Commit 8a4c6e19 authored by J. Bruce Fields's avatar J. Bruce Fields

nfsd: document kernel interfaces for nfsd configuration

These are only needed by nfs-utils.  But I needed to remind myself how
they worked recently and thought this might be helpful.  It's short and
incomplete for now as I was only interested in startup, shutdown, and
configuration of listening sockets.
Signed-off-by: default avatarJ. Bruce Fields <>
parent 65b2e665
Administrative interfaces for nfsd
Note that normally these interfaces are used only by the utilities in
nfsd is controlled mainly by pseudofiles under the "nfsd" filesystem,
which is normally mounted at /proc/fs/nfsd/.
The server is always started by the first write of a nonzero value to
Before doing that, NFSD can be told which sockets to listen on by
writing to nfsd/portlist; that write may be:
- an ascii-encoded file descriptor, which should refer to a
bound (and listening, for tcp) socket, or
- "transportname port", where transportname is currently either
"udp", "tcp", or "rdma".
If nfsd is started without doing any of these, then it will create one
udp and one tcp listener at port 2049 (see nfsd_init_socks).
On startup, nfsd and lockd grace periods start.
nfsd is shut down by a write of 0 to nfsd/threads. All locks and state
are thrown away at that point.
Between startup and shutdown, the number of threads may be adjusted up
or down by additional writes to nfsd/threads or by writes to
For more detail about files under nfsd/ and what they control, see
fs/nfsd/nfsctl.c; most of them have detailed comments.
Implementation notes
Note that the rpc server requires the caller to serialize addition and
removal of listening sockets, and startup and shutdown of the server.
For nfsd this is done using nfsd_mutex.
Markdown is supported
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment