SnapDrive problem - unable to find API

[SEJ]

Hi everyone,

We are currently deploying a new backup solution for our service, for a better render of Hyper-V backups and I had some trouble with SnapDrive. Note that the snapdrive account has admin rights on my netapp filer.

The problem occurs when I want to create a LUN with the "Create Disk Wizard", the first step is to add the Storage System so I put the storage system name then I press "Add" and bam I always get the same error "Failed to retrieve the root volume name from the storage system (xxxxxx). Error code : 13005. Error description :  Unable to find API : volume-get-root-name". I already checked on my storage system that the root volume exists and he does. 

So I'm requesting your help guys, maybe be some of you already experienced that.

Thanks in advance

[ghislanzoni]

Hi,

if you are in Cdot environment in the storage controller info of the SnapDrive you must enter the ip/user about the SVM that you are tring to log in.

So if you have a management lif of the SVM you must put in that ip.

Please tell me if this works successfully.

Regards

Roberto

[SEJ]

Yes, you're right we are in a clustered environment, I already registered those informations in SnapDrive but nothing changes.  And we don't have management LIFs, so i'm kind of blocked 😕

[ghislanzoni]

Hi,

you can try to follow these documents to do the correct procedure:

https://library.netapp.com/ecm/ecm_download_file/ECMP1804825

https://kb.netapp.com/support/index?page=content&id=1014912&actp=RSS

Let me know

Regards

Roberto

PS. Pay attention to the correct username vsadmin and not admin

PS2. Also the PTR record in DNS is needed

[georgevj]

This could be because of a BUG - 786228

More details here.

https://kb.netapp.com/support/index?page=content&id=2019826

If you cannot access this, please open a support case and refer to this KB ID 2019826.

The procedure involves privileged commands that has to be executed under guidance from NetApp support.

[SEJ]

---

@georgevj wrote:

This could be because of a BUG - 786228

More details here.

https://kb.netapp.com/support/index?page=content&id=2019826

If you cannot access this, please open a support case and refer to this KB ID 2019826.

The procedure involves privileged commands that has to be executed under guidance from NetApp support.

---

Hi,

thanks for your answer, unfortunately when I tried to open a case, it says that I can't because of my contract expired.

@ghislanzoni

I followed both of the procedures and I always got the same message.

[georgevj]

Oh. That sounds sad, but I hope you can still get the contract renewed.?

However, just to make sure that this is the issue, please try the following commands from cluster shell: (NB: login to the node management interface of each node separately and run these commands on every node.)

CLUSTER1::> set diag

CLUSTER1::*> debug smdb table cluster_cj_status show
create-join can-switch complete message                   cluster-uuid                         sitelist-uuid
----------- ---------- -------- ------------------------- ------------------------------------ ------------------------------------
join        no         no       "Past point of no return" ce9d7980-c799-11e2-9fcd-123478563412 ce9d7980-c799-11e2-9fcd-123478563412

If you see a 'no' under the "complete" header, this is the issue. Then you will have to rejoin that node back to the cluster. I would not post the instructions here, since it has to be run in realtime, seeing the response of each command. Especially if this is under production. 

[SEJ]

---

@georgevj wrote:

Oh. That sounds sad, but I hope you can still get the contract renewed.?

However, just to make sure that this is the issue, please try the following commands from cluster shell: (NB: login to the node management interface of each node separately and run these commands on every node.)

CLUSTER1::> set diag

CLUSTER1::*> debug smdb table cluster_cj_status show
create-join can-switch complete message                   cluster-uuid                         sitelist-uuid
----------- ---------- -------- ------------------------- ------------------------------------ ------------------------------------
join        no         no       "Past point of no return" ce9d7980-c799-11e2-9fcd-123478563412 ce9d7980-c799-11e2-9fcd-123478563412

If you see a 'no' under the "complete" header, this is the issue. Then you will have to rejoin that node back to the cluster. I would not post the instructions here, since it has to be run in realtime, seeing the response of each command. Especially if this is under production. 

---

Thanks for the fast answer, so I run the command over all nodes separately and fortunately I got "yes" on all nodes, but unfortunately it doesn't tell us the problem though.

[ghislanzoni]

Do you have already completed the documents that I 've posted in the first answers?

I think that the solution is inside it.

Regards

Roberto

[SEJ]

---

@ghislanzoni wrote:

Do you have already completed the documents that I 've posted in the first answers?

I think that the solution is inside it.

Regards

Roberto

---

Yes, the vsadmin is unlocked, the data LIF and management LIF are correctly configured, here's the firewall policy :

[ghislanzoni]

How you configure the integration, of the SVM, in the SnapDrive Transport Protocol Settings?

Can you post a screenshot please.

Regards

Roberto 

[SEJ]

---

@ghislanzoni wrote:

How you configure the integration, of the SVM, in the SnapDrive Transport Protocol Settings?

Can you post a screenshot please.

Regards

Roberto 

---

I use the admin account because for some reasons the dedicated user created for this purpose didn't reach the storage system. 

[ghislanzoni · Accepted solution]

I think the problem is here.

You must use the vsadmin user of the SVM where you want to create the LUN and the name of the SVM that is resolved by the DNS to the management LIF of the SVM.

If this step is not completed it doesn't work.

Regards

Roberto 

[SEJ]

---

@ghislanzoni wrote:

I think the problem is here.

You must use the vsadmin user of the SVM where you want to create the LUN and the name of the SVM that is resolved by the DNS to the management LIF of the SVM.

If this step is not completed it doesn't work.

Regards

Roberto 

---

Thanks a lot man, that solved the problem.

[georgevj]

Then its a different issue.. will try to see if there is any other possibilities.