Support

Some servers have a very strict CPU usage limit which could kill the backup, especially when it runs during the peak hours of the server. In order to work around such hosts, please go to your Akeeba Backup Configuration page and scroll all the way down to the Fine Tuning pane. Set the following:
- Minimum execution time: 1 second
- Maximum execution time: 7 seconds
- Runtime bias: 50%
If this doesn't work, please try these:
- Minimum execution time: 5 second
- Maximum execution time: 3 seconds (yes, maximum is less than minimum, it's not a typo)
- Runtime bias: 50%

If that still fails, there is one more plausible explanation. Your backup failed after exactly 24 minutes into the process, or 1440 seconds. It is possible that your CRON daemon –the programme which runs the CRON jobs on their scheduled time– is set up to not allow any job running for more than 1440 seconds. In this case, apart from using webcron.org and its maximum allowance of 3600 seconds per job, I don't have any solution.

Well, I do have a workaround, but you're not going to like it. The workaround is having two backup profiles. One which backs up the entire site except VirtueMart images (this is what is taking a lot of time), the other on a files only backup excluding everything else except VirtueMart's images. Splitting the backup in two would allow each half to run for less time, therefore not likely to hit the CPU usage and/or CRON timeout limits of your server. Restoring such a backup means that you extract the files-only backup first, then proceed to extract and restore the full site backup. More work during restoration, but should work around your server's limitations.