Version 8: reset CBT manually?

[dr.Koen]

Hi

In version 8 when resizing a virtual disk CBT is automatically reset for the virtual machine, for all virtual disks. Is it possible to disable this and reset CBT manually? Does CBT has to be reset for all virtual disks when there is only one that has been extended?
I have a VM with some huge virtual disks so an incremental backup takes a very long time when CBT is reset. I would like to have more control over this because this can seriously mess up my backup schedule.

Thanks,

Koen

[foggy]

Koen, there's no currently a way to disable this behavior, CBT is reset for all the VM disks.

Do you expand VM disks frequently? What you could do to get more control over this and avoid unexpectedly long incremental runs, is perform disk expansions right prior the full job runs.

[dr.Koen]

Hi Foggy

I can understand that the reset is automatic, but is it really needed to reset CBT on all virtual disks? I have this one VM with a a total of 10 TB spread over several virtual disks. The CBT reset causes the different backup jobs for this server to take several days, instead of a few hours. I don't take a full backup, I thought this was not needed anymore with forever incremental? Up to now I used to run the virtual disks close to their capacity and expand when needed (say once a month or so). But I will have to reconsider this strategy if it is not possible to fine tune the CBT reset behaviour of Veeam.

[Gostev]

is it really needed to reset CBT on all virtual disks?

Yes, as CBT is a global VM setting, it's not per-disk. Let's hope that VMware fixes this bug soon, so that we could remove automatic CBT reset.

[jim3cantos]

Hello:

in VEEAM BACKUP & REPLICATION 8.0 RELEASE NOTES, step 11 of the upgrade procedure to veeam 8 says:

Reset CBT for all VMs in the environment. For more information, refer to Veeam support article KB1940.

This is mandatory for the upgrade?
thanks.

[Gostev]

Yes. Otherwise, your backups may be impacted by known VMware issue and may not be recoverable. It is up to you if you want to take this risk.

[r.mckeon]

Yes. Otherwise, your backups may be impacted by known VMware issue and may not be recoverable. It is up to you if you want to take this risk.

My doubt is, and I think Jose was asking the same thing, is step 11 necessary IF I have never expanded vmdks in my environment? Naturally, if I never expanded a vmdk then I am not susceptible to the CBT bug, therefore why do I need to reset the CBT on all VMs during the upgrade process?

Also, I am currently using Veeam 6.5 and have never expanded a vmdk. Once I upgrade to Veeam 8.0 and confirm that its working, I plan on expanding a vmdk on one of my VMs. Will Veeam v8 reset the CBT and do a full scan on the next job automatically? Is this why step 11 is necessary or will the CBT be reset automatically regardless?

Thanks in advance for your patience, I know you guys have been bombarded with CBT bug questions these last few months...

[foggy]

Rick, if your disks have never been expanded, you do not need to perform any actions regarding that. After upgrading to v8, Veeam B&R will handle everything automatically.

[dr.Koen]

VMWare has issued a patch for the CBT-bug, and my system is updated. When can I expect Veeam B&R to not automatically reset CBT anymore?

[foggy]

As far as I know, there is no decision regarding this yet.

[dr.Koen]

I hope that this will not take too long (the next patch perhaps?).
And may I suggest that choices like these, that can have very significant consequences, are left to the user? A warning message explaining all possibilities and possible ramifications would have been far better than leaving no choice at all. At this moment I'm rather stuck and have to pull all sorts of tricks to avoid my main backup job to take serveral days instead of a few hours.

[Vitaliy S.]

Hi Koen, can you please tell us how frequent do you resize virtual disks of production VMs and what is the usual reason behind that? Thanks!

[dr.Koen]

Hi Vitaliy, I don't have exact statistics on that. Maybe once every three/four months. I only do it when available space is becoming scarce on the virtual disk. I try to keep the size of virtual disks in line with the real usage so that there's is not too much, not too little free space available. I could assign the maximum possible for all disks, but then I would be overcommitting a great deal, and I'm not comfortable with that.

[Vitaliy S.]

Got it, so currently I would suggest resizing these disks prior to the weekend as a temp solution, thanks for sharing your use case.

[dr.Koen]

Hi Vitaly

Concerning the upcoming Update 2, I'm reading in "The Word from Gostev":

"...as there is a workaround in place to prevent this issue (CBT-bug in VMWare) from corrupting your backups, and we are keeping it in place for Update 2".

I'm hoping that this does not mean that I will still have no choice to disable the CBT-reset in Veeam? This is a real pain for me, and our system is not even impacted by the CBT-bug anyway.

Koen

[Vitaliy S.]

Hi Koen, since vSphere 6 has this bug in-place the workaround from Veeam is also part of the update. Thank you for your understanding.

[Gostev]

Let me check with the devs - we may still be able to squeeze in a registry hack for disabling CBT reset.

[dr.Koen]

Hi Gostev

That would be most welcome.

Koen

[Gostev]

It is going to make it into Update 2.

Just wanted to clarify - is this because your hosts are patched against this issue already?

This is a real pain for me, and our system is not even impacted by the CBT-bug anyway.

Thanks!

[dr.Koen]

Fantastic news. And yes, my hosts are patched (Version 5.5).

Thanks,

Koen