Start a Conversation

Unsolved

M

24 Posts

2633

November 27th, 2018 12:00

Presenting Group/Volumes to New ESXi host

Hello Dell community.

I currently have a new EqualLogic Group where I have set up our new vSphere 6.5 Host.

I need to be able to connect to our old EqualLogic Group so I can share the datastore that contains the virtual machines between the old esxi and the new environment.

Then I can un-register the vm from the old esxi and register it to the new esxi 6.5 and from their I can migrate them to a new datastores on the new EqualLogic Group.

Thank you

1 Rookie

 • 

1.5K Posts

November 27th, 2018 14:00

Hello, 

 This is very easy to do.  I do it all the time in my lab.   With ESXi I prefer using CHAP credentials on the volumes. 

So if I add new servers I don't need to update the Volume Access Control list. 

For ESXiv5 servers I created a CHAP account on the EQL array "esxv5-chap"    For 6.5 I used "esxv65-chap"  

 So to add the ESXi v6.5 servers, you would create a CHAP USER 'esxv65-chap' and set a password 12 characters min.  

 Then on the ESXi servers, add the discovery address of the new group.  And add in the CHAP information 'esxv65-chap' in the Discovery Address CHAP credentials. 

 Run rescan on your ESXi servers and those volumes will show up as Datastores.  So you can do your un-register, re-register.  If the old servers are ESXi v5 you might also want to move them to VMFS v6 formatted Datastore for best performance.   Then if you empty and old VMFS v5.x Datastore you can re-format it with VMFS v6 and re-use the volume.  

 Regards, 

Don 

 

24 Posts

November 30th, 2018 08:00

Don,

Thanks for all the info.  I was able to do everything but when I rescan it does not show the new group.

When I look at the EqualLogic Event Log I can see where the new ESXi host tries to connect but then errors out with...

Session was Closed

iSCSI initiator connection failure

Connection was closed by peer

 

1 Rookie

 • 

1.5K Posts

November 30th, 2018 08:00

Hello, 

 Are they all on the same subnet/switches? 

 Also make sure that all the ESXi servers are configured for best practices. 

http://downloads.dell.com/solutions/storage-solution-resources/BestPracticesWithPSseries-VMware%28TR1091%29.pdf

Regards,

Don 

24 Posts

November 30th, 2018 09:00

I just realized that the error message is listing the IP 10.0.0.128 (eth3 of one of the SANs) and not the group IP of 10.0.0.151

24 Posts

November 30th, 2018 09:00

They are all on the same network/swtiches but different VLANS.

 

Severity Date and Time Member ID Message
-------- --------------------- ------- ------------------------ --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Info 11/30/2018 9:39:27 AM ICSSAN2 7.2.15 | 7.2.24 | 7.2.27 iSCSI session to target '10.0.0.128:3260, iqn.2001-05.com.equallogic:0-8a0906-720ad0008-9740015a02e5130b-dphtos01' from initiator '10.1.201.13:14694, iqn.1998-01.com.vmware:prvm01-62f45332' was closed. | iSCSI initiator connection failure. | Connection was closed by peer.

1 Rookie

 • 

1.5K Posts

November 30th, 2018 10:00

Hello, 

That's completely normal.  No logins ever occur on the Group IP address.  EQL uses an iSCSI feature called "Login redirect"   It returns an error to the initiator and the IP address of the port to use for the actual login. 

 When the array balances connections later, it uses another iSCSI command "async logout" to tell the initiator to logout but come right back to the Discovery Address.  Then login redirect sends it to the new port. 

 Regards,

Don 

No Events found!

Top