Rotated hard drives support in v8

[omerlin98000]

Hello all

I have updated to v8 patch 2, in order to get the new rotational feature.

I have backup jobs to a Synology NAS, then backup copy jobs to external e-Sata disk(s) connected to this NAS:

All repositories are addressed through CIFS.

Actually, I used a trick not to misuse each external disk, I created my two repositories with different subdirectories names (1 and 2), and accordingly 2x3 backup copy jobs that I enable/disable each week when rotating my disks:


Now, I want to create a SINGLE backup job and tell B&R that it is a rotational disk, so I can easily swap my two external disks: I can't see how to do this!

[marco.horstmann]

You can Edit your backup repository and in Advanced Settings there is a checkbox for saying
that Repository is backed by a removable device.

[omerlin98000]

Many thanks, Marco, I missed it!

[foggy]

Olivier, you can also review the thread above for some considerations on implementing the rotated drives scenario. Thanks.

[foggy]

- In v7, I was unable to have Backup Copy Jobs re-use an existing .vbk already on the disks. Can the job do this now with a week old .vbk that's not in the Veeam Backup retention records?

Chris, provided you have Veeam B&R v8 Update 2 and repository is not CIFS, this is possible. Just enable the rotated drives option for the corresponding backup repository.

[souperstar]

Alexander, I can't thank you enough. I will use this, and the previous information in this thread to get me up and going. I will probably post back on my success/failures. Also, thank you to whomever merged my question with this thread. I've no idea how I missed it but it didn't come up with the searches I was doing, so I did something wrong there.

[larry]

I have seven Veeam 8 repositories with one USB drive per repository. How do I add the second disk to the set? With 7 removable drives running at a time when I remove one how do I add a second one(rotation for one). When I am done I want 7 sets of drives, two drives per set. One drive of each set will be in the vault the other one spinning. I did read this http://helpcenter.veeam.com/backup/80/v ... ub=rotated but I can’t find how to say a new drive belongs to which group.

[foggy]

Larry, you just need to enable rotated drives scenario in repository settings and can switch drives after that. No additional actions are required. If you want to save yourself from consequences of the drive letter change, make sure the letter is the same during the first switch.

[jplavertu]

Quick questions :

- Repository scan grayed out : Is it because option This repository is backed up by rotated drives is enabled ?

- How to do test restore ? Whenever you plug back a USB drive, Veeam ( at least on my side ) would not recognized the contents on it and there is no way to rescan the repository. I tried restoring to points I had, which seem less than all the restore points ( 5 ) I should have had ( which leads me to think Veeam did not register the second USB HD, even though it backed up to it and everytime it would tell me it cannot find the vbk or vib file on E:\Backups.
I tried importing from 2nd usb HD and then I had other restore points but the other ones disappeared after Veeam told me : backup already exists

[foggy]

- Repository scan grayed out : Is it because option This repository is backed up by rotated drives is enabled ?

Correct. Rescan is disabled for rotated repositories to avoid possible issues with restore points in database (rotated drives logic is implemented in the job itself and "external" rescan could spoil it).

- How to do test restore ? Whenever you plug back a USB drive, Veeam ( at least on my side ) would not recognized the contents on it and there is no way to rescan the repository. I tried restoring to points I had, which seem less than all the restore points ( 5 ) I should have had ( which leads me to think Veeam did not register the second USB HD, even though it backed up to it and everytime it would tell me it cannot find the vbk or vib file on E:\Backups.
I tried importing from 2nd usb HD and then I had other restore points but the other ones disappeared after Veeam told me : backup already exists

I recommend asking support for assistance in reviewing your job/repositories configuration.

[larry]

I am running a backup copy job to removable disks, one disk for each day. One job of each day of week, with 7 disks pre- loaded. How can I stop the job after it is done so I can eject the disk( safety remove) . It would be great if Veeam could safety removed the disk. Now the job which only runs one day a week keeps the files open, which requires to disable the job to be able to then eject the disk. We then need to install new disk and remember to re-enable the backup copy job. I tried switching to filecopy jobs or robocopy but it takes hours longer. I don’t want to use backup because it will pull from production and over the wan again. I really just need backupcopy to run and then close files when the job is in its wait for permitted network usage period.

[foggy]

Larry, am I understanding you right, you have seven backup copy jobs (with weekly sync interval) for each day of the week writing each to its own disk (that are rotated daily within the same repository)? In this case, why not to use just a single backup copy job with daily interval? Remember, backup copy jobs are capable of resuming existing chain on the rotated drive and in case of daily interval it will release files right after interval expires.

Another alternative is disabling metadata caching.

