HP NCU NFT reports Failed Rx Path

[Lmm_1]

You should be able to use NFT, thats the most basic teaming type, the only requirement will be have both ports under same broadcast domain, if you are getting RX paath validation it could be related with the stacking or vlan config on the switches, have you try to put that port to the same physical switch?

[Lieven Verdonck]

No I have not tried to put the two cables in the same switch, i will try this.

Is it necessary to make a trunk on the nortelswitch if I want to activate NFT (or SLB) on the HP-tool on the server? If not, what is the advantage of making a trunk (better balancing, I presume)?

[Lmm_1]

No I have not tried to put the two cables in the same switch, i will try this.

Is it necessary to make a trunk on the nortelswitch if I want to activate NFT (or SLB) on the HP-tool on the server? If not, what is the advantage of making a trunk (better balancing, I presume)?

Basically NFT doesnt require any switch config nor intelligence, you can do it using a regular hub, with NFT you will get 1GB TX/RX and the other NIC will be standby, just waiting for NIC1 failure.

When using SLB, it will require the Nortel Trunking or Cisco Ethercahnnel, in this case you will get both NICs TX/RX, basically 2 GB load balancing and failover.

rgds.
lmm

[James Prendergast]

I'm still getting this error. We use Cisco 2950s at the access layer.

[Lieven Verdonck]

My error has been fixed thanks to the suggestion of LMM (I use NFT without Trunks now).

[James Prendergast]

We have the interfaces in an NFT team with each NIC connected to a separate switch. This way if a NIC or a switch fails we are okay.

[Lmm_1]

It looks like the same as I mentioned at the begining, the NIC is not getting the heartbeat, you need to make sure the port on each switch belong to the same broadcast domain and then make sure you got communication for tha VLAN switch to switch, then you must get the heartbeat frame.

rgds,
lmm