IBM Support

IBM Storage Fusion 2.5.2 Hot Fix

Download


Downloadable File

File link File size File description

Abstract

This document provides the install instructions for hotfix 2.5.2 and commands to resolve the backup failure of the file-api-claim PVC.

This hotfix is applicable when you use IBM Storage Fusion 2.5.2 Backup & Restore (Legacy) service to take IBM Cloud Pak for Data backups and any backup of CEPHS PVC with more than 1000 files.

Prerequisites

  • This hotfix requires OpenShift Container Platform v4.12 and OpenShift Data Foundation v4.12.

Installation Instructions

  • Follow the instructions mentioned in Step 1 to download and copy the images to a local private registry for an air-gapped environment.
  • Follow the instructions mentioned in Step 2 to apply the patch using the images that are downloaded to the local private registry.
Procedure:
  1. Follow the steps to apply the patch in an air-gapped environment: 
    1. Log in to the IBM Entitled Container Registry using the IBM entitlement key.
      docker login cp.icr.io -u cp -p <your entitlement key>
    2. Set the following environment variables.
      export LOCAL_SECRET_JSON='<relative path to pull-secret.json>' 
      export LOCAL_ISF_REGISTRY="<Your enterprise registry host>:<port>" 
      export LOCAL_ISF_REPOSITORY="<Your image path>" 
      export TARGET_PATH="$LOCAL_ISF_REGISTRY/$LOCAL_ISF_REPOSITORY"
      
    3. Run the following command to log in to the Docker registry with your enterprise registry credentials.
      docker login $LOCAL_ISF_REGISTRY -u <your enterprise registry username> -p <your enterprise registry password>
      
      
      
    4. Run the following command to install Skopeo (if it is not already installed).
      yum install skopeo
      
    5. Use the following skopeo commands to copy the patch images from the IBM production registry to the local private registry.
      skopeo copy --insecure-policy --all docker://cp.icr.io/cp/sppc/baas-spp-agent@sha256: 1b82996d9e0abf0dc7efa4395a0f246d4712af8a756acffcf057a738d54e73c2 docker://$TARGET_PATH/sppc/baas-spp-agent@sha256: 1b82996d9e0abf0dc7efa4395a0f246d4712af8a756acffcf057a738d54e73c2
      
      skopeo copy --insecure-policy --all docker://cp.icr.io/cp/sppc/baas-transaction-manager-redis@sha256: fb78128466fba7995040900358931f1281a09e151928c8e22cd30ade9ee389d3 docker://$TARGET_PATH/sppc/baas-transaction-manager-redis@sha256: fb78128466fba7995040900358931f1281a09e151928c8e22cd30ade9ee389d3
      
      skopeo copy --insecure-policy --all docker://cp.icr.io/cp/sppc/baas-transaction-manager-worker@sha256: dd662390762aba4f2ef56b56822a182519c28c703736e8bf59b2aca8919c84ba docker://$TARGET_PATH/sppc/baas-transaction-manager-worker@sha256: dd662390762aba4f2ef56b56822a182519c28c703736e8bf59b2aca8919c84ba
      
      skopeo copy --insecure-policy --all docker://cp.icr.io/cp/sppc/baas-transaction-manager@sha256: 2caafc0fb0c40c05555a3cab11818fd7b8a0c68918b0f4287b091157d36aba54 docker://$TARGET_PATH/sppc/baas-transaction-manager@sha256: 2caafc0fb0c40c05555a3cab11818fd7b8a0c68918b0f4287b091157d36aba54
      
      skopeo copy --insecure-policy --all docker://cp.icr.io/cp/sppc/baas-datamover@sha256: ce4a4d80624b0c4bf7e58031fe056624db728f3450af72bca8a0320f2984fd85 docker://$TARGET_PATH/sppc/baas-datamover@sha256: ce4a4d80624b0c4bf7e58031fe056624db728f3450af72bca8a0320f2984fd85
      
      
      
  2. Follow the steps to install the hotfix using the images downloaded to the local private registry:
     
    1. Run the following command to scale down operator ibmsppc-operator-controller-manager in baas namespace.
      oc scale --replicas=0 deployment/ibmsppc-operator-controller-manager -n baas
       
    2. Run the following command to update spp-agent cluster role.
      oc edit clusterrole baas-spp-agent
      
      Search for storageclasses section and add ‘create’ role.
      - apiGroups:
       - storage.k8s.io
       resources:
       - storageclasses
       verbs:
       - get
       - list
       - create
      
    3. Open baas-spp-agent deployment in baas namespace in the OpenShift Container Platform web console and save a copy of the current YAML so that you can revert the hotfix if required.
       
    4. Update the digest for the baas-spp-agent with the digest from hotfix.
      image: >-
                  cp.icr.io/cp/sppc/baas-spp-agent@sha256: 1b82996d9e0abf0dc7efa4395a0f246d4712af8a756acffcf057a738d54e73c2
      
      
      
    5. Open baas-transaction-manager deployment in baas namespace in the OpenShift Container Platform web console and save a copy of the current YAML so that you can revert the hotfix if required.
       
    6. Update the digest for the baas-transaction-manager pods with the digest from hotfix.
      image: >-
                  cp.icr.io/cp/sppc/baas-transaction-manager-redis@sha256: fb78128466fba7995040900358931f1281a09e151928c8e22cd30ade9ee389d3
               
       image: >-
                  cp.icr.io/cp/sppc/baas-transaction-manager-worker@sha256: dd662390762aba4f2ef56b56822a182519c28c703736e8bf59b2aca8919c84ba
      
                image: >-
                  cp.icr.io/cp/sppc/baas-transaction-manager@sha256: 2caafc0fb0c40c05555a3cab11818fd7b8a0c68918b0f4287b091157d36aba54
      
      
      
    7. Open config-map baas-configmap from baas namespace in the OpenShift Container Platform web console and save a copy of the current YAML so that you can revert the hotfix if required.
       
    8. Update the data-mover image with the digest from hotfix.
      datamover-image: >-
          cp.icr.io/cp/sppc/baas-datamover@sha256: ce4a4d80624b0c4bf7e58031fe056624db728f3450af72bca8a0320f2984fd85
       
      Update the following entry in baas-configmap to update the memory allocated for restic.
      testflags: '{"use_jetstack": True, "dm-memory": "50000Mi", "dm-ephemeral-storage": "6000Mi" }'
      
      During the restore, if you need to use a local recipe (instead of the recipe from backup), then the following flag can be added to the test flags.
      "useLocalRecipe": True
      
      
    9. Wait for baas-spp-agent pod and transaction-manager pods to restart.
       
    10. Run the following command to increase the memory limit of the cephfs mds pods to 32GiB.
      oc edit storagecluster -n openshift-storage <storagecluster_name> 

      Add the following specification details to the storage class cluster CR:
      spec:
        resources:
          mds:
            limits:
              memory: 32Gi
To revert the hotfix changes:
If you want to revert the hotfix, then change the digests back to the original values from the saved versions of YAML from #step3 to #step8. Also, revert the changes made under #step10.

Off

Document Location

Worldwide

[{"Type":"MASTER","Line of Business":{"code":"LOB26","label":"Storage"},"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SSG4YK","label":"IBM Spectrum Fusion"},"ARM Category":[{"code":"a8m3p0000000rXCAAY","label":"SW"}],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"All Versions"}]

Document Information

Modified date:
15 November 2023

UID

ibm17060466