[larry]

is every thing on the drive needed for a full restore of that day? I do it this way to get the full on each days disk so each days disk is a standalone restore (a requirement). I also preload all seven disks which I could not figure out how to do with one job. I all so need to remove one disk each day ( take off line - air gap requirement ). If I could do with one job would be much cleaner I just couldn't get to work. Still testing the switch from tapes to these drive so any suggestions would be timely.

[foggy]

is every thing on the drive needed for a full restore of that day? I do it this way to get the full on each days disk so each days disk is a standalone restore (a requirement).

Each disk will contain its own backup chain including corresponding full, so you will be able to restore from every disk to any weekly restore point stored on it.

I also preload all seven disks which I could not figure out how to do with one job.

No specific actions are required, disks should be recognized automatically, once introduced to the job for the first time.

[larry]

I am looking at a way to trip off sync now for a backup copy job. In the morning when I place in a new blank disk I want the backup copy job to start syncing when I switch drives ( 3 tb of data to sync) , Veeam sees that I switched drives, warns me but waits for new restore point, I know I have a good backup but just do it. During the day the serve is idle and I want it to copy lasts night backups to the clean disk then after the new nightly backups it can do just an incremental to the removable disk of only 350gig this way (clean disk ) . I have everything automated but the sync now. I want the job to think it needs the sync now even though it does not.


I am using the one daily job now but am stuck at this step.

Hope I am clearer than mud.

[foggy]

Switching the disks itself does not trigger the new copy interval so you either need to right-click the job and select Sync Now manually or schedule the job the way so that the new interval starts after the moment you typically switch disks.

[larry]

I changed the drive and it just says "waiting new restore points" but there are none on the disk now. I tried switch to hourly from daily but always continuous but sync now just says start new, click yes, stopping, Idle, Waiting backup Repository , rotated Drive letter not change, New Copy interval started Building VM list, Waiting for new restore point points. I go to the disk there is only 93 meg in the folder, the vm's are 800 gig. waits for new restore point until the backup job runs.


I did find if I change the job setting to copy from copy every 1 day to 1 hour save, back to 1 day save I get a yellow warning job canceled by user then job restarts and syncs now. Tried on each jobs and works each time, 4 for 4.

[larry]

I changed the drive and it just says "waiting new restore points" but there are none on the disk now. I tried switch to hourly from daily but always continuous but sync now just says start new, click yes, stopping, Idle, Waiting backup Repository , rotated Drive letter not change, New Copy interval started Building VM list, Waiting for new restore point points. I go to the disk there is only 93 meg in the folder, the vm's are 800 gig. waits for new restore point until the backup job runs.


I did find if I change the job setting to copy from copy every 1 day to 1 hour save, back to 1 day save I get a yellow warning job canceled by user then job restarts and syncs now. Tried on each jobs and works each time, 4 for 4.

skip this post..or delete, the weird results was from having a copy mounted on a different drive letter which would not be normal. Doing a sync now with Sync-VBRBackupCopyJob works as expected if I dismount the extra copy.

[bernardw]

Larry,

So you are able to get Veeam to start syncing the backups to your drives in the mornings even when your backups happen in the evening? I would really like the same thing to happen but I have not figured out how to do it from the GUI. I do not get the data to sync even after right clicking on the job and selecting "Sync Now". How did you achieve getting the system to start copying data?



Also for the people that have been working on figuring out how to make Veeam have 1 full backup on the drives even though they are rotated every day I am experimenting with a backup interval of 4 hours. This causes Veeam to copy all the data over and then merge the data. (seeing how we can't setup a repository with only 1 backup). I am having to "force" the merge because my backups are 4 TB and the drives are 6 TB and a single full backup of everything takes 16-18 hours to complete. As Foggy or Gustov mentioned earlier, depending on how the incremental backups are "seen" they can be larger then normal - I have had 1.4 TB incrementals, however I have had fewer issues with these "large" incremental backups when I have a complete full VBK file on the drive.

My biggest frustration with the rotated drives have been:
1) forcing Veeam to have a single good VBK file on the drive
2) to start backing up as soon as a "new" drive is inserted (which Veeam obviously detects since it reports the content has changed).

[larry]

I created a backup job that backs up a 2 meg vm host at 11am ( just to use the gui) . When job finishes a ps1 script starts the backupcopy job with yesterdays data with time if day restrictions which will stop the job if still running when the nightly job runs. then at night the real job kicks off which is only changes for the day .

Code: Select all

asnp VeeamPSSnapin

$CopyJob = Get-VBRJob -Name "USB-Groton-A-C"
Enable-VBRJob -Job $CopyJob
Sync-VBRBackupCopyJob -Job $CopyJob

If the main ps1 I check for errors and disable jobs if needed.

