Re: [BUG?] Maybe NFS bug since 2.6.37 on SPARC64

[Lukas Razik <linux@razik.name>]

> Jim Rees <rees@umich.edu> wrote:

> Lukas Razik wrote:
> 
>   > So, once again...
>   > 
>   > What kind of switch is this client connected to, and how is that switch
>   > configured?
>   
>   I can't answer this question because I have no physical access to the
>   hardware.  Hence I think it's logical to stop here until the nodes are
>   connected directly (without any hardware between them).  I'll write here
>   as soon as I know if it works...
>   
>   For the rest I thank all contributors for the fast help!
> 
> I think the implication here is that you're plugged in to a port that's
> running spanning tree (stp) and is not set to portfast.  Such a port can
> take up to a minute to start forwarding packets.  If so, you should fix
> that, but it would also be nice if nfsroot could be made to work in this
> setup.
> 
> By the way, if you want to send me that machine with its 64 processors and
> 32GB of memory, I'd be glad to investigate.
>

Hello again!

I've connected the systems directly (without a switch between them).
It seems that the problem came from the switch because now the newer kernels also mount their NFSROOTs during start up without problems.

Assuming that the ports of the switch really run in STP mode - two questions:
1. Is it unusual that the ports of a switch where computers are connected to run in STP mode?
2. If that's not unusual: Wouldn't it be useful if the new linux kernels could mount an NFSROOT even when the system is plugged into an STP port?

The problem is that we'll have 16 such systems which can't be connected directly - they must communicate through the switch. And I've no access to the switch.

Best regards,
Lukas