Backup and Restore for external blobstores

Page last updated:

This topic explains how to back up and restore external blobstores with BOSH Backup and Restore (BBR).

BBR supports Amazon S3 buckets, S3-compatible storage solutions, Azure storage containers, and Google Cloud Storage buckets.

The necessary configuration and supported restore scenarios differ depending on the type of blobstore you use. For details, see the sections below.

S3-compatible unversioned blobstores

BBR backs up external blobstores by copying blobs from live buckets to specified backup buckets. BBR uses the native copy functionality of your blobstore to transfer blobs between the live and backup buckets, without transferring the blobs to your BBR instance.

For resiliency and safety, store your backup buckets and live buckets in different regions.

Enable backup and restore of your unversioned S3-compatible blobstore

To back up a Cloud Foundry deployment that uses an unversioned S3-compatible external blobstore, you must co-locate the s3-unversioned-blobstore-backup-restorer job from the backup-and-restore-sdk-release as part of your deployment.

  1. Verify that your blobstore is either an Amazon S3 bucket or an S3-compatible bucket with support for AWS Signature Version 4.

    Note If your S3-compatible blobstore uses a custom CA certificate, see Configuring trusted certificates in the BOSH documentation. BBR automatically makes use of your configured trusted certificates.

  2. Create backup buckets for droplets, packages, and buildpacks. Cloud Foundry recommends that either the backup buckets or copies of them be in a different region than the live buckets.

  3. Include the enable-backup-restore-s3-unversioned.yml ops file in your deployment. The enable-backup-restore-s3-unversioned.yml file is in the cf-deployment GitHub repository.

    See vars-enable-backup-restore-s3-unversioned.yml in GitHub for information about how to configure the variables for your backup bucket locations.

    Note Apply enable-backup-restore.yml and use-s3-blobstore.yml before enable-backup-restore-s3-unversioned.yml. See Apply ops files in the correct order.


    If you do not use cf-deployment and ops files, you can still back up and restore external blobstores with BBR. See the contents of the enable-backup-restore-s3-unversioned.yml ops file as an example and review the backup-and-restore-sdk-release documentation in GitHub for further information.

Caution To minimize storage costs, Cloud Foundry recommends creating a lifecycle policy for your backup buckets that permanently expires your older backups. For more information, see the AWS documentation.

Supported restore scenarios for your unversioned S3-compatible blobstore

The restore process copies blobs from a directory in the backup bucket to the live buckets in use by the CF deployment you are restoring into.

If your original backup buckets are lost but you are restoring from copies of those original backup buckets, ensure that the variables listed in vars-enable-backup-restore-s3-unversioned.yml reference those copies.

Caution Restoring to an unversioned bucket overwrites blobs with the backed up copies of those blobs. BBR only modifies the blobs that were copied to the backup bucket. Other blobs are not changed during this process.

S3-compatible versioned blobstores

BBR supports S3-compatible buckets that are versioned and support AWS Signature Version 4. For more details about enabling versioning on your blobstore, see Enable Versioning on your S3-compatible blobstore below.

External blobstores are backed up by storing the current version of each blob, not the actual files. Those versions are set to be the current versions at restore time.

Caution Storing the current version of each blob makes backing up and restoring faster. However, this means that you can only restore if the original bucket still exists. If the original bucket is deleted, all of that bucket’s related versions are also deleted. If the original bucket is deleted, you can only restore from a replica. For more information, see Restore from replicas.

Caution The s3-versioned-blobstore-backup-restorer in backup-and-restore-sdk v1.5.1 and earlier does not support the backup and restore of an Amazon S3 blobstore with individual blobs greater than 5 GB.

Enable backup and restore of your versioned S3-compatible blobstore

To back up a Cloud Foundry deployment that uses an external blobstore, you must co-locate the s3-versioned-blobstore-backup-restorer job from backup-and-restore-sdk-release as part of your deployment.

  1. Verify that your blobstore is either an Amazon S3 bucket with versioning enabled or an S3-compatible bucket with versioning enabled and support for AWS Signature Version 4.

    Note If your S3-compatible blobstore uses a custom CA certificate, see Configuring trusted certificates in the BOSH documentation. BBR automatically makes use of your configured trusted certificates.

  2. Include the enable-backup-restore-s3-versioned.yml ops file in your deployment. The enable-backup-restore-s3-unversioned.yml file is in the cf-deployment GitHub repository.

    Note Apply enable-backup-restore.yml and use-s3-blobstore.yml before enable-backup-restore-s3-versioned.yml. See Apply ops files in the correct order.

    If you do not use cf-deployment and ops files, you can still back up and restore external blobstores with BBR. See the contents of the enable-backup-restore-s3-versioned.yml ops file as an example and see the S3-compatible Versioned Blobstores in the Blobstore Backup and Restore documentation in GitHub.

