Can we configure only a VIP for the corporate interface and no VIP for the Acquisition interface for a failover system?
[EDIT]
This would mean that you only need to have your corporate interfaces in the same subnet and can have acquisition interfaces in different subnets without VIP.
Hi Michiel,
TLDR: Yes
In a failover system, you need as a minimum one shared virtual IP (or hostname).
It’s not mandatory to configure it on all interfaces.
Hello all
I think the question is not that clear, from what i understood what they want to configure is the following, a Failover with Both Corporate and Aquisition, but they will be in different Datacenters and the Aquisition will be in different Sub Nets, and no Aquisition for the Virtual, depending on the site active the IP for Aquisition will be different
Site A
Corporate – 192.168.168.xxx
Aquisition – 10.201.10.xxx
Site B
Corporate – 192.168.168.xxx
Aquisition – 10.230.11.xxx
Virtual Dataminer 2
Corporate – 192.168.168.xxx
Aquisition – Not defined
Is this supported ?
Hello Ive
Thanks for the clarification but the problem here is that the Corporate network does not have access to the Devices only the Aquisition, not sure if i was clear with this detail.
So … not sure what to do, is the configuration proposed usable ?
What i can see as a disavantage is that depending on the active machine a different IP will “talk” to the devices, and this can cause issues with some devices, but i dont even know if Dataminer will accept this type of configuration and if was ever tested, not sure what issue it can cause.
AFAIK the VIP is mostly used for clients to access the system. They need to be in the same subnet as they need to be able to send ARP messages to see who has the VIP. The communication towards the devices should be initiated by DM and can go over the acquisition IF, so I wouldn’t expect any issues not having a VIP as it is DM that gives the IP to the devices. Just wanted to verify if you can configure it.
FYI, DM will pick the IF to communicate to your device depending on your routing table. When initiating a communication layer it will provide the IP (and MAC) of the IF used by the DM machine, so the device knows where it can send answers to. When you assign a VIP to your IF, you basically have two IPs on that IF, meaning that other devices can use both of those IPs to send messages to that network IF.
Thanks a lot for the help to all
you can configure a 2 VIPs, one for both interface
I also misunderstood the question there.
VIPs are used for two reasons.
– giving access to clients, typically on corporate network
– as a target IP for traps, typically on acquisition network.
You can work around this by sending the traps to both main and backup server
Hi Tiago,
In this case, you can use the corporate IP addresses (which are in the same subnet) to setup a virtual IP.
The users / other agents in the cluster can use this VIP to connect to the dataminer system.
There is no need for a VIP on the acquisition network, as long as both can access the same end devices and data sources.