I am searching for a dependable backup solution for my Linux server, which currently uses BackupRun and R1soft software. However, I experienced a problem when I attempted to restore a client's database and was unsuccessful. Although I had another local backup, my confidence in BackupRun has been diminished, and I am now uncertain about R1soft. My server contains about 1.2TB of data, and I require the ability to restore individual files and databases.
While a cloud service may provide a bare-metal restore option, it could take an extended period of time. I am also interested in a total server backup on a daily basis that enables the restoration of a single file. Although I have found options such as iDrive through searches, larger and more affordable backup alternatives do not appear to support Linux servers. Considering this, I have contemplated obtaining another low-cost dedicated server and creating a mirror. Would this be a practical choice?
For nearly a decade, I have used Duplicity software to conduct Ubuntu-based VPS's "incremental forever" backups, utilizing specific parameters such as --timeout 120, --num-retries 12, --backend-retry-delay 90, --allow-source-mismatch, and --no-encryption.
While relying on B2 for cloud object storage, I prefer to use S3 as an interface, as it is more dependable than the B2 interface with newer Duplicity versions. The only adjustment I've made was performing a mysqldump to simplify database recovery, prior to conducting night-time backups through Duplicity.
Based on my experience, I would suggest using an application such as Jetbackup as the ideal solution. Begin by selecting a high-capacity bare-metal server, and then install Jetbackup on your main server to determine how you would like data to be backed up.
That being said, it is unfair to solely blame R1soft for any issues. R1soft remains a reliable application, and if properly installed, no problems should arise.
The programs that you use are popular, then reliable. I don't believe that R1soft is somehow not reliable. Check again if you have installed the necessary files correctly. If something is still wrong, contact the software support service.
I can recommend CrashPlan. It has many backup options. There are such options: copies in a folder, copies in another network computer, you can also in a friend's computer or on your CrashPlan backup server. The main thing is to find a friend with a lot of free hard disk space, and you will be able to get your files absolutely free of charge right from home.
To create backups on a remote server, copy the backup made on the local server and transfer it to the remote webserver. For QA servers, backup production and restore it to the QA server. To ensure adequate backup storage space, consider using a dedicated server. If a dedicated backup server is utilized, backups can be initiated remotely without copying the backup to the local server.
To back up a database, make a backup on a local server and transfer it to a remote server, resulting in two separate locations should one backup be lost. Should a database fail on a local server, restoring from a backup on the same server may prove more convenient than copying the backup from a remote webserver to a local server. Utilize specific tools such as mysqldump and pg_dump for database backups.
When lacking a dedicated backup server to keep redundant copies of backups, consider implementing tape backups and store backups on tape. Backups should be made on local servers before being copied to tape during non-business hours or on weekends. While tape backup is slow, the benefit is that backups are portable and can be moved to any location.
Additionally, it is crucial to keep backups off-site to avoid complete loss of data in the event of a disaster. Utilize a backup data center where mission-critical applications are synchronized with an emergency data center, or at least store a copy of the backup tape in a remote location. Avoid physically changing tapes and storing them in the same datacenter, which is not useful during a disaster recovery scenario.
Good to work on this kind of backup server.
Software Development Company (https://www.techmahajan.com/)
Let me address your current setup with BackupRun and R1soft. While it's good that you had another local backup to fall back on, it's clear that you've lost confidence in BackupRun. R1soft, on the other hand, is a decent solution, but it's always good to have multiple options.
Now, regarding your requirements, you need a solution that can handle 1.2TB of data, provide daily full server backups, and allow for individual file and database restoration. You're also considering a bare-metal restore option, which is a great feature to have.
Cloud services can be a good option, but as you mentioned, restoring 1.2TB of data can take a significant amount of time. Some popular cloud backup services like AWS S3, Google Cloud Storage, or Microsoft Azure Blob Storage do offer Linux support, but they might not be the most affordable options.
iDrive is a decent choice, but as you mentioned, it might not be the most cost-effective solution for larger data sets. Other options like Backblaze or CrashPlan might be worth exploring, but again, they might not support Linux servers or have limitations on data size.
Now, regarding your idea of getting another low-cost dedicated server and creating a mirror, it's an interesting approach. This would essentially give you a redundant server that you can use for backups, and in case of a disaster, you can simply switch over to the mirrored server.
This approach has its pros and cons. On the plus side, you'd have a redundant server that can be used for backups, and you'd have full control over the server and the data. However, this approach also means you'd need to manage another server, which can add to your administrative overhead.
Another thing to consider is that a mirrored server would require you to set up replication, which can be complex, especially if you have a large dataset. Additionally, you'd need to ensure that the mirrored server is identical to the primary server, which can be a challenge.
In conclusion, while getting another dedicated server and creating a mirror is a viable option, it might not be the most practical choice. Instead, I would recommend exploring other backup solutions that are designed specifically for Linux servers and can handle large datasets.
Some options you might want to consider are:
Amanda (Advanced Maryland Automatic Network Disk Archiver): a popular open-source backup solution that supports Linux and can handle large datasets.
Bacula: another open-source backup solution that supports Linux and has a large community of users.
Duplicati: a free, open-source backup solution that supports Linux and can handle large datasets.
Veeam: a commercial backup solution that supports Linux and has a range of features, including bare-metal restore.