Skip to content

Latest commit

 

History

History
72 lines (55 loc) · 3.09 KB

Backups.md

File metadata and controls

72 lines (55 loc) · 3.09 KB
Raw

Backups

Backups are done via Velero.

Currently, my backups are done using the Velero helm chart. A Velero Schedule Custom resource is created for every day at 5 in the morning that backups a bunch of my services.

Manual steps and General Knowledge :

Installation:

  1. Create a bucket in AWS S3 ( Mine is called sgenov-backup )
  2. Create a IAM user with the correct policy for S3
  3. Get the IAM user's credentials
  4. Create a new file: velero-credentials in the root of the folder ( see how it should look down below )
  5. In the project root, run: velero install --use-restic --provider aws --plugins velero/velero-plugin-for-aws --bucket sgenov-velero-backup --secret-file ./velero-credentials --backup-location-config region=eu-west-1 --snapshot-location-config region=eu-west-1
    1. Make sure the bucket is empty!
    2. --use-restic is used to back up PVCs

velero-credentials:

[default]
aws_access_key_id=<AWS_ACCESS_KEY_ID>
aws_secret_access_key=<AWS_SECRET_ACCESS_KEY>

Creating backups:

Manual:

velero backup create uptimekuma --include-namespaces uptimekuma --snapshot-volumes=true Will backup the namespace uptimekuma

From Schedule:

velero create backup --from-schedule general

Restoring backups:

When restoring backups Velero will not restore objects that are already existing, so whatever you need to restore must be deleted first. If you are restoring pods from a deployment/statefulset/daemonsets/etc make sure to delete them too.

NOTE: Velero will hang up if it can't restore all the restic volumes, so you may need to remove the restore or it will be stuck in In progress

Manual:

  1. Clear up namespace
  2. Run velero restore create uptimekuma1 --from-backup uptimekuma --restore-volumes=true

From Schedule:

  1. Clear up any resources you wish to restore
  2. Run: velero restore create --from-schedule general --restore-volumes=true. Optionally add: --include-namespaces postgresql

OLD way of handling backup and restore:

Backups

Currently, I do my backups via node-red. Reason why I don't go for another solution is because I have not found a good FOSS to do what I want to do. I know that K10 is more than enough for me but even so I don't want to rely on proprietary software.

Restores

Restorations can be done via the restore playbooks however they are very opinionated towards my output from NodeRed.

  • ansible-playbook -i hosts/inventory playbooks/restore/main.yml --tags nodered
  • ansible-playbook -i hosts/inventory playbooks/restore/main.yml --tags changedetection
  • ansible-playbook -i hosts/inventory playbooks/restore/main.yml --tags nginxproxymanager
  • ansible-playbook -i hosts/inventory playbooks/restore/main.yml --tags vaultwarden
  • ansible-playbook -i hosts/inventory playbooks/restore/main.yml --tags simplesecrets
  • ansible-playbook -i hosts/inventory playbooks/restore/main.yml --tags postgresql
  • ansible-playbook -i hosts/inventory playbooks/restore/main.yml --tags trilium
  • ansible-playbook -i hosts/inventory playbooks/restore/main.yml --tags uptimekuma