VCM not provisioning NIC's correctly

[TSM1]

Greetings

I wish to share a problem I am having in the hopes a solution can be found. 

On a freshly imaged VC module, I have attached the uplinks, have created an ethernet network and created a server profile for the BL460c blades in my BladeSystem c7000 g3 Enclosure. Following this, I have booted a previously working Centos 7.6 system, that previously had 2 NIC's available, enumerated as eno49 and 50. 

Upon booting Centos, I am immediately presented with messages such as:

bnx2x - BAD MCP validity signature
bnx2x - FW failed to respond. 
Unregistered net device.... etc

The command "ip a" only shows the lo interface. 

 
Finally, running lshw -c network, shows all controllers but no interaces:

lshw -c network
*-network:0 UNCLAIMED
description: Ethernet controller
product: BCM57840 NetXtreme II 10/20-Gigabit Ethernet
vendor: Broadcom Inc. and subsidiaries
physical id: 0
bus info: pci@0000:06:00.0
version: 11
width: 64 bits
clock: 33MHz
capabilities: pm vpd msix pciexpress cap_list
configuration: latency=0

*-network:1 UNCLAIMED
description: Ethernet controller
product: BCM57840 NetXtreme II 10/20-Gigabit Ethernet
vendor: Broadcom Inc. and subsidiaries
physical id: 0
bus info: pci@0000:06:00.1
version: 11
width: 64 bits
clock: 33MHz
capabilities: pm vpd msix pciexpress cap_list
configuration: latency=0

All the way to number 6 unclaimed controllers... On an idential working system, I show network interfaces 0 and 1 only

This is being managed on a VC FlexFabric-20/40 F8 Module (4.80 FW).

Please let me know if I can provide any further information, or if there are any ideas. Thanks!

[rmay_bk]

E-fuse the blade.  Or, if you have physical access to the enclosure, remove the blade and re-insert it.

[Suvamay]

Hi,

How many servers are present in the enclosure?

Is this the only server you are facing the issue?

What is the NIC status in the VCM ?

Try unassigned and reassign a new profile.

Try efuse the blade.

Thanks

[TSM1]

Thanks for the replies. Based on your suggestion, I have tripped the e-fuse, but the situation remains the same: the blade is unable to enumerate its networking hardware.

I should mention there are 16 blades in the enclosure and this is affecting 2. I have attempted adding and removing these blades from profiles, tripping the e-fuse, attempting a network boot, and even attempting to install NIC drivers from HPE website (which also fails because card is not detected). I should also mention, booting from ISO of different OS, NIC's are also not detected. 

In the VCM when viewing the networking hardware of the individual blade, the information shown is the same as any working blade in the enclosure:

Server Ethernet Adapter InformationEthernet Adapter Flex NIC Location Module Port Model MAC Address Network Allocated VFs

Port 1 Max. Virtual Functions: 64, VF Increment: 8

LOM1:1-a FLB1 Bay 1:d1:v1 HP FlexFabric 20Gb 2-port 630FLB Adapter XX-XX-XX-XX-XX-XX Network_ 64
LOM1:1-b FLB1 Bay 1:d1:v2 HP FlexFabric 20Gb 2-port 630FLB Adapter XX-XX-XX-XX-XX-XX
LOM1:1-c FLB1 Bay 1:d1:v3 HP FlexFabric 20Gb 2-port 630FLB Adapter XX-XX-XX-XX-XX-XX
LOM1:1-d FLB1 Bay 1:d1:v4 HP FlexFabric 20Gb 2-port 630FLB Adapter XX-XX-XX-XX-XX-XX
Port 2 Max. Virtual Functions: 64, VF Increment: 8

LOM1:2-a FLB1 Bay HP FlexFabric 20Gb 2-port 630FLB Adapter XX-XX-XX-XX-XX-XX 0
LOM1:2-b FLB1 Bay HP FlexFabric 20Gb 2-port 630FLB Adapter XX-XX-XX-XX-XX-XX
LOM1:2-c FLB1 Bay HP FlexFabric 20Gb 2-port 630FLB Adapter XX-XX-XX-XX-XX-XX
LOM1:2-d FLB1 Bay HP FlexFabric 20Gb 2-port 630FLB Adapter



Thanks for your help!

[Suvamay]

Hi,

When you say NIC card is not detected, can you check by swapping the server from this bay to any other bay bay where server is working. If that does not work can you remove swap the NIC card to rule out the hardware issue.

Let me know the outcome.

Thanks