Veeam 11a issues with Dell SC integration

[squebel]

Hello,

I'm having issues getting backups to integrate properly with our Dell SC array. I have previously set this up in other environments without issue and I've done everything the same way as before. The storage array is defined and rescanned, the proxy server(s) are set up properly with access to the iscsi network for access to the array, transport mode has been changed on the proxies I want to use, job has been set to use storage snaps and the particular proxy I want it to use, but I continue to get a message about "not being able to find a suitable proxy server". One thing I have noticed that is different is the server object/initiator that Veeam has created in older versions is not showing up in the SC array configuration; a different "rescan" initiator is but not any others. I have opened case 05290301 regarding this but wanted to see if anyone knew of something else before I wait for response from support.

[rennerstefan]

Hi

I think you already did the right thing opening a ticket with our support. If no initiator shows up support should check.

The message you mention always means either a connectivity or access issue.

Thanks

[squebel]

I don't think it's either. Everything is set up properly and all access is there that it needs. Scan are good and it even creates that one server object named "severname_Rescan". I just tested an integrated backup to an Infinidat array and it works perfectly fine with the same proxy server. Something is busted in the SC integration.

[Andreas Neufert]

Do you use iSCSI or FC ?

[rennerstefan]

... Everything is set up properly and all access is there that it needs. .... Something is busted in the SC integration.

Yea I was just saying that usually I see this error with communication and access issues such as wrong IPs being used or other network/fc related topics.
Unfortunately its hard to assist in such a case on the forum that's why having support involved is very good.
And as far as I can see in the ticket support is already on it.

Feel free to add a entry here once it is solved so others may find help in case they run into the same situation.

[squebel]

Based off what support has showed me, it appears that Veeam is attempting to use a different fault domain on the SC than the proxy is configured to use. The proxy is configured with access to fault domains on vlan 220 and 230 but Veeam keeps attempting communication on the proxy to fault domains on vlans 3100 and 3150. Is there any way to control this?

[Origin 2000]

I have a working Dell SC and Veeam backup from storage snapshot within a iSCSI environment. For sure i have 2 FD and i always add the proxy in a manual way to connect to the compellent. When checked later there was always a 2nd. "Server" object which looks like one created by veeam. Looks weird but works.

We will upgrade to veeam 11a later this month.

Regards,
Joerg

[squebel]

But do you have multiple iscsi fault domains on your SC array? This is what it's tripping up on.

[Origin 2000]

Two FDs.

[Andreas Neufert]

Veeam Backup & Replication uses 2 iSCSI initiators during Proxy processing.
The Microsoft Windows iSCSI initiator and an own one.
Usually the hosts entries are autocreated for this but there is only so much logic that you can do arround this and I guess in your cases here it is created in the wrond FD.

Potentially do the following:
Note the Veeam Proxy Server IQNs (both) from the hosts entries created within the storage.
Remove the hosts entry and recreate them manually in the correct FD. Or if this is possible move them into the other FD.

The Windows iSCSI initiator IQN can be started within the Windows menue (search for iSCSI).
The Veeam iSCSI initiator IQN can be found after first rescan on the Veeam Backup Server in one of the log files.
C:\ProgramData\Veeam\Backup\ScanRescan\Util.SanSnapshot.xxx.log
You can search there for a LUN name from your specific storage or just search for the iqn with the veeam name in it.
The above method by checking the existing entries for it is the simplest way.

[QoX8x]

Also want to mention I'm having issues with this on version 11a. My version 11 test box seems to work fine.

Case: #02444018

[rennerstefan]

@QoX8x thanks for sharing this info.
Can you please check the case number. Seems to be the wrong one.
I'd like to have a look to it.

Thanks

[foggy]

There's a chance that Veeam B&R gets incorrect IP-LUN mapping data from the plug-in/storage for some reason. If this is the case, support will escalate the case and our R&D will find out if there's an issue on the Veeam B&R or the plug-in side.

[fastide]

Looks like we have the same problem with 11a and our Dell SC3020 with four fault domains.
It fails to do a rescan with the error: VMware Backup Proxy Error: An item with the same key has already been added
In the Util.SanRescan.*** log I see that it looks like it is trying to connect on the wrong fault domains.

[foggy]

Hi Martijn, do you have a support case open for this?

[fastide]

Yes, it is Case 05486242