iSCSI Initiator Config

[JeffJD]

I'm looking at a vSphere iSCSI setup that a vendor created utilizing a single host and 2 SANs.  They have used port binding to create a single iSCSI initiator for 2 vmnics and the setup looks correct on the vSphere configuration side (1 nic per switch, no teaming, same subnet).

The problem (I believe) is they have direct connected the 2 physical NICs to the SANs.  There is no switch and therefore no connectivity between the individual SANs or between the host adapters.  Ignoring the obvious lack of redundancy isn't there an issue with port binding the 2 vmkernels together when they are technically on completely separate networks?  Shouldn't I be seeing errors in the logs regarding one of the paths being down?  Or have issues when scanning for datastores?

Here's a simplified diagram of the setup:

Thanks for any insight :smileyhappy:

[Nick_Andreev]

You have only one path to each of the datastores. So technically none of the paths are down. But this is obviously not the best practice.

How many paths do you see when go to the datastore properties?