Skip to content
This repository has been archived by the owner on May 4, 2024. It is now read-only.

Latest commit

 

History

History
63 lines (58 loc) · 2.73 KB

test2procedure.md

File metadata and controls

63 lines (58 loc) · 2.73 KB
Raw

Leave ibm-common-services and cpd-operators namespaces Recover cpd-instance namespace from VRG(s) and hooks

how to use the hooks: https://ibm.ent.box.com/notes/1001273872485?s=u4u190uwzzhs943t2frtpzkyyc6uo7wy install spp: https://www.ibm.com/docs/en/spp/10.1.7?topic=platform-installing-operator-in-online-environment

backup: recipe:

 workflows:
  - name: backup
    sequence:
    - hook: br-service-hooks/checkpoint
    - hook: br-service-hooks/pre-backup
    - group: cpd-instance-volumes
    - hook: br-service-hooks/post-backup
    - group: cpd-instance-resources

procedure:

1. run hook br-service-hooks/checkpoint (checkpoint.sh)
2. run hook br-service-hooks/pre-backup (pre-backup.sh)
3. run hook br-service-hooks/post-backup (post-backup.sh)
4. create [vrg-capture](vrg-capture.yaml) in cpd-instance ns as primary and wait for clusterdataprotected
5. check minio content: vrg, persistent volume, kubeobjects
6. set [vrg-capture](vrg-capture.yaml) as secondary
7. delete cpd-instance ns following instructions here: (delete-cpd-instance.sh)
https://www.ibm.com/docs/en/cloud-paks/cp-data/4.0?topic=bured-scenario-backing-up-restoring-instance-cloud-pak-data-same-cluster

vrg backup: https://github.com/RamenDR/cpd-testing/tree/main/cpd-instance/ci

restore: recipe:

 - name: restore
  sequence:
  - group: cpd-instance-volumes
  - group: cpd-instance-pre-workload-resources
  - hook: br-service-hooks/pre-workload
  - group: cpd-instance-workload-resources
  - hook: br-service-hooks/post-workload
  - group: cpd-instance-operator-resources

procedure:

1. create cpd-instance ns
2. modify ns to the backuped ns uid: 1000740000 ($kubectl edit ns cpd-instance)
(3. re-restore: kubectl annotate pv $i volumereplicationgroups.ramendr.openshift.io/ramen-restore- )
3. create [vrg-restore-1](vrg-restore-1.yaml) in cpd-instance ns as primary and wait for clusterdataready
4. check if pods are coming back in cpd-instance ns
4.5. https://www.ibm.com/docs/en/cloud-paks/cp-data/4.0?topic=dbs-worker-pods-are-not-running-after-you-do-backup-restore-oadp-backup-restore-utility follow this solution to restore wkc
5. run hook br-service-hooks/post-workload (post-workload.sh)
6. delete [vrg-restore-1](vrg-restore-1.yaml) and create [vrg-restore-2](vrg-restore-2.yaml) in cpd-instance ns as primary and wait for clusterdataready
7. check if cpd-instance ns resources are restored: 
(cpd-cli can be installed here: https://github.com/IBM/cpd-cli)
cpd-cli manage get-cr-status --cpd_instance_ns=cpd-instance
8. check web console of cp4d: 
cpd-cli manage get-cpd-instance-details --cpd_instance_ns=cpd-instance --get_admin_initial_credentials=true