Uploaded image for project: 'OpenShift Migration Toolkit for Containers'
  1. OpenShift Migration Toolkit for Containers
  2. MIG-853

DVM pod limits should never breach cluster limitranges


    • False
    • False
    • Passed
    • Sprint 212+213

      From: https://coreos.slack.com/archives/CFN0XJEPR/p1632835181185500

      > When the DVM PODs start on source namespaces sometimes fails because our LimitRange has a max value to containers of 500m but DVM needs 1000m to run. This make us to change the LimitRange in all our projects. Is there a way to reduce this default value to DVM PODs?

      We currently parameterize this value: https://github.com/konveyor/mig-operator/blob/master/docs/usage/MigrationControllerConfigQuickView.md#rsync-specific-configuration

      However, we should always modify the default based on the limitrange since we know that values out side of this range are going to fail. If the value is so low that it's going to practically prohibit DVM performance, we should consider either failing the migration with an explicit error, or at the very least, an explicit warning so the user understands what the problem is and how to address it.

            pgaikwad1@redhat.com Pranav Gaikwad
            ernelson@redhat.com Erik Nelson (Inactive)
            Prasad Joshi Prasad Joshi
            1 Vote for this issue
            4 Start watching this issue