Enable versioning on your S3-compatible blobstore

BBR only supports the backup and restore of blobstores stored in versioned Amazon S3 buckets and in S3-compatible buckets that are versioned and support AWS Signature Version 4.

Three Cloud Foundry buckets are backed up by BBR, so you only need to enable versioning of:

  • droplets
  • packages
  • buildpacks

Enabling versioning of the resource_pool bucket is not required.

To enable versioning of Amazon S3 buckets, see How Do I Enable or Suspend Versioning for an S3 Bucket? in the Amazon Simple Storage Service documentation.

If you prefer to use the AWS CLI, use the put-bucket-versioning command. For more information, see put-bucket-versioning in the AWS CLI Command Reference.

  1. If your blobstore buckets are not empty, run the following command on each one of them, using the AWS CLI:

    aws s3 cp s3://BUCKET-NAME/ s3://BUCKET-NAME/ --recursive --metadata bump=true
    

    Where BUCKET-NAME is the name of your bucket.

    For example:

    $ aws s3 cp s3://my-bucket/ s3://my-bucket/ --recursive --metadata bump=true
    

    This ensures that each file in your buckets has a valid version ID.

  2. After enabling versioning, you have a current version and zero or more non-current versions for each object.

Note Cloud Foundry recommends setting an expiration policy on your buckets to delete old non-current versions and minimize storage costs. For more information, see the AWS documentation in the Amazon Simple Storage Service documentation.

Note If your blobstore uses S3-compatible buckets that are not from Amazon, see the documentation for your storage provider regarding enabling versioning and setting an expiration policy.

Enable replication on your versioned S3-compatible blobstore

BBR does not download your blobs to the backup artifact when performing a backup. Instead, BBR notes the current version identifier of each blob and stores the identifiers in the artifact.

When restoring, BBR reverts your blobs to the original versions using the version identifiers. Because the backup artifact contains only identifiers, not blobs, BBR can only restore the blobs if the original bucket containing the blob versions still exists.

If a bucket is deleted, all of that bucket’s versions are also deleted and you cannot restore using that bucket. To prevent this from happening, set up Cross-Region Replication. See the put-bucket-replication command if you prefer to use the AWS CLI.

If your original buckets are lost, you can restore from a replica. Replication results in buckets that are identical to the original buckets, including the original version identifiers. To restore from a replica, see Restore from Replicas below.

Supported restore scenarios for your versioned S3-compatible blobstore

Determine which of the following scenarios applies to your deployment before restoring:

Note When restoring to a bucket, BBR only modifies blobs that are recorded in the backup artifact. Other blobs are not affected.

In-place sestore

When backing up an external blobstore, the backup consists of a snapshot of the objects’ IDs and versions. If you are doing an in-place restore and your destination Cloud Foundry for that restore uses the original buckets of the backed-up Cloud Foundry, those versions are retrieved and set to be the current versions in the buckets.

Restore to new buckets

If your destination Cloud Foundry for a restore uses different buckets, then you can also restore into those new buckets, if the original buckets still exist. During restore, the original versions are copied from the original buckets to the destination buckets.

In order to restore to new buckets, those new buckets must be versioned. For more information about versioning an Amazon S3 blobstore, see Enable versioning on your S3-compatible blobstore above.

Before restoring, ensure that the s3-versioned-blobstore-backup-restorer job in the backup-and-restore-sdk is configured to point to the destination buckets.

Restore from replicas

To protect yourself from losing your blobstore buckets, you should enable replication, as described below. This allows you to perform restores from the replicas, in case your original buckets are lost.

To restore from replicas, you must modify the backup artifacts to point to the replicas before beginning a restore.

To modify the backup artifacts:

  1. In your terminal, change into the directory for your backup artifact.

  2. Extract the blobstore backup archive by running:

    tar xvf BLOBSTORE-ARTIFACT.tar
    

    Where BLOBSTORE-ARTIFACT is the name of your blobstore artifact.

    For example:

    $ tar xvf backup-restore-0-s3-versioned-blobstore-backup-restorer.tar

  3. Modify the blobstore.json to point to the replicated buckets.

  4. Recalculate the shasum of blobstore.json by running:

    shasum -a 256 blobstore.json
    
  5. Update the metadata file entry for that blobstore.json with the new checksum.

  6. Re-create the archive by running:

    tar cvf BLOBSTORE-ARTIFACT.tar ./blobstore.json
    

    Where BLOBSTORE-ARTIFACT is the name of your blobstore artifact.

    For example:

    $ tar cvf backup-restore-0-s3-versioned-blobstore-backup-restorer.tar ./blobstore.json

The backup artifact is now ready to be restored from the replicated buckets.

Azure blobstores

BBR backs up Azure storage containers by storing the ETags of each blob, not the actual blobs, which makes backups and restores faster. However, this means that restores work only if the original containers still exist.

