VMware Cloud Community
hernandez80
Contributor
Contributor
Jump to solution

ESXi 4.1 - HP Blades and NFS mounts

  Hello,

I'm having a strange issue regarding NFS access

We are currently using HP Blades BL680c G5 servers, in C7000 enclosures. Cisco Switch configuration (3120x) are the same for all ports / ESXs.

I'm using VMNIC0 and VMNIC1 in etherchannel / IP Hash for providing access to NFS datastores on NetApp filers.

I now have two servers side by side that I want to invert for test reasons. These two servers are just moving to the place of the other.

Once rebooted, both of these hosts ar not seeing the datastores anymore. They appear as inactive. I can't reconnect them (VI client or esxcfg-nas -r)

vmkping to the Netapp filers doesn't work anymore also.

As soon as i put back these 2 servers in their original slot, everything's back to normal.

On cisco switch side, ports are still configured the same. None of them are shut down.

I'm missing something, but don't know where to search .

Thanks in advance

0 Kudos
1 Solution

Accepted Solutions
bulletprooffool
Champion
Champion
Jump to solution

Hi Hernandez,

Are you repatching the hosts or are they still using the same Network cables as before (perhaps an IP change, or an export rule on the NFS host)

remember, you'll have policies created on the NFS mounts to 'export' those mounts to specific hostnames, or IPs - if you are repatching and changing IP etc you may be causing this export to be invalidated,

One day I will virtualise myself . . .

View solution in original post

0 Kudos
3 Replies
bulletprooffool
Champion
Champion
Jump to solution

Hi Hernandez,

Are you repatching the hosts or are they still using the same Network cables as before (perhaps an IP change, or an export rule on the NFS host)

remember, you'll have policies created on the NFS mounts to 'export' those mounts to specific hostnames, or IPs - if you are repatching and changing IP etc you may be causing this export to be invalidated,

One day I will virtualise myself . . .
0 Kudos
hernandez80
Contributor
Contributor
Jump to solution

Hi,

No repatch, not any modification to the hosts. just move them from on slot to the other. So network "cables" can be considered as change, but the configuration of the switch ports are exactly the same.

All ips are the same

Working environment :

Blade 1 in slot 1 - VMnic0 and VMnic1 in etherchannel on NFS Vlan

Blade 2 in slot 2 - VMnic0 and VMnic1 in etherchannel on NFS Vlan

If Blade 2 goes in slot 1 and blade 1 in slot 2, i have inactive datastore and no vmkping...

Strange

0 Kudos
hernandez80
Contributor
Contributor
Jump to solution

All right, forget about it.

I fell dumb but the network switches were not configured exactly the same.

On one set, i had a vlan tagging, and on the other i had the vlan set in access mode . Configuration was not exactly the same !!!

Thanks for reading 🙂

0 Kudos