Uploaded image for project: 'OpenShift Container Platform (OCP) Strategy'
  1. OpenShift Container Platform (OCP) Strategy
  2. OCPSTRAT-725

LVM Storage: replace upstream topolvm

XMLWordPrintable

    • 100% 100%
    • 0

      Outcome Overview

      LVM Storage is the primary CSI driver used by SNO / MicroShift, but also other edge deployments where node local storage is good enough. It is based on upstream topolvm, which turns out to be not of enterprise grade quality, causing lots of blockers for new features (like encryption or RAID support) or fixing of bugs. The goal is to replace topolvm with a new implementation which can reach operator level "Full Lifecycle" and beyond.

      The outcome for customers is a much more stable and robust, trustworthy local storage implemenation.

       

      Success Criteria

      Provide a seamless migration to the new implementation that provides feature parity with the existing one. 

      Requirements

      • Feature parity with current implementation (esp. CSI Snapshots)
      • Multi-Node Support (local volumes per node, NO distributed / remote storage)
      • stick with k8s PVC sizes (e.g. MB disk size)
      • resilience/self healing (e.g. deleting PVC/LVMCluster objects, adding/remove disks does not need a uninstall/re-install)
      • CSI / K8S best practises (e.g. reconciling all entities)
      • Keep the fooprint (CPU, Memory, Image Size, number of images, number of pods) as low as possible to be a fit for resource constraint edge deployments (e.g. SNO with 1 core control plane).
      • Be prepared for operator Level 3 and above (Full Lifecycle)
      • Seamless upgrade from old topolvm to new implemenation
      • Make it simple to implement advanced features like Disk Encryption, RAID support etc.

      Expected Results (what, how, when)

      What incremental impact do you expect to create toward the company's Strategic Goals by delivering this outcome?  (possible examples:  unblocking sales, shifts in product metrics, etc. {}{} provide links to metrics that will be used post-completion for review & pivot decisions). {}For each expected result, list what you will measure and +when you will measure it (ex. provide links to existing information or metrics that will be used post-completion for review and specify when you will review the measurement such as 60 days after the work is complete)

       

      This will require at least two releases. 

      • Dev Preview stage with only MVP release, very basic CSI driver capabilities.
      • TP/GA release with feature parity and seamless migration.

      We expect bugs/support cases  being opend due to race conditions, instabilities, lack of robustness  etc. to be reduced by 95%

       

      Post Completion Review – Actual Results

      TBD.

            dfroehli42rh Daniel Fröhlich
            dfroehli42rh Daniel Fröhlich
            Chad Scribner Chad Scribner
            Steve Gordon Steve Gordon
            Votes:
            0 Vote for this issue
            Watchers:
            6 Start watching this issue

              Created:
              Updated:
              Resolved: