CBT failing on drives other than Hard Disk 1

[rhnb]

I just rec'd a download link for this patch (3) as I'd opened a support call I guess.
My call had been closed earlier too, but I pointed out it wasn't fixed and they re-opened it.
I'll be installing the patch tomorrow, so fingers crossed!

[rhnb]

Installed patch 3 this morning and ran one of my smaller jobs (just 3 VMs which all failed CBT on 2nd/3rd disks previously).
Happy to report the job completed successfully and used CBT on ALL disks.
Thankyou, Thankyou, Thankyou!
This was causing real headaches for me in terms of backup window.

The patch install was very slick - I'd just have liked a run down on what fixes were in the patch but I guess that will be available when it's officially released.

[cag]

[merged]

This is displayed as a warning

Cannot use CBT: Soap fault. A specified parameter was not correct. . deviceKeyDetail: '<InvalidArgumentFault xmlns="urn:internalvim25" xsi:type="InvalidArgument"><invalidProperty>deviceKey</invalidProperty></InvalidArgumentFault>', endpoint: ''

What does this mean and should a worry that the backup is bad ?

[Sethbartlett]

http://www.veeam.com/KB1113

You can search our KB at www.veeam.com/support also.

[tbullock]

Just to add my 2 cents, as I have been using this post to try to resolve a similar issue.

Some background - for a variety of reasons we cannot upgrade to v6 yet, so this problem is ocurring on v5. It does not produce an error - it merely refuses to use CBT, so the backups take forever (18 hours for our data servers). I have been sending in tickets with support for months, and they have never mentioned this thread, or admitted it's a common problem. They have suggested all the usual things - delete the ctk files and run the backup twice; shut down the VM, reset the flags, power on, power off, power on, etc etc. I have even moved my ERP sysem machines from using NFS to iSCSI because that was supposed to solve problems. Mostly to no effect.

It has, however allowed me to make some observations:
- the problem only seems to occur on machines with more than one drive
- storage on iSCSI seems to allow the HOTADD/CBT combination to work (even on machines with more than one drive)
- disabling HOTADD and forcing Network mode seems to fix the problem (eg my mail server went from average 2.5hrs to under 10 minutes)

Again, this is on v5, not v6 - I understand that there is a patch on the way for v6 that should correct the problem, but for all those on v5 maybe this post will help.

PS Somebody please tell support, so they don't wast a month of somebody's time!!!!

[rhnb]

Well the patch 3 for v6 has fixed this problem for me.
I never saw the issue under v5, but reading other posts in this thread it seems like the problem could occur in v5 also.

[aicpcuwes]

I had the CBT problem as well and I was VERY FRUSTRATED. Actually contemplated going back to version 5 and scrapping all my replicas. I just applied patch 3 and all seems to work great. Thanks Veeam for putting this out!!

-Wes

[jonwhiteman]

Got the patch 3 also after logging a support call, CBT backups all working fine now.

Thanks Petr for your help.

JW

[mattf]

http://www.veeam.com/KB1113
You can search our KB at http://www.veeam.com/support also.

Is this also the fix for users on v6? I am having this issue on three of my VMs. (I do have the latest patch)

[foggy]

If you have the latest patch installed, then you most likely encountering a different issue than the one originally discussed in this topic (which cannot be fixed by resetting CBT but is fixed by applying the patch). While you can still try resetting the CBT for the VM, I would suggest to open the case and provide the log files to investigate the exact reason of your issue. Thanks.