vCloud Support Details

[dcampregher]

Hello,

1) In the documentation there are only two options on the Platform section:

*VMware Cloud on AWS
*VMware Cloud on Dell EMC

This means that only on these environments ( AWS and DellEMC) the integrations will be supported by the Veeam Support ? Other environments like a on premisse or in anything other hardware won't be supported ?

2) About the versions:
*vCloud Director 8.20–10.1

The "10.1" is only the "10.1" or the "10.1.X" versions like "10.1.2" ?

[Andreas Neufert]

Hi Diogo,

1)

VMware Cloud on AWS, VMware Cloud on AWS Outpost and VMware Cloud on Dell EMC have a special vSphere version that need specific API calls. This is why it is listed as separate versions in our documentation. See as well https://www.veeam.com/kb2414

All other Cloud based offerings with vSphere base are supported automatically because we support the vSphere Versions that run on it.
This include but is not limited to:
Azure VMware Solution by Cloud Simple https://www.veeam.com/kb2999
Azure VMware Solution (AVS) - We are waiting for the initial official release from vendor
Google Cloud VMware Engine 1.0 (GCVE) - Cloud Simple based offering https://www.veeam.com/kb3178
Google Cloud VMware Engine 2.0 (GCVE) - Google native offering - We are waiting for the initial official release from vendor
IBM Cloud for VMware Solution (CVS) https://www.veeam.com/kb4006
Oracle Cloud VMware Solution (OCVS) https://www.veeam.com/kb4007

Same for the AHV based offerings.
This include but is not limited to:
Nutanix Cluster on AWS https://www.veeam.com/kb4005

2)
The patch versions never broke an integration for vSphere or vCD and are automatically supported.
It is 10.1.x

[maximeg]

Hi Andreas
It seem's that v10.1.2 break the VCD provisionning wizzard in veeam 10:
[17.11.2020 10:06:54] <45> Error Guid should contain 32 digits with 4 dashes (xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). (System.FormatException)
[17.11.2020 10:06:54] <45> Error at System.Guid.GuidResult.SetFailure(ParseFailureKind failure, String failureMessageID, Object failureMessageFormatArgument, String failureArgumentName, Exception

When i use postman and api to check VCD response:

API CALL on vCloud Director version: 10.0.0.16081830
GET https://atlas-fr-mar.jaguar-network.com ... 5bc-43b7-8...
<vmext:Url>https://vcd-vca-rg10-mar.pms.as30781.net:443</vmext:Url>
<vmext:Uuid>f8b92c4a-41ae-4c4e-b0a3-b755...</vmext:Uuid>

API CALL on vCloud Director version: 10.1.2.16779297
https://atlas-fr-par.jaguar-network.com ... d05-8149-4....
<vmext:Url>https://vcd-vca-rg10-par.pms.as30781.net</vmext:Url>
<vmext:Uuid>https://vcd-vca-rg10-par.pms.as30781.ne ... vmext:Uuid>

Now, the Uuid return an url not an uid...

I have a support case: 04499476
Support feedback:
So I think you are right - looks like Veeam doesn't provide support for last 10.1.2 vCD version yet. And maybe it's also can be the problem with vCenter version - I know that our R&D has already working on the vSphere 7.0 Update 1 support in the Cumulative patch for Update A. (It is next patch)

[Andreas Neufert]

Looks like I forgot to knock 3 times on wood when I made the above statement.
Thanks for sharing and I will look into this support call.

[Andreas Neufert]

It looks like a VMware bug that they send a URL instead of a UUID.

[maximeg]

We opened a support case with VMware
The Uuid value in our vCloud postgresql database is wrong, there's an URL instead of an UUID
So it's a VMware issue, maybe we'll be able to manually change it
At the moment, i don't know if it's a bug in vcloud 10.1.2 version or in our install.

[Sturniolo]

@maximeg What is the VMware case number for this issue?

Thanks

[maximeg]

Hi Sturniolo
this is our SR: SR 20169916111
We have initailly a problem with VCAV and with Veeam, after several tests and research we found that our issues had the same cause
Response from VMware support:
"After confirming the vCloud Director cell information and registered resource vCenter information from the vCD Portal, we took a snapshot of each vCD cell and a backup of the postgres database.

We then connected to the vCD database and ran the select query again against the register resource vCenters to verify we are still seeing the behaviour with the UUID column containing a URL instead.

Using the vCenter instance ID you previously gathered we then updated the database entry for this vCenter Server, setting the UUID to be the same as the instance ID present on the vCenter VM and in the Lookup Service registration.

Once this was done, we went back to the vCloud Availability Portal and saw we were able to configure replications now without issue."

Then we were able to add VCD into our Veeam server.

[Sturniolo]

This is the VMware BugId: 2680344. We are working on escalating this with VMware.

[anpa]

Can you please hint to where in the vCenter database you found this? Need to take a look in our lab, we seem to have encountered a GUID bug for a quite fresh set up for a lab.

EDIT:

Found it, have the same issue:

Code: Select all

vcloud=# select name,uuid from virtual_center;
     name     |                 uuid                  
--------------+---------------------------------------
 Labb-vCenter | 0034117e-b58b-4aca-8f96-50774aba6d1f
<new vcenter> | https://<url>/sdk

Exiting, and this is 10.2.2 we're running. The new vCenter was added later than the first one. So yes, the bug is still here. Annoying!

[Andreas Neufert]

Looks like it came back, as 10.2 had no issues. Only 10.1.2 was affected. We will follow up with VMware.

[Andreas Neufert]

Hi everyone,

I have an update for this issue.

Challenge:
If you are facing the following error:
[17.11.2020 10:06:54] <45> Error Guid should contain 32 digits with 4 dashes (xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). (System.FormatException)

Cause:
VMware vCloud Director 10.1.2 stores information wrong in it´s database, filling out a UUID field with a URL.

You can verify this with the following SQL request on the vCD database:
select name,uuid from virtual_center;
There should be a GUID number. Instead you will get a UUID.

VMware BugId: 2680344

Resolution:
VMware has fixed the issue in vCloud Director 10.2.2 and 10.3.
The fix will not correct the wrong entries, so that you could see the same error message above as well with newer vCD versions if the entries where created with the affected vCD versions.

1) Upgrade your vCD (together with it´s database) to 10.2.2, 10.3 or later.
2) Reconnect the affected vCenter in vCD. It will update the wrong entry.

VMware shared with us that they do not intent to fix this in a 10.1.x based version.

[d3nz]

Andreas,
What's about support vCD 10.3 from Veeam?

[Andreas Neufert]

As with any other platform release, we target to support newer vCD versions within 90 days. vCD 10.3 was just released on July 15th.

[akoznov]

Still no news about support of 10.3 Cloud Director?
We found that there is Veeam KB4193 and it contains info about recommendations to upgrade to 10.2.2 or 10.3. Veeam still doesn't support 10.3 or we can schedule an upgrade of sites to fix a lot of Cloud Director issues with that upgrade, but we can't because the main backup solution (VBR) could be broke in case of the unsupporting latest version of Cloud Director.

[Gostev]

The release vehicle for 10.3 support will be version 11a, which we are planning to release next month.

[benhaeloim]

Some news regarding the new version 11a? Any scheduled date? Thanks in advance

[Gostev]

We're starting the Early Availability phase for 11a next week.