Bug 2074819
| Summary: | Change description of state migration in MTC documentation | ||
|---|---|---|---|
| Product: | Migration Toolkit for Containers | Reporter: | Richard Hoch <rhoch> |
| Component: | Documentation | Assignee: | Richard Hoch <rhoch> |
| Status: | CLOSED CURRENTRELEASE | QA Contact: | ssingla |
| Severity: | medium | Docs Contact: | Richard Hoch <rhoch> |
| Priority: | medium | ||
| Version: | 1.7.0 | CC: | akarol, ernelson, mberube, ssingla, xjiang |
| Target Milestone: | --- | Flags: | rhoch:
needinfo?
|
| Target Release: | 1.7.1 | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2022-05-18 14:49:34 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Description
Richard Hoch
2022-04-13 07:38:12 UTC
Marco -- Section 8.2.5 of the MTC documentation ("Running a migration plan in the MTC web console") https://access.redhat.com/documentation/en-us/openshift_container_platform/4.9/html-single/migration_toolkit_for_containers/index#migration-running-migration-plan-cam_migrating-applications-with-mtc has note after stage 2 that says "IMPORTANT: Do not use state migration to migrate a namespace between clusters. Use stage or cutover migration instead." Is that still correct for the web console or should I remove it and adjust the definition of "state" there (add that you can migrate between clusters, as well as removing the reference to Kubernetes resources)?
I don't know what the intention of this note was when it was originally added.
> ...migrate a namespace between clusters
This isn't specific enough to mean anything of value. If I read between the lines, I *think* what this is saying is that you should use stage/cutover to "migrate your workloads, including their state, on a namespace basis". The significance here is that the user needs to understand that in order to migrate the application itself (it's k8s objects), the images, AND the state, they need to use stage and cutover. If they only care about migrating data because their workload's deployments are managed by a CD system like Argo, they should use state only migration to get their data transferred to their new destination.
When doing a state only migration, it is not possible to migrate k8s resources that could be considered application state (like secrets/configmaps) as well as PVCs. You need to adjust the MigPlan on the cli to include the k8s resources.
So to try to answer your question, I think this NOTE should remain, and that "migrate a namespace between clusters" needs more specificity. I will also defer to Marco however, and I'm trying to provide the context here as best as I can.
please change doc to: State copies selected persistent volume claims (PVCs) instead of : State copies selected persistent volume claims (PVCs) and Kubernetes resources that constitute an application’s state. You can use state migration to migrate a namespace within the same cluster. (Even if the current version could be technically correct, it's misleading. Most people at this point just want to know what state only migrate PVCs). Marco. Please review thids PR: https://github.com/openshift/openshift-docs/pull/44543 Looks good. thanks. Aziza or Xin, Please review this PR: https://github.com/openshift/openshift-docs/pull/44543 Sachin, Please Review Merged. |