On Oct 14, 2008, at Oct 14, 2008, 5:54 AM, Neil Brown wrote:
On Monday October 13, chuck.lever@xxxxxxxxxx wrote:
NFSD start-up code makes two separate and unconditional calls to
lockd_up(): one for UDP and one for TCP. So whether or not NFSD
actually honors the "-T" and "-U" flags, lockd certainly does not
honor them in current (unpatched) code.
Not exactly correct. The calls are not unconditional.
I assume we are looking at nfsd_init_socks in nfssvc.c
Note the condition at the top:
if (!list_empty(&nfsd_serv->sv_permsocks))
return 0;
If a UDP port has previously been created by a write to "portlist",
then when the first thread is stared, no new sockets will be created,
and nfsd (and lockd) will just listen on UDP.
So your patches do make a real functional change which is more than
just "always listen on UDP".
OK, I'm already collecting a set of additional fixes and cleanups in
this area for 2.6.29. We can add this to the list.
Why are there two separate and undocumented kernel interfaces for
starting NFSD?
I agree that the man page should reflect reality.
Several people have said that, but no-one has suggested what specific
parts of the man page are problematic.
--
Chuck Lever
chuck[dot]lever[at]oracle[dot]com
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
