Merge release v0.1.17

Release v0.1.17

Author: ajfloeder

docs/repo-guides/release-nnf-sw/crd-upgrade-helpers.md

@@ -8,6 +8,85 @@ Together, these services implement [Upgrade existing objects to a new stored ver
 
 The ArgoCD `Application` bootstraps for the two services can be found in the [argocd-boilerplate](https://github.com/NearNodeFlash/argocd-boilerplate) repository under `environments/example-env/0-bootstrap1/`. NNF release manifests include the manifests for these services.
 
+## Debugging
+
+Monitor the progress of `kube-storage-version-migrator` or `nnf-storedversions-maint` by watching their logs and by watching changes to a specific CRD.
+
+For example, we may want to check the status of the `NnfStorageProfile` CRD after an upgrade that has added a new API version named "v1alpha7".
+
+Begin by finding the full name of the CRD:
+
+```console
+kubectl get crds -o custom-columns=NAME:.metadata.name | grep nnfstorageprofile
+```
+
+That will return a name of "nnfstorageprofiles.nnf.cray.hpe.com". Next, query the `status.storedVersions` list on that CRD:
+
+```console
+kubectl get crds nnfstorageprofiles.nnf.cray.hpe.com -o json | jq -rM '.status.storedVersions'
+```
+
+That might return a list of API versions:
+
+```console
+[
+  "v1alpha6",
+  "v1alpha7"
+]
+```
+
+This indicates that `kube-storage-version-migrator` has not finished migrating the resources of this type, or that `nnf-storedversions-maint` hasn't caught up.
+
+Check the progress of `kube-storage-version-migrator`. Begin by looking for the `StorageVersionMigration` resource it uses to track this CRD:
+
+```console
+kubectl get storageversionmigrations --no-headers -o custom-columns=NAME:.metadata.name | grep nnfstorageprofile
+```
+
+In this case it might return a name that looks like "nnfstorageprofiles.nnf.cray.hpe.com-mcsnm". Note the auto-generated suffix. The migrator replaces this resource any time it works on the CRD, and then it'll get a new suffix.
+
+Dump out the `StorageVersionMigration` resource to see the currrent status:
+
+```console
+apiVersion: migration.k8s.io/v1alpha1
+kind: StorageVersionMigration
+metadata:
+  creationTimestamp: "2025-04-28T21:12:23Z"
+  generateName: nnfstorageprofiles.nnf.cray.hpe.com-
+  generation: 1
+  name: nnfstorageprofiles.nnf.cray.hpe.com-mcsnm
+  resourceVersion: "35307472"
+  uid: ead6c3cf-8554-4400-aeea-e8ea5a59bf40
+spec:
+  resource:
+    group: nnf.cray.hpe.com
+    resource: nnfstorageprofiles
+    version: v1alpha7
+status:
+  conditions:
+  - lastUpdateTime: "2025-04-28T21:12:26Z"
+    status: "True"
+    type: Succeeded
+```
+
+That looks like it completed the migration of these resources to the new "v1alpha7" API version. In this case, the "lastUpdateTime" just happens to be a few seconds ago.
+
+The `nnf-storedversions-maint` should have had a chance to respond to this by now, so look at the CRD again:
+
+```console
+kubectl get crds nnfstorageprofiles.nnf.cray.hpe.com -o json | jq -rM '.status.storedVersions'
+```
+
+The list contains only the "v1alpha7" API:
+
+```console
+[
+  "v1alpha7"
+]
+```
+
+These activities can also be followed in the logs of the respective services.
+
 ## References
 
 ### Kubernetes