Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

Is it possible to enforce a maximum rsize/wsize in nfsd? I'm having issues related to IP fragmentation (yes, I'm stuck with NFS-over-UDP, contrary to the warnings in the manpage), and have no practical access to the client mount command (buried in one of many TFTP boot images).

http://nfs.sourceforge.net/nfs-howto/ar01s05.html lists a kernel source parameter limiting the maximum block size, but I'm not gong to get away with recompiling the nfsd kernel module so that's not really an option either :-(

share|improve this question
    
So I realised I probably should have posted this on ServerFault instead of SuperUser... is crossposting acceptable? –  paulw1128 Apr 13 '12 at 18:21

1 Answer 1

I would think IP fragmentation would need to be addressed by the MTU parameter on the NICs. I'm not even aware of a way to adjust the UDP packet size for just NFS even if you had control of both sides.

share|improve this answer
    
We're limited to the 'standard' 1500 byte MTU size. What I would love to have is a way for the NFS daemon to tell the client "Yeah sure you requested an 8K rsize, but you are actually only getting 1K blocks" (and thus avoid fragmentation). Sadly I'm not sure that this exists :( –  paulw1128 Apr 12 '12 at 23:02

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.