for baremetal enhancement proposal (andfasano)

    This change updates some of the implementation details of the previously approved enhancement to create a cluster-baremetal-operator to manage the bare metal provisioning subsystem.

  • 496: (0/0) [metering] Add slim install metering enhancement (timflannagan1)

    Due to the heavy footprint of the current Hadoop-based metering stack, it's difficult to install in smaller clusters, like a demo or developer environment, which is affecting potential adoption and opens the door to replicating existing metering functionality, with an in-house solution. This proposal outlines a path towards a slimmer metering stack, both in terms of reducing complexity and the overall resource footprint, and is centered around replacing the properties that the Hadoop, Hive, and Presto components offer, with Postgresql acting as both the underlying storage and compute engine. This will allow metering to set more reasonable defaults and reduces the barrier of entry when attempting to install metering on smaller or medium-sized environments as it requires little-to-none user intervention.

    In addition to replacing the existing storage and compute layers, metering will now utilize recording rules for its out-of-the-box Prometheus metrics it tracks, imports, and stores long term in Postgresql. Recording rules are a way to precompute frequent promQL queries, spreading the compute overhead over time, which increases metering performance at query-time and reduces the query-time load on Prometheus. In the case a user wants to extend the reporting capabilities of metering to track and store additional Prometheus metrics, those promQL queries will need to be evaluated at query-time instead of utilizing recording rules.

Active Enhancements

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

There were 12 Active pull requests:

  • 463: (80/319) [machine-api] Describing steps to support out-of-tree providers (Danil-Grigorev)
  • 486: (23/46) [local-storage] Adds proposal for auto partitioning in LSO (rohan47)
  • 432: (17/21) [ingress] Add Default Apps Domain enhancement (dustman9000)
  • 482: (16/40) [general] Add single-node-developer Cluster Profile (rkukura)

Single-node deployments of various styles are a hot topic right now. See also #440 - "single-node control plane based on installer bootstrap" and #302 - "thought-experiment: single-node cluster static pod creation"

  • 441: (12/42) [console] dynamic plugins for console (spadgett)

Enable operator authors to provide GUI components that integrate with the OpenShift console? Yes, please!

  • 411: (10/52) [installer] run the Assisted Installer on-premise as opposed to utilizing a cloud service (mhrivnak)

Being able to run the tools that implement the assisted installer workflow on premise, instead of requiring cloud service access, is a key requirement for their GA release, currently slated for OCP 4.7.

  • 467: (6/33) [machine-config] Add MCO Flattened Ignition proposal (hardys)
  • 453: (4/96) [network] Add a high level BGP use cases overview (russellb)
  • 447: (4/14) [insights] insights: Insights-gateway (iNecas)
  • 473: (4/70) [network] [WIP] Enable IPsec support in OVNKubernetes (markdgray)
  • 369: (2/22) [scheduling] Add new pod pod priority class (lilic)

This enhancement appears to be ready to merge.

  • 137: (2/99) [general] CLI in-cluster management (sallyom)

-- Russell Bryant Doug Hellmann