CDP failover not working

[oasisnet]

I recently started using CDP technology to replicate to DR site for some critical machines.
When I try to make a faiover attempt the operation fails due to the vCenter on the production site not being available. Which is a logical situation in the event of the fall of the production site.
Even when I turn on the vCenter machine that was normally replicated on the DR site, the failover operation still fails because in this situation there is no communication to the ESXI servers at the production site.
I can not find any documentation on the net how to perform failover properly.
I have applied for support and I am not getting an answer to the issue.
By the way when the production site is available there is no problem to make a Failover.
Does anyone know and have done this that can help?

This is the error message I'm getting
"Failed to login to "vcenter.lab", port 443, user "administrator@vsphere.local", proxy srv: port:0
Unable to connect to the remote server
A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond"

[HannesK]

Hello,
and welcome to the forums.

Even when I turn on the vCenter machine that was normally replicated on the DR site

that sounds like you have only one VCenter and that one VCenter is not available when failover would happen? If my assumption is correct, then a second VCenter is needed to cover that scenario. And if you are not on 11a yet, then an update to 11a is also required.

Best regards,
Hannes

[oasisnet]

Your assumption is incorrect.
I have two vCenters. One at the production site and one at the DR site. When the production site is not available I stay with vCenter only available on the DR site. After I turn on the vCenter of the production site and try to perform cdp failover the operation fails because the ESXI servers of the production site are not available which is a logical situation when the production site is not available. My version is completely updated.
The use of CDP technology does not seem to be effective in a situation where the production site is not available.
Unfortunately I have been trying to get guidance from the support staff for a long time and get no satisfactory answer.

[HannesK]

Hello,
that's wrong behavior... what is the Veeam support case number for that? I remember bug #320719 (please refer to it), but that should be fixed.

Best regards,
Hannes

[oasisnet]

hi
Case #05504551

Just to clarify that I have tried to perform the operation many times also in a test environment and got the same result
Thanks

[halllowien]

Hello,

i had the same problem (Desastertest with Source vCenter offline).
"It looks like we might face a known issue here, CDP failover tries to connect to the source vCenter server, and if the source vCenter is not available, failover fails with "Failed to login" "

I got a Private fix for this after a week (my case ID was 05493456).

Greets
David Brunner

[HannesK]

Hello,
@oasisnet: I see that support is waiting for logs. If you could export the logs from the DR cluster, that would help. That is also possible via the log export wizard (export for these objects -> vsphere -> CDP cluster).



@halllowien: thanks for posting the case number as reference and good to hear, that there was a solution for you.

Best regards,
Hannes

[oasisnet]

hi
If so it seems I am not the only one who is having this problem.
And I understand that you have a fix for the problem.
Can't you just provide me with the repair?
In any case, I'll send you the logs you requested tomorrow.
Thanks

[FrenchBlue]

Hello,

We did the tests this morning and we have the very same issue. If the source Vcenter is shut down, the cdp failover of VMs on the target site fails, saying it cannot contact the source Vcenter. This is of course nonsense for a DRP tool. A Veeam case will be created soon by IBM cloud support.

[HannesK]

Hello,
agree, bugs are annoying. I just checked again with developers and between detection of bug #363011 and today, there was no cumulative update. So the only way today is going via support to get the fix.

Best regards,
Hannes

[veremin]

Can't you just provide me with the repair?

It seems that the hot fix was provided to you some time ago, unfortunately we have not heard back from you so are still wondering whether it's solved the issue for you?

Good news - the fix is included in v12, so the issue won't bother you again, once v12 is available.

Thanks!

[FrenchBlue]

Hello,

This leads to the next question: is there something a least a little more precise than "2H" for V12 release date?

Thanks.

[veremin]

No, we cannot provide you with a date or month. Thanks!

[FrenchBlue]

OK thanks, well, at least it now works as expected once the private fix has been applied.

[veremin]

Glad to hear that the private fix has worked for you. Let us know, if other help is needed.