Typical backup errors and how to avoid them. Part two
In the last article, we began a review of typical errors when working with backups. Continuing this topic, we propose to dwell on other problems of backup and consider ways to solve them. Some errors arise from old habits, while others are due to a misunderstanding of the very concept of backup. We continue:
6. Replacing backup archiving or replication
These procedures are somewhat similar, but have a fundamentally different purpose. Backup is used to guarantee the recovery of data lost for any reason in the production system (the system in which they directly work with files). Archiving is used to free up space in the workload by compressing data that is no longer relevant and transferred to slower media with a low cost of storing files. Replication is the constant duplication of all data from a product to a parallel system (local or cloud platform) in synchronization mode. It is needed to increase resiliency, but only reduces the severity of the consequences of hardware failures.
')
In the event of a virus attack (in particular, actions of a cryptographer Trojan) or accidental overwriting of data, they will not be found anywhere except for backup. Synchronous replication will lead to the simultaneous deletion or damage of files in the production and replica, and in the archives there will always be only old files that were not planned to be actively used in the near future.
7. Storing backup and raw data together
Writing backups to the same disk from which the source data is being read is already blocked by most backup programs. However, there is a less obvious mistake: storing backups and products on the same physical device. Even if they are written to a separate hard drive, there is still a high risk of losing the original data and its backup copies immediately. For example, a burned-out power supply sometimes burns out the controllers of all disks connected to it. Slightly better storage on an external disk or NAS located in the same room. In case of theft, fire, flooding or other force majeure circumstances, it is likely that both the production and all backup copies will be destroyed.
8. Using only one backup
According to rule “3-2-1”, all valuable data must have at least three backup copies recorded on two types of media, and one of the backups is additionally stored remotely (for example, in the cloud). This allows you to guarantee data recovery under any circumstances, including the physical destruction of the server or the removal of equipment during the "mask show". As long as you have only one backup, you can assume that it does not exist at all. For example, Acronis Backup (Advanced) has the ability to automatically duplicate a backup, saving it in parallel to another media or loading it into a proprietary cloud.
9. Exclusion from backup of files stored in the cloud
Cloud file storage services are growing in popularity. Many believe that there is no point in backing up data already duplicated in the cloud. However, cloud failures also occur. Once a mistake in updating the Yandex.Disk application led not only to data loss, but also to the removal of system files — computers stopped loading. Google Docs sometimes saves new documents to nowhere, Dropbox is sometimes unavailable, and Box always slows down. Therefore, the cloud is only an additional place to store data, from where they should also be regularly placed in their own backups.
10. Creating a backup during the operating system update
During the update of system files, access to them is suspended. After applying some updates may require a reboot. Therefore, the process of updating the OS can disrupt the creation of a backup and cause its damage, especially if it is a copy of the system partition or the actively used database. The problem is further complicated by the fact that in Windows 10 updates are applied automatically. To avoid failures, it is recommended to use deferred updates, as well as measure the typical backup creation time and schedule their creation so that they do not overlap with the OS update procedure. A more modern approach is based on the use of shadow copy technologies. While the OS is being updated, a backup is created from the shadow blocks of the file system. In Windows, you can use Microsoft's Volume Shadow Copy (VSS) for this. In Acronis Backup (Advanced), if it fails or stops, you can additionally use the SnapAPI module, which is responsible for all I / O operations on the hard disk.
Findings:
1) Backup schemes should guarantee a low RPO time. The forced stopping of IT services for several hours is critical, and for several days is fatal for modern business with high transaction rates (banks, retail, telecom operators, service providers and many others).
2) Creating backups manually can not be considered a reliable method and does not meet the current needs of companies, even small and not having a developed IT infrastructure.
3) Archiving, replication and backup - conceptually different technologies for improving fault tolerance. They can be used together, but they do not replace each other.
4) Modern backup software allows you to automate all procedures: creating backups on a schedule, checking them, deleting the oldest copies to clear the place and sending notifications about each stage.
5) Acronis solutions make it easy to implement the “3-2-1” rule by automatically duplicating backups to additional storage devices and to your own cloud.
