Automatically end backup after x hours?

[csanburn]

We have a situation where backups sometimes run long and they slow down the users. They'd rather cancel the backup than deal with the slow response of the servers while backups finish. Is there a way to have Veeam automatically cancel it's backup if it runs past a set amount of hours, like you can do with Windows task scheduler?

[Gostev]

Chris, no we do not have such option. I don't think cancelling backup is good idea, because this will make you unable to restore this latest backup. Did you consider introducing additional backup server to be able to fit your backup window?

[glennsantacruz]

Is there no "eject button" to cancel a bad backup? I agree that it's not a good idea to cancel a backup on the principle that it would render that restore point invalid, but in the event a backup spins off into never-never-land, you really should have some means to cancel it. What if someone "accidentally" moved a huge VM into the wrong folder and Veeam decides to start backing it up? Maybe that huge VM is a SQL server and is using some other native backup agents to perform a different type of backup... and would very likely exhaust the allocated space on the backup target.

[tsightler]

You can manually stop a job in progress, there's just no automated way based on time of day to stop a job that's run. You could probably script something easily enough though.

[glennsantacruz]

Ok, thanks for the clarification. Being new to the product definitely shows through, doesn't it? Although I still agree with the original post -- a feature of "hard cutoff" or "soft cutoff" would be nice. Hard cutoff would kill any currently running jobs at the window, ans soft cutoff would allow a job to finish the current stage, but not process further ( imagine a job consisting of 12 VMs - 8 might make it within the window, the 9th may be running as the window closes; the 9th would be allowed to complete, but 10 thru 12 would not even attempt ). Does Veeam have this type of "soft cutoff" baked in to the product already?

[tsightler]

I'm not aware of any cutoff at all. I've used a lot of backup products over the years and I don't know of any that had a "cutoff" like that. Part of designing your backup solution has to include getting the backup done within the backup window or implementing a solution that doesn't interfere with normal processing. I'm not really sure how stopping the job would really help much, certainly it would allow the system response time to return to normal, but would the problem just occur again the next day since all the previous nights backups did not complete? Seems like you should figure out why you're seeing so much variance.

I'm also amazed that Veeam is putting so much load on the storage that it's impacting performance that much. We've found that Veeam, at least with change block tracking, is pretty light on the storage overall. I can kick off backup jobs during the day with minimal impact on our production array from a performance perspective since Veeam only has to pick up the changed blocks. You might consider throttling I/O to the Veeam server if you have that capability, that might allow Veeam to run with less of an impact on production. Are you perhaps running a lot of jobs in parallel?