Considerations and Limitations
This section lists considerations and known limitations for object storage repositories.
General Considerations and Limitations
Consider the following limitations for object storage repositories:
- Make sure to open required ports to communicate with object storage repositories in advance. Consider that a backup server and a gateway server must have internet access to verify that the certificates installed on object storage repositories are valid. For more details, see Ports.
- If you use default network security configuration for helper appliances, make sure that they are compliant with your internal security policies.
- You can add an object storage repository to a second backup server using credentials with the read-only access permissions that allows you to perform data recovery options. If you use credentials with full-access permissions, it will lead to unpredictable behavior and data loss.
Consider the following:
- Object storage gateway appliances that are used to store backup data in filer (SMB (CIFS)/NFS) or block device mode (iSCSI/FC/SAS) are not supported if the backup data is offloaded to object storage and is no longer stored directly on the appliance.
Such gateway appliances are only supported in the following cases:
- All of the backup data is stored on the appliance altogether (that is, all of the backup chains are stored on the appliance as a whole and not scattered across multiple devices) and only additional copies of the backup data are transported to object storage.
- These appliances emulate a tape system (VTL) as an access protocol for Veeam Backup & Replication.
- Data in object storage bucket/container must be managed solely by Veeam Backup & Replication, including retention and data management. Enabling lifecycle rules is not supported, and may result in backup and restore failures.
- Multi-factor authentication (MFA) is not supported for object storage repositories.
- If a backup chain contains backup files that are marked as corrupted by Health Check, then such corrupted files, as well as all subsequent files that go after the corrupted one are never offloaded. In such a scenario, offload is only possible starting from the full backup file that succeeds the backup chain with corrupted backups.
- For optimal processing, we recommend to set the default block size to 1MB in the storage settings of a backup job. Larger block size can lead to multiple times larger incremental backups, while smaller block sizes will create extra IO pressure on the object storage.
- Different object storage repositories mapped to the same cloud folder can be used for storing both the capacity tier backups and the unstructured data backups.
Consider the following:
- Within a scale-out backup repository, the mount server of a performance extent will act as a gateway server of the capacity extent if all of the following is true:
- You use SMB share/NFS share/deduplicating storage appliances as performance extents of your scale-out backup repository.
- You have chosen Automatic selection for the gateway server at the Specify Shared Folder Settings step of the New backup repository wizard.
- For the object storage that you use as the capacity extent, you have not selected to connect to object storage using a gateway server at the Account step of the New Object Repository wizard.
- The backup proxy that processes backup data must meet the following requirements:
- It must be an on-premises server as close as possible to a backup server.
- It must have access to the cloud storage that you use as an object storage repository.
- You cannot switch an object storage repository to Sealed Mode and to Maintenance Mode unless it is an extent of a scale-out backup repository.
- Veeam Cloud Connect service providers can not use Azure Data Box and AWS Snowball Edge storage as object storage repositories.
Considerations and Limitations for Direct Backup to Object Storage
- Reverse incremental backup method is not supported.
- Synthetic full backup method is not supported.
- Compact full backup file option is not supported.
- You cannot use direct backup object storage repositories for backups of applications running on Kubernetes persistent volumes created by Kasten K10 Plug-in for Veeam Backup & Replication.
- You cannot use direct backup object storage repositories to keep backups created with Veeam Plug-ins for Enterprise Applications.
- You cannot use Amazon S3 Glacier Storage, S3 compatible with Data Archiving and Azure Archive Storage as direct backup repositories.
- You cannot use direct backup object storage repositories to keep backups created with Veeam Agent for Oracle Solaris and Veeam Agent for IBM AIX.
Note that you can use direct backup object storage repositories to keep backups created with backup copy jobs for Veeam Agent for IBM AIX and Veeam Agent for Oracle Solaris.
- Make sure that you add an S3 compatible object storage device fully compatible with the AWS S3 operations and AWS S3 Signature Version 4 standard.
- For some S3 compatible object storage repositories we do not recommended to use more than one bucket for each scale-out backup repository. If you use multiple folders within one bucket for several scale-out backup repositories, it will slow down data processing since the metadata generated by S3 compatible is handled for each bucket. To check requirements for your object storage, contact your S3 compatible vendor.
- [For Amazon S3] Only the Standard, Standard-IA and One Zone-IA storage classes are supported. For more information about Amazon S3 storage classes, see AWS Documentation.
- S3 compatible storage systems with the eventual data consistency model are not supported.
- You cannot add as performance extents one Wasabi bucket as an S3 compatible Object Storage and another Wasabi bucket as a Wasabi Cloud Object Storage to the same type of tier (either performance tier or capacity tier) of a scale-out backup repository.
- To be able to work with S3 compatible with Data Archiving object storage, the SOSAPI functionality must be enabled. To enable this, contact your S3 compatible vendor.
Limitations for Microsoft Azure Object Storage
- [For Microsoft Azure Blob storage] Veeam Backup & Replication supports specific types of storage accounts and tiers. For more information, see Microsoft Azure Storage Accounts.
- [For Microsoft Azure Blob storage] Veeam Backup & Replication does not support Azure Germany since Microsoft closed Microsoft Cloud Deutschland.
- Veeam Backup & Replication supports the Versioning feature for Microsoft Azure object storage repositories for which immutability is enabled.
- [For Microsoft Azure Blob storage] Veeam Backup & Replication does not support soft delete for blobs.
- [For Microsoft Azure Archive storage] Microsoft Azure has certain limits (quotas) on maximum amount of resources used. The quotas depend on the type of proxies you have selected. If you exhaust a quota, you will be unable to use Microsoft Azure Archive storage. For more information about Microsoft Azure quotas, see Microsoft Docs.
- Veeam Backup & Replication performs operations only on a blob level. You cannot create Azure containers or storage accounts from the backup infrastructure.
- Veeam Backup & Replication does not support object-level immutability and default immutability policies assigned to Azure storage accounts. You must set immutability for an Azure container where backed-up objects will reside. For more information, see Microsoft Docs.
Limitations for Google Cloud Object Storage
Currently, Veeam Backup & Replication does not support the Object Versioning and Bucket Lock features for Google Cloud object storage.
Enabling either any or both of these features on the bucket may result in unpredictable system behavior and data loss, as well as in extra costs for storing objects that have been removed by the retention policy. For more information, see Object Versioning and Bucket Lock.
Limitations for IBM Cloud Object Storage
- For IBM Cloud Object Storage on-premise, Veeam Backup & Replication supports versions starting from 220.127.116.11.
- Veeam Backup & Replication is supported on all IBM Cloud Object Storage (COS) deployment models. This includes on-premise, public cloud, and hybrid models. For the IBM public cloud, the following storage classes are supported: Standard, Vault, Cold Vault and Smart Tier.
- Veeam Backup & Replication does not support Archive and Accelerated Archive storage classes in the IBM public cloud.
Limitations for Immutability
For more information, see the Immutability Considerations and Limitations section.