Useful links:
» Webinars on backup and data protection
» Demo version for backup of corporate servers Acronis Backup Advanced
» Comparing Acronis and Veeam
Do not forget to backup in time, Friends!
6. Replacing backup archiving or replication
These procedures are somewhat similar, but have a fundamentally different purpose. Backup is used to guarantee the recovery of data lost for any reason in the production system (the system in which they directly work with files). Archiving is used to free up space in the workload by compressing data that is no longer relevant and transferred to slower media with a low cost of storing files. Replication is the constant duplication of all data from a product to a parallel system (local or cloud platform) in synchronization mode. It is needed to increase resiliency, but only reduces the severity of the consequences of hardware failures.
')
In the event of a virus attack (in particular, actions of a cryptographer Trojan) or accidental overwriting of data, they will not be found anywhere except for backup. Synchronous replication will lead to the simultaneous deletion or damage of files in the production and replica, and in the archives there will always be only old files that were not planned to be actively used in the near future.
7. Storing backup and raw data together
Writing backups to the same disk from which the source data is being read is already blocked by most backup programs. However, there is a less obvious mistake: storing backups and products on the same physical device. Even if they are written to a separate hard drive, there is still a high risk of losing the original data and its backup copies immediately. For example, a burned-out power supply sometimes burns out the controllers of all disks connected to it. Slightly better storage on an external disk or NAS located in the same room. In case of theft, fire, flooding or other force majeure circumstances, it is likely that both the production and all backup copies will be destroyed.
8. Using only one backup
According to rule “3-2-1”, all valuable data must have at least three backup copies recorded on two types of media, and one of the backups is additionally stored remotely (for example, in the cloud). This allows you to guarantee data recovery under any circumstances, including the physical destruction of the server or the removal of equipment during the "mask show". As long as you have only one backup, you can assume that it does not exist at all. For example, Acronis Backup (Advanced) has the ability to automatically duplicate a backup, saving it in parallel to another media or loading it into a proprietary cloud.
9. Exclusion from backup of files stored in the cloud
Cloud file storage services are growing in popularity. Many believe that there is no point in backing up data already duplicated in the cloud. However, cloud failures also occur. Once a mistake in updating the Yandex.Disk application led not only to data loss, but also to the removal of system files — computers stopped loading. Google Docs sometimes saves new documents to nowhere, Dropbox is sometimes unavailable, and Box always slows down. Therefore, the cloud is only an additional place to store data, from where they should also be regularly placed in their own backups.
10. Creating a backup during the operating system update
During the update of system files, access to them is suspended. After applying some updates may require a reboot. Therefore, the process of updating the OS can disrupt the creation of a backup and cause its damage, especially if it is a copy of the system partition or the actively used database. The problem is further complicated by the fact that in Windows 10 updates are applied automatically. To avoid failures, it is recommended to use deferred updates, as well as measure the typical backup creation time and schedule their creation so that they do not overlap with the OS update procedure. A more modern approach is based on the use of shadow copy technologies. While the OS is being updated, a backup is created from the shadow blocks of the file system. In Windows, you can use Microsoft's Volume Shadow Copy (VSS) for this. In Acronis Backup (Advanced), if it fails or stops, you can additionally use the SnapAPI module, which is responsible for all I / O operations on the hard disk.
Findings:
1) Backup schemes should guarantee a low RPO time. The forced stopping of IT services for several hours is critical, and for several days is fatal for modern business with high transaction rates (banks, retail, telecom operators, service providers and many others).
2) Creating backups manually can not be considered a reliable method and does not meet the current needs of companies, even small and not having a developed IT infrastructure.
3) Archiving, replication and backup - conceptually different technologies for improving fault tolerance. They can be used together, but they do not replace each other.
4) Modern backup software allows you to automate all procedures: creating backups on a schedule, checking them, deleting the oldest copies to clear the place and sending notifications about each stage.
5) Acronis solutions make it easy to implement the “3-2-1” rule by automatically duplicating backups to additional storage devices and to your own cloud.
Useful links:
» Webinars on backup and data protection
» Demo version for backup of corporate servers Acronis Backup Advanced
» Comparing Acronis and Veeam
Do not forget to backup in time, Friends!
Source: https://habr.com/ru/post/316596/
All Articles