> You may want to use NFSv3 over TCP instead.  NFS over UDP has a bit of
> an uneven history.

I agree, but from some reason we couldn't just leave well enough alone on this 
one.  The mystery here was too much...

> Even when (or _especially_ when) the DF bit is set, it's expected and
> acceptable to have the _sender_ emitting fragments.  The DF bit is an
> instruction to intermediate routers -- "please don't fragment this _any
> further_" -- it is not something that the destination host or any
> intermediate node has any business whatsoever attempting to "verify."

This is logical.  We'll probably turn this "feature" off.

>> 1) Why does Solaris set the DF bit on all packets, even fragmented packets?  
>> This seems odd to me.
>
>Path MTU discovery.  See RFC 1191.
>
>Most modern systems do this.

Solaris's behavior in setting the DF flag on all packets appears to be somewhat 
unique.

IP Flags from a packet trace copying a file off a RHEL 5 NFS server via udp:
IP:   Flags = 0x2 (May Fragment, More Fragments)
IP:   Flags = 0x2
...
IP:   Flags = 0x2
IP:   Flags = 0x0 (May Fragment, last Fragment)

Repeat with AIX 6.1 nfs server:
IP:   Flags = 0x2 (May Fragment, More Fragments)
IP:   Flags = 0x2
...
IP:   Flags = 0x2
IP:   Flags = 0x0 (May Fragment, last Fragment)

And Solaris: 
IP:   Flags = 0x6 (Don't Fragment, More Fragments)
IP:   Flags = 0x6
...
IP:   Flags = 0x6
IP:   Flags = 0x4 (Don't Fragment, last fragment)

-Alex
_______________________________________________
networking-discuss mailing list
[email protected]

Reply via email to