Issues with "Clone Backup to Backblaze" Job Performance and Scheduling Conflicts

[SysErr]

I am encountering a significant issue with a "Clone Backup to Backblaze" job that I've set up. The task is designed to transfer all my backups, totaling 1.4 TB, to Backblaze using a 275mbit/s internet connection. Based on my calculations, I anticipated the job would complete within 13-14 hours. However, after 43 hours and 30 minutes, the progress is only at 88.5%.

Initially, the transfer rate peaks between 200-300mbit/s, which aligns with my expectations. But, a few hours into the job, the speed dramatically drops to as low as 0.01 mbit/s and stays minimal for extended periods, vastly underutilizing my available bandwidth as confirmed by speedtest.net results.

Additionally, there's a scheduling conflict that exacerbates the situation. While the "Clone Backup to Backblaze" job is ongoing, the original backup process starts, writing data to the same source targeted by the clone job. I've tried to mitigate this by scheduling the clone job to run over the weekend, ensuring it has ample time to complete before the original backup resumes on Monday. Despite these adjustments, the issue persists.

Could you please provide guidance or solutions to address the following concerns?

1. The drastic fluctuations in data transfer speed, particularly the periods of significantly reduced speed.
2. Strategies to prevent scheduling conflicts between the clone and original backup jobs, ensuring both can complete without interference.

Thank you for your assistance.

[The Official Moderator]

7 hours ago, SysErr said:

I am encountering a significant issue with a "Clone Backup to Backblaze" job that I've set up. The task is designed to transfer all my backups, totaling 1.4 TB, to Backblaze using a 275mbit/s internet connection. Based on my calculations, I anticipated the job would complete within 13-14 hours. However, after 43 hours and 30 minutes, the progress is only at 88.5%.

Initially, the transfer rate peaks between 200-300mbit/s, which aligns with my expectations. But, a few hours into the job, the speed dramatically drops to as low as 0.01 mbit/s and stays minimal for extended periods, vastly underutilizing my available bandwidth as confirmed by speedtest.net results.

Additionally, there's a scheduling conflict that exacerbates the situation. While the "Clone Backup to Backblaze" job is ongoing, the original backup process starts, writing data to the same source targeted by the clone job. I've tried to mitigate this by scheduling the clone job to run over the weekend, ensuring it has ample time to complete before the original backup resumes on Monday. Despite these adjustments, the issue persists.

Could you please provide guidance or solutions to address the following concerns?

1. The drastic fluctuations in data transfer speed, particularly the periods of significantly reduced speed.
2. Strategies to prevent scheduling conflicts between the clone and original backup jobs, ensuring both can complete without interference.

Thank you for your assistance.

Hello @SysErr, to assist you with your request, please send us a support bundle. It contains log files and system information that will help us identify the problem and provide a resolution.

Please remember to include the main database and specify your ticket ID #244546.

The instructions on how to generate and send a support bundle can be found in the NAKIVO Help Center at the following link:

https://helpcenter.nakivo.com/User-Guide/Content/Settings/Support-Bundles.htm 

We look forward to your reply and assisting you further.
 

[The Official Moderator]

On 3/11/2024 at 9:28 AM, SysErr said:

I am encountering a significant issue with a "Clone Backup to Backblaze" job that I've set up. The task is designed to transfer all my backups, totaling 1.4 TB, to Backblaze using a 275mbit/s internet connection. Based on my calculations, I anticipated the job would complete within 13-14 hours. However, after 43 hours and 30 minutes, the progress is only at 88.5%.

Initially, the transfer rate peaks between 200-300mbit/s, which aligns with my expectations. But, a few hours into the job, the speed dramatically drops to as low as 0.01 mbit/s and stays minimal for extended periods, vastly underutilizing my available bandwidth as confirmed by speedtest.net results.

Additionally, there's a scheduling conflict that exacerbates the situation. While the "Clone Backup to Backblaze" job is ongoing, the original backup process starts, writing data to the same source targeted by the clone job. I've tried to mitigate this by scheduling the clone job to run over the weekend, ensuring it has ample time to complete before the original backup resumes on Monday. Despite these adjustments, the issue persists.

Could you please provide guidance or solutions to address the following concerns?

1. The drastic fluctuations in data transfer speed, particularly the periods of significantly reduced speed.
2. Strategies to prevent scheduling conflicts between the clone and original backup jobs, ensuring both can complete without interference.

Thank you for your assistance.

@SysErr Thank you for your patience as we looked into your backup copy job to Backblaze B2. We've reviewed the logs and found no errors with the "Clone-Nakivo-to-Backblaze" job.

With backups to S3 storage types like AWS, Backblaze, or Wasabi, there may be periods of time when no data transfers are happening. This typically happens at the end of the backup process. At this stage, NAKIVO makes an API call to move data from a temporary "transit" folder to the repository. Depending on how much data there is, this moving step can take a while, and during this time, you won't see any data transfers happening on our side.

To help with this and improve your backup job performance, we recommend setting your backup jobs to the "active" full backup type for any backups to S3 storage. Also, to prevent any overlap and ensure both your clone and original backup jobs run without issues, it's a good idea to schedule them with more time in between.

This gives the clone job enough time to finish up, considering the extra time needed for data to move to the cloud storage. We're here if you need more help or have other questions.

Your feedback helps us make our service better for everyone.