Document change to update known issue. (vmware-tanzu#560)

* Document change. Signed-off-by: Liping Xue <[email protected]> * Address comment from Xing. Signed-off-by: Liping Xue <[email protected]> * Remove files that are not needed for this change. Signed-off-by: Liping Xue <[email protected]> --------- Signed-off-by: Liping Xue <[email protected]>
lipingxue · Mar 11, 2024 · bc27a1c · bc27a1c
1 parent 3e6cb89
commit bc27a1c
Show file tree

Hide file tree

Showing 2 changed files with 24 additions and 0 deletions.
diff --git a/docs/known_issues.md b/docs/known_issues.md
@@ -64,3 +64,24 @@ Users might require ways to provide registry credentials when deploying velero a
 ## Volume Expansion Not Supported when there are Snapshots
 
 If upload fails during a backup, the local snapshot will not be deleted. Velero vSphere Plugin will keep retrying the upload. If the user wants to expand volume, it will fail because volume expansion is not supported when there are snapshots on the volume.
+
+## Backup and Restore with vSphere with IPV6
+
+Users may hit the following error `too many colons in address` if they specify VC IP in IPV6 format in `csi-vsphere.conf` file and create `velero-vsphere-config-secret` from this file when deploying Velero Plugin for vSphere. 
+
+```
+time="2023-11-06T10:53:21Z" level=error msg="Failed at copying to remote repository" Local PEID="ivd:fdc85e45-7c61-453f-9487-b7e47d020cd9:a6183167-d52d-4a83-932f-77104fac93e3" error="dial tcp: address 2620:124:6020:c304:0:a:0:763:443: too many colons in address" logSource="/go/src/github.com/vmware-tanzu/velero-plugin-for-vsphere/pkg/dataMover/data_mover.go:120"
+```
+The workaround to this problem is to use VC FQDN instead of IPV6 format in `csi-vsphere.conf` file. See the following example.
+
+```
+[Global]
+cluster-id = "cluster0"
+
+[VirtualCenter "vCenter-FQDN"]
+insecure-flag = "true"
+user = "[email protected]"
+password = "password"
+port = "443"
+datacenters = "dc0"
+```
diff --git a/docs/supervisor-notes.md b/docs/supervisor-notes.md
@@ -16,6 +16,8 @@ Below is the networking diagram of velero-plugin-for-vsphere in vSphere with Tan
 
 Certain resources with the **vSphere with Tanzu** Supervisor Cluster need to be created by the Supervisor Cluster and cannot be restored. The **Velero Plugin for vSphere** blocks backup and restore of these resources and will generate errors if an attempt is made to backup or restore these resources and the backup or restore will be marked as "Partially Failed".
 
+Similarly, certain resources with the **Tanzu Kubernetes Grid** multicloud (TKGm) need to be created by the TKGm cluster and cannot be restored. The **Velero Plugin for vSphere** blocks backup and restore of these resources and will generate errors if an attempt is made to backup or restore these resources and the backup or restore will be marked as "Partially Failed".
+
 A configmap named `velero-vsphere-plugin-blocked-resources-list` will be created in the namespace where you install velero during **Velero Plugin for vSphere** installation. Exclude the resources in the configmap from your backups to avoid these errors. After the plugin is deployed, users can edit the configmap to remove resources from being blocked, or add more resources to be blocked. Users should not delete the configmap.
 
 **Note**: The configmap `velero-vsphere-plugin-blocked-resources-list` will be reset to the default list only when user try to reboot velero pod.
@@ -32,6 +34,7 @@ The default list of blocked resources in configmap is:
  	certificaterequests.cert-manager.io
  	certificates.cert-manager.io
  	challenges.acme.cert-manager.io
+	clusterclasses.cluster.x-k8s.io
  	clusterissuers.cert-manager.io
  	clusternetworkinfos.nsx.vmware.com
  	clusterresourcesetbindings.addons.cluster.x-k8s.io