Array Setup and Networking
1827948 Members
3665 Online
109973 Solutions
New Discussion

Live editing Network settings Question

 
SOLVED
Go to solution
Darkzadow
Occasional Contributor

Live editing Network settings Question

Hello,

I have a HF40 already up and running with many (80) production systems on it. Connected to an 5 Host VMware cluster. For reasons unkown, at commisioning, they only used 2 out of the 4 available ISCSI ports. I would like to add the unused ports to my iscsi interfaces. Will this tank my system? 

Thank you,

BrandonScreenshot 2023-04-11 090423.png

2 REPLIES 2
Sheldon Smith
HPE Pro
Solution

Re: Live editing Network settings Question

It should be no problem at all. "Any networking port can be used for any purpose."
Assuming the ESXi hosts have the latest HPE Storage Connection Manager (NCM) for VMware installed on each of them, they will simply have more array data ports from which the discovery port can draw.
You may want to contact Nimble Support and have a support engineer in a virtual room while you enable the two remaining ports.

That's an interesting port layout. HPE Best Practices would typically have eth0A and eth0B both doing management, preferably using two switches. In such a layout, all ports ending with "A" or "C" would go to switch 1, and all "B" or "D" to switch 2. This spreads the communications in a way to reduce or eliminate single points of failure. Of course if everything is connected to a single switch, then it's academic.


Note: While I am an HPE Employee, all of my comments (whether noted or not), are my own and are not any official representation of the company

Accept or Kudo

Darkzadow
Occasional Contributor

Re: Live editing Network settings Question

Hello Sheldon,

The ISCSI go to a redundant dedicated 2 switch ISCSI stack and the Management go to a seperate redundant 2 switch Management stack. Both are so isolated that they could realistically be merged.

The thought process was that they wanted to protect against Nic card (built in or 4 port Nic) failure and controller failure, and network equipment failure since our systems interface with heavy production equipment.  The only reason I dont have a second nimble is that it was not in the budget.

All Hosts are running NCM 7.0.2. When i tried this on the site that was being commissioned I saw high latency but no system drop out, but do not know if that was just reporting blip as connections and network were adjusted.

 

Thank you for your reply. Ill ask HPE directly and post results (most likely in a week while an equipment stand down can be scheduled)