Enable soft delete in your Azure storage account

BBR requires that you enable soft delete in your Azure storage account. With soft delete, you can recover your data when blobs or blob snapshots are deleted. To enable soft delete in your Azure storage account, follow the instructions in the Azure documentation. You should set a reasonable retention policy to minimize storage costs.

Enable replication for your Azure blobstore

BBR does not download your blobs to the backup artifact when it performs a backup. Instead, BBR records the current ETag of each blob and stores the identifiers in the artifact.

When restoring, BBR reverts your blobs to the original versions using the ETags. Because the backup artifact stores identifiers, not blobs, BBR can restore the blobs only if the original container with the blob versions still exists.

An Azure data center failure can render your original container unavailable. To mitigate this threat, configure replication for your container. For more information, see Azure Storage replication in the Azure documentation.

Enable backup and restore of your Azure blobstore

To back up a Cloud Foundry deployment that uses an Azure external blobstore, you must include the enable-backup-restore-azure.yml ops file in your deployment. This co-locates the azure-blobstore-backup-restorer job from backup-and-restore-sdk-release as part of your deployment.

Note Apply enable-backup-restore.yml and use-azure-storage-blobstore.yml before enable-backup-restore-azure.yml. For more information, see Apply ops files in the correct order.

If you do not use cf-deployment and ops files, you can still back up and restore external blobstores with BBR. See the contents of the enable-backup-restore-azure.yml ops file as an example and see the backup-and-restore-sdk-release documentation in GitHub for details.

Supported restore scenarios for your Azure blobstore

Determine which of the following scenarios applies to your deployment before restoring it:

Note When restoring to a container, BBR modifies only blobs that are recorded in the backup artifact. Other blobs are not affected.

In-place restore

When BBR backs up an Azure blobstore, the backup consists of a snapshot with the IDs and ETags of the objects.

In an in-place restore, you reuse the same containers. BBR replaces the current blob versions in the container with those corresponding to the ETags recorded in the backup artifact.

Restore to a new container in the same storage account

When you restore to different containers, BBR copies the versions recorded in the backup artifact from the backed-up containers to the new containers. In this scenario, the backed-up containers must exist during the restore.

Restore to a new container in a different storage account

You can restore blobs to a container in a different Azure storage account from the original. In this scenario, BBR copies the blobs from the backed-up containers in the source Azure storage account to the new containers in the storage account used for restore. You must use the enable-restore-azure-clone.yml ops file in your deployment.

Note Apply enable-backup-restore.yml and use-azure-storage-blobstore.yml before enable-restore-azure-clone.yml. For more information, see Apply ops files in the correct order in Configuring Cloud Foundry for BOSH Backup and Restore.

After restoring, replace the enable-restore-azure-clone.yml ops file in your deployment with the enable-backup-restore-azure.yml ops file and redeploy.

Google Cloud Storage (GCS) blobstores

BBR backs up GCS blobstores by copying blobs from live buckets to specified backup buckets. BBR uses the native copy functionality of your blobstore to transfer blobs between the live and backup buckets, without transferring the blobs to your BBR instance.

Enable backup and restore of your GCS blobstore

To back up a Cloud Foundry deployment that uses a GCS external blobstore, you must co-locate the gcs-blobstore-backup-restorer job from the backup-and-restore-sdk-release as part of your deployment.

  1. Create backup buckets for droplets, packages, and buildpacks. Cloud Foundry recommends that either the backup buckets or copies of them be in a different region than the live buckets.

  2. Include the enable-backup-restore-gcs.yml ops file in your deployment. The enable-backup-restore-gcs.yml file is in the cf-deployment GitHub repository.

    See vars-enable-backup-restore-gcs.yml for information about how to configure the variables for your backup bucket locations.

    Note Apply enable-backup-restore.yml and use-gcs-blobstore-service-account.yml before enable-backup-restore-gcs.yml. See Apply ops files in the correct order in Configuring Cloud Foundry for BOSH Backup and Restore.


If you do not use cf-deployment and ops files, you can still back up and restore external blobstores with BBR. See the contents of the enable-backup-restore-gcs.yml ops file as an example and review the backup-and-restore-sdk-release documentation in GitHub for further information.

Caution To minimize storage costs, Cloud Foundry recommends creating a lifecycle policy for your backup buckets that permanently expires your older backups. For more information, see the GCP documentation.

Supported restore scenarios for your GCS blobstore

The restore process copies blobs from a directory in the backup bucket to the live buckets in use by the CF deployment you are restoring into.

If your original backup buckets are lost but you are restoring from copies of those original backup buckets, ensure that the variables listed in vars-enable-backup-restore-gcs.yml reference those copies.

Caution Restoring to a GCS bucket overwrites blobs with the backed up copies of those blobs. BBR only modifies the blobs that were copied to the backup bucket. Other blobs are not changed during this process.

View the source for this page in GitHub