[ofa-general] [NFS/RDMA] Can't mount NFS/RDMA partition

[Tziporet Koren]

Talpey, Thomas wrote:
> At 04:15 PM 12/17/2008, Tziporet Koren wrote:
>   
>> Talpey, Thomas wrote:
>>     
>>>>> If I try "ibstat" after that, I have a kernel panic message :
>>>>> "ibpanic: [4826] main: stat of IB device 'mlx4_0' failed: (Device 
>>>>>           
>> or resource
>>     
>>>>> busy)" because device is in use.
>>>>>       
>>>>>           
>> We should release FW 2.6.0 soon - so it will be worth to try
>>     
>>>> I can't explain this - certainly I've never seen it. I am going to 
>>>> guess it's an
>>>> OFED issue, or something in your setup. Do you have any other detail? Stack
>>>> trace of the oops?
>>>>
>>>>
>>>>     
>>>>         
>> Tom
>> Have you used our ConnectX cards when testing NFS/RDMA?
>>     
>
> Yes, I have and it works fine in the mode Bull is using. We did have
> some interoperability problems between ConnectX and mthca, but
> those were back in May, and fixed by Roland a short time later.
>
> In any case, the Bull kernel log message indicates NFS/RDMA has
> connected successfully, so I believe the problem lies elsewhere.
>
> Tom.
>
>   
Jack noticed that ib_mad1 uses 100% of the cpu in the bug report.
And the error -16 is -EBUSY, and this is returned if the command times 
out when using events.
Maybe the HCA or the switch are sending a flood of traps and then the 
HCA is busy handling all of them and does not complete the command on time?

Can you check you do not have errors on your line?

Tziporet