Backup Copy - Fail to generate points error

[shuji]

Hi there,

I found the backup copy job cannot merge full backup file.

Red lines are:

Code: Select all

2016-04-12 11:10:11 AM :: Failed to merge full backup file Error: Backup copy interval for storage 92222d24-XXXX-XXXX-XXXX-3b7c0d7e23eb not found
2016-04-12 11:10:36 AM :: Failed to generate points Error: Backup copy interval for storage 92222d24-XXXX-XXXX-XXXX-3b7c0d7e23eb not found 
2016-04-12 11:11:38 AM :: Failed to generate points Error: currentGroup is null. Point is Point '172f974d-XXXX-XXXX-XXXX-3eba20a7a712' is empty 

I have performed a backup file health check in the beginning. Why does this still happen?
More importantly, how can I fix it?

Thank you.
Ryan

[foggy]

Ryan, please contact technical support with this directly, as it is always advised to do when you have some sort of technical issue. Posts about technical issues without support case ID will be eventually deleted by moderators. Thanks.

[hyvokar]

I had couple of these.
With first one we had to manually edit the veeam database, to get rid of the null restore point.
Second one was solved by installing v9 update1.

[pkelly_sts]

Unless I'm missing a way of doing it, I'd like to be able to initiate an active full COPY for a single VM, without having to create a full copy of ALL vms in the job.

In my case it's because of a (most-likely self-inflicted) issue where a previous pass of the copy job failed for a small number of VMs but I hadn't noticed it.

Whilst tidying jobs up after some significant changes I was getting the following error at the end of the copy interval:

"Failed to generate points Error: Points group does not contain full point, oldest point date <date>"

The date in question was where the job failed part-way thought (can't remember why now) so after checking the RPs I found a number of them with a full RP with a status of "Incomplete".

So, rather than have to re-seed the whole lot, I tested deleting the backup copy (ALL rps) of a small minor VM, re-synced & as expected it started a full copy.

I'd like to be able to stage/manage such a full copy in future without having to mess around disabling/re-enabling the copy job (i.e. I had to disable the job to delete the backup but then re-starting the job started the copy immediately whereas I'd like to be able to get it to start later/OOH. This way I had to edit the schedule to block out a time window, then I'll have to edit the schedule again to block it back in again.

It would be SO much easier if I could somehow right-click a VM & say "perform active full copy of this VM on the next pass".

Hopefully makes sense!

[PTide]

Hi,

The Backup Copy job is configured to use a backup job as a source, not VMs, is that correct? Also

a previous pass of the copy job failed for a small number of VMs but I hadn't noticed it.

Why did it fail, what was the error message?

Thanks

[pkelly_sts]

Correct, copy job is using backup job as source (I thought that was the only option anyway for a copy job?)

As I mentioned, in fairness to Veeam in this case it was self-inflicted "Job was stopped due to backup window setting". It was around the time that we were moving office/datacenter amongst other things.

My reasoning is, sometimes we have failures for various reasons (network glitch, repo server crash, disk space issues are a few possibilities that spring to mind) and if a few VMs within a larger job have failures in their chain, then other than a full re-seed of the whole job (which is awkward enough, especially if you're using SoBR repos) this seemed the only way to get a fresh .VBK to the copy site, at the cost of a bunch of RPs (30 days in my case).

[pkelly_sts]

Actually checking status this morning I'm still having issues - I'm going to have to open a ticket (will post case id here when I do).

As mentioned above, one of the things I did this week on a small VM was to:
1) Disable copy job
2) Delete a VM (and all its RPs) from the copy repo
3) Restart the job

This created a .VBK for just that VM as expected.

I then wanted to do the same for more of the VMs, but I wanted to manage bandwidth a little more so, on top of the network bandwidth throttling I already have configured (which is basically no more than around half of the bandwidth of our P2P 100Mb link) I also edited the allowed hours to "stagger" the hours for the job, so from around that time, let's say 11:00 until 18:00 the "allowed" window within the copy job was disabled for every other hour.

Now I've just checked the job, I noticed more than I expected of the error "Failed to generate points Error: Points group does not contain full point, oldest point date 17/03/2017" (which is the oldest full RP on disk) even though I believed I only had two more (larger) servers to repeat the above process on.

Now that I've checked the few I've caused to re-copy this week, all of the ones that I processed during the "staggered" window have 3 RPs, but the full has a status of "Incomplete". So, the only two that copied successfully are the ones I allowed to fully copy during business hours (i.e. I didn't interfere with the schedule).

I thought the whole idea of the scheduling window was to allow things to cleanly stop/pause until the available window came back? This doesn't seem to be the case in my situation for some reason?

[PTide]

I thought that was the only option anyway for a copy job

VMs and backups can also be specified as source for BCJ.

I'm going to have to open a ticket

Yes, please do so. The error that you've got means that you're likely to have an issue with Veeam's database:

Code: Select all

"Failed to generate points Error: Points group does not contain full point, oldest point date <date>"

So, starting active full for particular VMs probably won't help.

Thanks