P. Larry Nelson wrote:
From what I've gleaned about the two protocols from googling, it appears that TCP has advantages on a lossy network but that's not our scenario. It also is not a stateless protocol, like UDP, so if a server crashes in the middle of a packet transmission, the client will hang and filesystems will need to be unmounted and remounted. So it would seem UDP is better, at least in our case.
I've found a lot of the information on the net regarding NFS parameters to be fairly inconsistent (or possibly outdated). Some of the tuning guides do recommend using tcp, others suggest udp. Unfortunately Redhat's own documentation gives no useful; guidance on this, though I hope they had some good reason to change the default. I've left our SL systems using tcp without any noticeable performance hit, and I can't say I've noticed any client hangs either. OTOH I did recently find we still had explicit "rsize=8192,wsize=8192" values in our nfs mounts - probably from SL3 days! Bumping these to more current values (32768) made a huge performance difference.
Graham -- ------------------------------------------------------------------------- Graham Allan - I.T. Manager School of Physics and Astronomy - University of Minnesota -------------------------------------------------------------------------