Code: Select all

$CopyJob = Get-VBRJob -Name "USB-Groton-A-C"
Disable-VBRJob -Job $CopyJob

I found if I disable the jobs and the main job is still running or other errors I have time to fix before everything is automatic.
On error I send an email

[souperstar]

I have to say, I find the Backup Copy Job scheduling confusing. I'm not sure when to schedule it?

I have normal backups writing to a physical Veeam server, with a local repository comprised of DAS. There are multiple COTS disks attached via SAS appliance that I then use a Backup Copy Job to write to. These COTS disks are rotated daily, the five working days of the week. Backups and Backup copies are scheduled during the evening of the five working days of the week.

Breaks down like this:

• Disks get loaded into the SAS appliance in the morning sometime (no set schedule, but they usually do it before 10AM)
• Backups kick off about 5:15 PM. Secondary destinations are configured to also write to the COTS disks
• Currently the Backup Copy Jobs are scheduled to sync daily at 10PM, however I can see a job running right now that in a few minutes is going to "fail" because it isn't finished yet. It's busy "Merging oldest incremental backup into full backup." I only really need the one restore point on the rotated disks, so I don't think I care about this, and it's also taking foreverrrr.

I've tried 4AM, I've tried 4PM... They generate errors because they try to sync and don't have enough time before they're either removed or the Backup Jobs start. I just don't know what "time" to schedule the Copy Job in my case. In fact, I don't really care to schedule the darn thing at all. I just want it to start copying AS SOON AS the Backup Job is done. That one time, one backup on those disks is all I need.

Is this possible, or need to be a feature request, or is there something I'm missing about how to schedule the Backup Copy Jobs?

I don't want to resort to scripting because that'll take me forever and a day to write, but I can if I have to. I tried something similar in V7 and I couldn't pull it off, where I was disabling the Copy jobs until the Backup job was successful and then I would enable the Copy job, and post-script the Copy job to disable itself on success. It just didn't work the way I'd hoped.

[foggy]

I just don't know what "time" to schedule the Copy Job in my case. In fact, I don't really care to schedule the darn thing at all. I just want it to start copying AS SOON AS the Backup Job is done. That one time, one backup on those disks is all I need.

Just set the copy interval start time to somewhat several minutes after the source backup starts. Backup copy will start new interval and wait until new restore point is available. Enable forward lookup for the backup copy job to avoid copy of previous points, that are already on disk at the time of backup copy job creation.

[souperstar]

This is exactly what I was looking for, and thank you very much foggy. I did both, and look forward to tomorrow's Enterprise Manager report!

[Stonki]

As of B&R 8.0, rotated drives functionality can now be enabled with a check box in the advanced backup repository settings, as opposed to being controlled by ForceCreateMissingVBK registry settings (which, however, is still supported for upgrade compatibility reasons).

ForceDeleteBackupFiles reg key also works as before, not to change the product behavior after v8 upgrade. However, we strongly recommend deleting this key, because backup files clean up is now integrated, and is done automatically for all Veeam backup files which are no longer under retention policy (see v8 What's New document).

Additionally, v8 backup jobs now track known external hard drives, and "follow" them transparently in case drive letter changes, instead of failing.

sorry jump into this topic, but our backup with three rotating drives stopped working two days ago with the error message: ": Error: Full storage not found "
what makes me wondering, that the Veeam support ask me to add the "forceDeleteBackupfiles" to the registry. Of course it did not change anything... So should I remove it again?

[foggy]

Judging on the error, there's no full backup (VBK) on the inserted drive, can you confirm that? How rotated drives are configured? Do you have the corresponding option in repository settings enabled or ForceCreateMissingVBK registry key specified?

[Gostev]

Yes, please remove the key - bad suggestion by our support engineer this...

[Stonki]

Hello,

after two (three?) days of errors, tonight backup was running fine. Although I tried yesterday evening manually before dinner. Strange. Not sure if this is now permanently solved or will occur on the next change of the hard drive..

1) The backup repository is set as "This repo is backed by rotated hard drives"
2) The backup itself is set to "reverse Incremental"

I added all logs to the support request (00973510). Even the logs from todays succesful backup. Wondering what caused the problem... (might be 20cm in front of the monitor, might not *g*)

[veremin]

It's probably worth observing situation for some time and reaching the support engineer, should the problem re-occur. Thanks.

[vpock]

Hello all,

So, this is the first time a customer asked for this, so I'm a little puzzled.

The customer use a physical Windows server as a repository. He would like to archive the backups on a set of USB drives using backup copies. I'm not sure how to do this reliably.

Anyone using USB drive for backup copies?

[foggy]

Dominic, you need to enable the rotated drives option in the corresponding repository settings. Please review this thread for details. Thanks.