From: Jeff Layton <jlayton@poochiereds.net>
To: bfields@fieldses.org, trond.myklebust@primarydata.com
Cc: schumaker.anna@gmail.com, linux-nfs@vger.kernel.org
Subject: Re: [PATCH 0/4] nfs/nfsd/sunrpc: enforce requirement for congestion control protocols in NFSv4
Date: Thu, 23 Feb 2017 12:17:33 -0500 [thread overview]
Message-ID: <1487870253.3448.4.camel@poochiereds.net> (raw)
In-Reply-To: <20170223170337.10686-1-jlayton@redhat.com>
On Thu, 2017-02-23 at 12:03 -0500, Jeff Layton wrote:
> RFC5661 says:
>
> Where an NFSv4.1 implementation supports operation over the IP
> network protocol, any transport used between NFS and IP MUST be among
> the IETF-approved congestion control transport protocols.
>
> ...and RFC7530 has similar verbiage. The NFS server has never enforced
> this requirement, however, so a user could issue NFSv4 calls against
> the server via UDP.
>
> This patchset adds a small bit of infrastructure to the sunrpc layer
> to enforce this requirement, and has the nfs and nfsd layers set the
> appropriate flags for it. It also has knfsd skip registering a UDP
> port for NFSv4, using the same flags.
>
> Lightly tested by hand, but it's fairly straightforward.
>
> Jeff Layton (4):
> sunrpc: flag transports as using IETF approved congestion control
> protocols
> sunrpc: turn bitfield flags in svc_version into bools
> nfs/nfsd/sunrpc: enforce congestion control protocol requirement for
> NFSv4
> sunrpc: don't register UDP port with rpcbind when version needs
> congestion control
>
> fs/nfs/callback_xdr.c | 6 ++++--
> fs/nfsd/nfs2acl.c | 1 -
> fs/nfsd/nfs3acl.c | 1 -
> fs/nfsd/nfs4proc.c | 13 +++++++------
> include/linux/sunrpc/svc.h | 12 ++++++++----
> include/linux/sunrpc/svc_xprt.h | 1 +
> net/sunrpc/svc.c | 22 +++++++++++++++++++++-
> net/sunrpc/svcsock.c | 1 +
> net/sunrpc/xprtrdma/svc_rdma_transport.c | 2 ++
> 9 files changed, 44 insertions(+), 15 deletions(-)
>
I probably should have sent this as an RFC first. I'm not 100% clear on
whether PROG_MISMATCH is the right return code there.
Also, there is still a small wart after this patchset. The high/low
program versions reported look a little odd:
$ rpcinfo -T udp knfsdsrv nfs 4
rpcinfo: RPC: Program/version mismatch; low version = 3, high version = 4
program 100003 version 4 is not available
We could try to fix this and report different values depending on the
socket type, but I'm not sure I really care. AFAIK, this is just
informative anyway, and it's not _technically_ wrong. The server does
support version 4, just not the UDP socket where we sent the RPC ping.
Thoughts?
--
Jeff Layton <jlayton@poochiereds.net>
next prev parent reply other threads:[~2017-02-23 17:27 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-02-23 17:03 [PATCH 0/4] nfs/nfsd/sunrpc: enforce requirement for congestion control protocols in NFSv4 Jeff Layton
2017-02-23 17:03 ` [PATCH 1/4] sunrpc: flag transports as using IETF approved congestion control protocols Jeff Layton
2017-02-23 19:42 ` Tom Talpey
2017-02-23 20:00 ` Jeff Layton
2017-02-23 20:06 ` Tom Talpey
2017-02-23 20:11 ` J. Bruce Fields
2017-02-23 20:26 ` Jason Gunthorpe
2017-02-23 20:33 ` Tom Talpey
2017-02-23 20:55 ` Jason Gunthorpe
2017-02-24 15:08 ` Tom Talpey
2017-02-24 17:17 ` Jeff Layton
2017-02-24 18:03 ` Jason Gunthorpe
2017-02-23 20:32 ` Jeff Layton
2017-02-23 20:17 ` Chuck Lever
2017-02-23 20:15 ` Chuck Lever
2017-02-23 17:03 ` [PATCH 2/4] sunrpc: turn bitfield flags in svc_version into bools Jeff Layton
2017-02-23 17:03 ` [PATCH 3/4] nfs/nfsd/sunrpc: enforce congestion control protocol requirement for NFSv4 Jeff Layton
2017-02-23 17:03 ` [PATCH 4/4] sunrpc: don't register UDP port with rpcbind when version needs congestion control Jeff Layton
2017-02-23 17:17 ` Jeff Layton [this message]
2017-02-24 18:25 ` [PATCH v2 0/4] nfs/nfsd/sunrpc: enforce NFSv4 transport requirements Jeff Layton
2017-02-24 18:25 ` [PATCH v2 1/4] sunrpc: turn bitfield flags in svc_version into bools Jeff Layton
2017-02-24 18:25 ` [PATCH v2 2/4] sunrpc: flag transports as having both reliable and ordered delivery, and congestion control Jeff Layton
2017-02-24 18:25 ` [PATCH v2 3/4] nfs/nfsd/sunrpc: enforce transport requirements for NFSv4 Jeff Layton
2017-02-24 18:25 ` [PATCH v2 4/4] sunrpc: don't register UDP port with rpcbind when version needs congestion control Jeff Layton
2017-02-24 18:38 ` [PATCH v2 0/4] nfs/nfsd/sunrpc: enforce NFSv4 transport requirements Chuck Lever
2017-02-24 18:53 ` Jeff Layton
2017-02-24 21:23 ` J. Bruce Fields
2017-02-24 18:53 ` Tom Talpey
2017-02-24 21:22 ` J. Bruce Fields
2017-02-24 21:25 ` J. Bruce Fields
2017-02-24 21:34 ` Jeff Layton
2017-02-24 21:44 ` J. Bruce Fields
2017-02-27 11:59 ` Jeff Layton
2017-02-27 12:08 ` Tom Talpey
2017-02-27 12:55 ` Jeff Layton
2017-02-27 14:20 ` J. Bruce Fields
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1487870253.3448.4.camel@poochiereds.net \
--to=jlayton@poochiereds.net \
--cc=bfields@fieldses.org \
--cc=linux-nfs@vger.kernel.org \
--cc=schumaker.anna@gmail.com \
--cc=trond.myklebust@primarydata.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).