Newbie ? - 18VM's - how long before the backup starts?

[sbishop]

Hi everyone,

We are testing veeam backup and so far so good with individual vm's. I decided yesterday to start a complete backup of all vm's but got worried after 50min of no backup happening.

after 50min, In the job progress window, all 18 VM's had "pending" has status (creating vm snapshot and collecting disk files location data had green check mark)

in Vcenter, recent task had "map disk region and renew disk lease" every few seconds for different servers.

is this normal for 50minutes? when I've tested with 3 vm's, it starts backing up in about 4minutes. Should I have just waited longer before cancelling the job?

Thanks for any suggestions.
Sbishop

[Vitaliy S.]

Hi Steve,

No need to cancel the job. How many allowed concurrent tasks to do you have on the proxy, repository servers? Do you want to start all 18 VM backup at once or put them in the queue, so that these VMs can be backed up according to the number of allowed tasks?

Thanks!

[sbishop]

Thanks for the reply.

concurrent proxy task is set to 2, backup repository is set to 4 maximum concurrent task. I don't think I would want all 18vm backing up at the same time...I'd like to put them in queue..maybe 2-3 vm's at the time. is there a setting for that?

Sbishop

[sbishop]

I just selected 5 vm's to test and watched the vcenter to see what happens. within 4 minutes it had created 5 snapshot, whole bunch of map disk region, some renew disk lease / release disk lease and lastly remove snapshot for those 5vm's.

It then proceeded to backup the first 2 vm's, the other 3 has a status of pending. back to my original question, 50 minutes and still not backing up..is that normal?

Thanks
Sbishop

[Vitaliy S.]

It then proceeded to backup the first 2 vm's, the other 3 has a status of pending. back to my original question, 50 minutes and still not backing up..is that normal?

It depends. What was the operation in the job log session that took most time? You can check that in the Veeam console. Was it a snapshot creation or preparation for the snapshot? If you have concurrent tasks set to 2, then only two virtual disks will be processed at a time, all other VMs will be queued.

[sbishop]

the job started at 12:13pm, in the action section of the log, building vm list took 53 seconds. There is not other times listed for any other action. The last green check mark list preparing next vm for processing (no time)...then error operation was canceled by user (that was me) at 1:05pm

below is a screenshot

thanks
Sbishop

[sbishop]

[Vitaliy S.]

Ok, thanks! Can you please also post a screenshot of the first/second VM processing log? Seems like either the first or the second VM can be the reason for this unexpected behavior.

P.S. do you use storage snapshot integration feature by any chance?

[sbishop]

no haven't setup storage snapshot feature yet. Here are 3 screenshots.

[sbishop]

I did a bit of research and i might have found what the cause was....or maybe not

4 of those 18vm's had a previous snapshot, 2 of them were databases over 1TB.

what type of backup does Veeam perform? is it a vcb?

http://kb.vmware.com/selfservice/micros ... Id=1007331

Thanks
Sbishop

[foggy]

Backup is done through vStorage API (the successor of VCB). I'd let our engineers investigate the job logs.

[sbishop]

Did a full backup last night of all VM's, no issues...took 13hrs and the job started backing up after 17minutes.

Thanks
Sbishop