From 0a86a4772068e201241a8188b4c6abdb136988a0 Mon Sep 17 00:00:00 2001 From: Michael Mattsson Date: Wed, 1 May 2024 09:09:24 -0700 Subject: [PATCH] Minor updates and fixings (#204) Signed-off-by: Michael Mattsson --- .../hpe_alletra_6000/index.md | 1 + .../hpe_alletra_storage_mp/index.md | 2 +- docs/partners/redhat_openshift/index.md | 49 ++++++++++++++++++- 3 files changed, 50 insertions(+), 2 deletions(-) diff --git a/docs/container_storage_provider/hpe_alletra_6000/index.md b/docs/container_storage_provider/hpe_alletra_6000/index.md index 72254000..184ea750 100644 --- a/docs/container_storage_provider/hpe_alletra_6000/index.md +++ b/docs/container_storage_provider/hpe_alletra_6000/index.md @@ -92,6 +92,7 @@ Consult the [compatibility and support](../../csi_driver/index.md#compatibility_ - Striped volumes on grouped arrays are not supported by the CSI driver. - The CSP is not capable of provisioning or importing volumes protected by Peer Persistence. +- When using an FC only array and provisioning RWX block volumes, the "multi_initiator" attribute won't get set properly on the volume. The workaround is to run `group --edit --iscsi_enabled yes` on the Array OS CLI. ## StorageClass Parameters diff --git a/docs/container_storage_provider/hpe_alletra_storage_mp/index.md b/docs/container_storage_provider/hpe_alletra_storage_mp/index.md index fe23f962..cd7eb550 100644 --- a/docs/container_storage_provider/hpe_alletra_storage_mp/index.md +++ b/docs/container_storage_provider/hpe_alletra_storage_mp/index.md @@ -299,7 +299,7 @@ Use this parameter to specify a subset of Fibre Channel (FC) ports on the array | Parameter | Option | Description | | ----------- | ------ | ----------- | -| fcPortsList | Text | Comma separated list of available FC ports. | +| fcPortsList | Text | Comma separated list of available FC ports. Example: "0:5:1,1:4:2,2:4:1,3:4:2" | ### Support diff --git a/docs/partners/redhat_openshift/index.md b/docs/partners/redhat_openshift/index.md index 877536b1..21282df9 100644 --- a/docs/partners/redhat_openshift/index.md +++ b/docs/partners/redhat_openshift/index.md @@ -77,6 +77,7 @@ Once the steps have been followed for the particular version transition: `oc delete crd/hpecsidrivers.storage.hpe.com` - [Uninstall](#uninstall_the_hpe_csi_operator) the HPE CSI Operator for Kubernetes - Proceed to installation through the [OpenShift Web Console](#openshift_web_console) or [OpenShift CLI](#openshift_cli) +- Reapply the [SCC](#scc) to ensure there hasn't been any changes. !!! important "Good to know" Deleting the `HPECSIDriver` instance and uninstalling the CSI Operator does not affect any running workloads, `PersistentVolumeClaims`, `StorageClasses` or other API resources created by the CSI Operator. In-flight operations and new requests will be retried once the new `HPECSIDriver` has been instantiated. @@ -92,7 +93,7 @@ oc new-project hpe-storage --display-name="HPE CSI Driver for Kubernetes" !!! important The rest of this implementation guide assumes the default "hpe-storage" `Namespace`. If a different `Namespace` is desired. Update the `ServiceAccount` `Namespace` in the SCC below. -Deploy or [download]({{ config.site_url}}partners/redhat_openshift/examples/scc/hpe-csi-scc.yaml) the SCC: +
Deploy or [download]({{ config.site_url}}partners/redhat_openshift/examples/scc/hpe-csi-scc.yaml) the SCC: ```text oc apply -f {{ config.site_url}}partners/redhat_openshift/examples/scc/hpe-csi-scc.yaml @@ -300,6 +301,52 @@ oc get pvc -n openshift-virtualization-os-images -w !!! hint These steps might be removed in a future release in the event access mode transformation become a supported feature of the CSI driver. +# Live VM migrations for Alletra Storage MP + +With HPE CSI Operator for Kubernetes v2.4.2 and older there's an issue that prevents live migrations of VMs that has `PVCs` attached that has been clones from an OS image residing on Alletra Storage MP backends including 3PAR, Primera and Alletra 9000. + +Identify the `PVC` that that has been cloned from an OS image. The VM name is "centos7-silver-bedbug-14" in this case. + +```text +oc get vm/centos7-silver-bedbug-14 -o jsonpath='{.spec.template.spec.volumes}' | jq +``` + +In this instance, the `dataVolume` is the same name as the VM. Grab the `PV` name from the `PVC` name. + +```text +MY_PV_NAME=$(oc get pvc/centos7-silver-bedbug-14 -o jsonpath='{.spec.volumeName}') +``` + +Next, patch the `hpevolumeinfo` `CRD`. + +```text +oc patch hpevolumeinfo/${MY_PV_NAME} --type=merge --patch '{"spec": {"record": {"MultiInitiator": "true"}}}' +``` + +The VM is now ready to be migrated. + +!!! hint + If there are multiple `dataVolumes`, each one needs to be patched. + +# Unsupported Version of the Operator Install + +In the event on older version of the Operator needs to be installed, the bundle can be installed directly by [installing the Operator SDK](https://sdk.operatorframework.io/docs/installation/). Make sure a recent version of the `operator-sdk` binary is available and that no HPE CSI Driver is currently installed on the cluster. + +Install a specific version (v2.4.2 in this case): + +```text +operator-sdk run bundle --timeout 5m -n hpe-storage quay.io/hpestorage/csi-driver-operator-bundle:v2.4.2 +``` + +!!! important + Once the Operator is installed, a `HPECSIDriver` instance needs to be created. Follow the steps using the [web console](#openshift_web_console) or the [CLI](#openshift_cli) to create an instance. + +When the unsupported install isn't needed any longer, run: + +```text +operator-sdk cleanup -n hpe-storage hpe-csi-operator +``` + # Unsupported Helm Chart Install In the event Red Hat releases a new version of OpenShift between HPE CSI Driver releases or if interest arises to run the HPE CSI Driver on an uncertified version of OpenShift, it's possible to install the CSI driver using the Helm chart instead.