Copy Job FAILS!??

[swilsonbsb]

My business requirement is to get my backup data copied offsite as soon as possible. Getting backup information offsite is of utmost importance to make sure that we can recover from a local disaster. Backup copy jobs seem to be built without the importance of getting data copied in mind, and this blows my mind for a backup product.

Given a setup where I need backup data to be copied every day and so I have my copy interval set to 1 day, if the copy job takes longer than 1 day then the behavior is to FAIL on the copy?! Finishing the copy is the most critical piece of a backup operation, why would you not finish the copy?! Throw a warning, email me about a job running too long, whatever... but in the name of all that is holy what BACKUP software company would ever green light a behavior that FAILS on anything when it could have finished... just because it might be running long??

If someone out there has a business requirement that they can only have data be copied for X hours and if it takes longer than that then we must stop the copy and give up on our backup data being copied, then I'd love to talk to them because that's just ludicrous to me. Even if it exists, then you handle that with a setting that says jobs are prohibited from running during certain hours, or you have a separate setting (like Windows scheduled tasks) that says to stop the operation if it takes longer than X hours.

Assuming that the best course of action would be to fail on a copy job just because it's taking a long time is just insane.

Can someone please tell me that this is something that can be overcome or is in the works for a fix?

[PTide]

Hi,

Could you please specify your source backup job schedule settings? Backup copy job should not give any errors/warnings only if there were no new restore points detected on the source side. Please also provide the message that the backup copy job ended with.

Thanks

[swilsonbsb]

One example is this:

Backup job named Backup CORE Folder ran at 1800. Copy job is set to 1 day at 1815. This backup usually finishes in 20-30 min.

The Backup Copy Job kicked off at 1900 on day 1 and ran until 1955 on day 2 and then failed.

There is a warning on the copy job that says Some VMs were not processed during the copy interval.
There is an error on the first VM that says Copy interval has expired.
The other VMs just say job has failed, see logs.

This Backup Copy Job then ran again and completed.

Copy jobs should just start whenever new data is available and run until the copy is finished. Failing and starting again just because the copy took a long time is aggravating.

[Thernlund]

If your source backup is daily, and your backup copy is also daily, and the copy takes longer than a day, your backup copy will be forever falling further and further behind your source until your source retention expires and fails it anyway. Failing it early on makes sense.

Sounds me to that your problem isn't Veeam, but having adequate bandwidth to ship your off-site copy. I'd get more bandwidth. In fact, that is precisely what I had to do.

Maybe try using the WAN Accelerator if you don't want to get more bandwidth. Might help.

[PTide]

<...>This Backup Copy Job then ran again and completed.

Did it start again immediately?

Copy jobs should just start whenever new data is available and run until the copy is finished. Failing and starting again just because the copy took a long time is aggravating.

If that was the very first synchronization then the backup copy job had to start copying a new increment + the remaining part of the full right after it had detected that the new restore point within the copy interval (24 hours) was available on the source. According to your backup job settings it had to happen immediately. If that was one of the subsequent synchronizations then the backup copy job had to wait for the expiration of the synchronization interval and start a new one right away to finish copying. In other words in both cases there should be no significant delay between copy job failure and its resumation. If that wasn't the case then please consider contacting support team and post your case ID here.

Thanks