The percentage complete isn't accurate when the job includes a synthetic backup. It goes to 99% for the main part of the backup but then sticks there for about six hours in my case as the synthetic backup completes. There is a percentage done on the job log itself, but it should really include the amount for the synthetic backup in the overall percentage. After all, it knows it's going to be doing a synthetic backup at the start of the job and it knows how much it's going to read from the prior backups so it could blend the overall capacity into the job progress bar. Also, the "Processing rate" during a full backup isn't update either.
Hello,
There is no tracker for issues on GitHub. The forums, and support tickets, are the best way to ask for feature requests, etc.
I should admit that it makes sense and these are two different tasks themselves, talking about an incremental job + the synthetic operation to merge them to a full.
I will ask to see if there are some plans to better differentiate those or not, I know that works this way so meanwhile the synthetic is generated, the job does not run and triggers another incremental, etc. But as said, perhaps that operation can be chained like the offload or something similar.
Changing a bit the topic, and this is why we do not get much requests of this, what are you using for backup repository? As I can not see the fast-clone tag, I am assuming it is not ReFS, or XFS. And if you are using synthetic for this, and other backups. I would strongly recommend to take a look at those, as the synthetic will take much less time, always depending of backup repo disks, raid, etc. But certainly less than normal synthetic. Another option is to produce a few more synthetics during the week, like tuesday/friday or wed/sat so you have less to merge.
Thanks for the feedback.
Jorge de la Cruz
Senior Product Manager | Veeam ONE @ Veeam Software
Ohh this is just a personal development server backup sat in my cellar. It's not a serious production environment These days most of my clients have moved 100% to the cloud where having to worry about backups isn't top of the list anymore. Kind of miss BIG metal In fact, all my mission critical stuff is in M365 SharePoint so the server could blow up right now and I'd not cry too much.
That said, my main client uses iLand hosted Veeam and we're reviewing that. Having run Veeam M365 for a couple of smaller clients (e.g. 10 people), I know it's not THAT hard to manage yourself. So we're considering hosting it ourselves. Would love to throw some big metal together for that but it's just not going to happen. So finding the right virtual host that doesn't charge £££ for backup storage is going to be critical. As this will become their historical backup (needed for regulatory environment), we also need to mirror the backup system. When a backup is mainly a live snapshot and you don't care about history, having the backup disappear is annoying but not deadly. Just a bit exposed whilst you fix and get the first full backup done again. But if you need historical backups, then it's critical to have a copy of the entire backup system. I like Veeam for that.
>As this will become their historical backup (needed for regulatory environment)
In fact, having it hosted with iLand is a bit of a problem if we do move elsewhere. They effectively host the historical backup. We'd preferably want to migrate that to our own host. I doubt they're going to be open do doing that. That's the downside of cloud hosted...
>I will ask to see if there are some plans to better differentiate those or not
Did you ever get anything from this? Thought it might have been improved in V12. Weekly synthetic backup is running again and the following is wrong:
Job progress is 99% - nowhere near that. The synthetic backup will take another six hours
Transferred figure is stuck at 10GB
Throughput is zero but but it's currently writing at 300Mb/s
The % done against the synthetic backup is completely inaccurate (at least fix this!). Says 50% but it's only been going an hour - takes ~6 hours so it's more like 10%
Basically the display doesn't update at all for the synthetic backup stage. Probably worth thinking about separating the synthetic backup from the backup job like they've done with the health check.