problem with nfs ls latency during high IO

[Judith Flo Gaya]

Hello,

Just wanted to let you know that I got answers from the linux-nfs list. 
It seems that at some point the code from the patch that was provided to 
fix the problem was removed between the 2.6.32 version and 2.6.35.

They suggest me to file a bug to bugzilla[1]

By the moment they suggest me to modify the variable from the kernel 
vm_dirty_ratio and lower it, so that the amount of data that the client 
waits to send to the nfs is less, indeed it improves the performance of 
the ls, I don't know yet about the possible collateral-effects.

Let's see what happens.
Thanks anyway for your help!
j

[1] https://bugzilla.redhat.com/show_bug.cgi?id=688232