After certain failure modes of an NFS mount, an NFS client should send
a MOUNTPROC_UMNT request to remove the just-added mount entry from the
server's mount table. While no-one should rely on the accuracy of the
server's mount table, this is just being a good neighbor.
Since NFS mount processing is handled in the kernel now, we will need
a function in the kernel's mountd client that can post a MOUNTRPC_UMNT
request, in order to handle these failure modes.
Signed-off-by: Chuck Lever <[email protected]>
---
fs/nfs/internal.h | 1 +
fs/nfs/mount_clnt.c | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 61 insertions(+), 0 deletions(-)
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index fa9a0d0..d3823d7 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -102,6 +102,7 @@ struct nfs_mount_request {
};
extern int nfs_mount(struct nfs_mount_request *info);
+extern int nfs_umount(const struct nfs_mount_request *info);
/* client.c */
extern struct rpc_program nfs_program;
diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index b5ed325..67148e4 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -218,6 +218,52 @@ out_mnt_err:
goto out;
}
+/**
+ * nfs_umount - Notify a server that we have unmounted this export
+ * @info: pointer to umount request arguments
+ *
+ * MOUNTPROC_UMNT is advisory, so we set a short timeout.
+ */
+int nfs_umount(const struct nfs_mount_request *info)
+{
+ static const struct rpc_timeout nfs_umnt_timeout = {
+ .to_initval = 1 * HZ,
+ .to_maxval = 3 * HZ,
+ .to_retries = 2,
+ };
+ struct mountres result;
+ struct rpc_message msg = {
+ .rpc_argp = info->dirpath,
+ .rpc_resp = &result,
+ };
+ struct rpc_clnt *clnt;
+ int status;
+
+ clnt = nfs_mount_rpc_create(info, &nfs_umnt_timeout);
+ if (IS_ERR(clnt))
+ return PTR_ERR(clnt);
+
+ dprintk("NFS: sending UMNT request for %s:%s\n",
+ (info->hostname ? info->hostname : "server"),
+ info->dirpath);
+
+ if (info->version == NFS_MNT3_VERSION)
+ msg.rpc_proc = &clnt->cl_procinfo[MOUNTPROC3_UMNT];
+ else
+ msg.rpc_proc = &clnt->cl_procinfo[MOUNTPROC_UMNT];
+
+ status = rpc_call_sync(clnt, &msg, 0);
+ rpc_shutdown_client(clnt);
+
+ if (status < 0) {
+ dprintk("NFS: UMNT request failed, status=%d\n", status);
+ return status;
+ }
+
+ dprintk("NFS: UMNT request succeeded\n");
+ return 0;
+}
+
/*
* XDR encode/decode functions for MOUNT
*/
@@ -416,6 +462,13 @@ static struct rpc_procinfo mnt_procedures[] = {
.p_statidx = MOUNTPROC_MNT,
.p_name = "MOUNT",
},
+ [MOUNTPROC_UMNT] = {
+ .p_proc = MOUNTPROC_UMNT,
+ .p_encode = (kxdrproc_t)mnt_enc_dirpath,
+ .p_arglen = MNT_enc_dirpath_sz,
+ .p_statidx = MOUNTPROC_UMNT,
+ .p_name = "UMOUNT",
+ },
};
static struct rpc_procinfo mnt3_procedures[] = {
@@ -428,6 +481,13 @@ static struct rpc_procinfo mnt3_procedures[] = {
.p_statidx = MOUNTPROC3_MNT,
.p_name = "MOUNT",
},
+ [MOUNTPROC3_UMNT] = {
+ .p_proc = MOUNTPROC3_UMNT,
+ .p_encode = (kxdrproc_t)mnt_enc_dirpath,
+ .p_arglen = MNT_enc_dirpath_sz,
+ .p_statidx = MOUNTPROC3_UMNT,
+ .p_name = "UMOUNT",
+ },
};
On Mon, 2009-07-13 at 15:17 -0400, Chuck Lever wrote:
> After certain failure modes of an NFS mount, an NFS client should send
> a MOUNTPROC_UMNT request to remove the just-added mount entry from the
> server's mount table. While no-one should rely on the accuracy of the
> server's mount table, this is just being a good neighbor.
>
> Since NFS mount processing is handled in the kernel now, we will need
> a function in the kernel's mountd client that can post a MOUNTRPC_UMNT
> request, in order to handle these failure modes.
>
> Signed-off-by: Chuck Lever <[email protected]>
> ---
>
> fs/nfs/internal.h | 1 +
> fs/nfs/mount_clnt.c | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 61 insertions(+), 0 deletions(-)
>
> diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
> index fa9a0d0..d3823d7 100644
> --- a/fs/nfs/internal.h
> +++ b/fs/nfs/internal.h
> @@ -102,6 +102,7 @@ struct nfs_mount_request {
> };
>
> extern int nfs_mount(struct nfs_mount_request *info);
> +extern int nfs_umount(const struct nfs_mount_request *info);
>
> /* client.c */
> extern struct rpc_program nfs_program;
> diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
> index b5ed325..67148e4 100644
> --- a/fs/nfs/mount_clnt.c
> +++ b/fs/nfs/mount_clnt.c
> @@ -218,6 +218,52 @@ out_mnt_err:
> goto out;
> }
>
> +/**
> + * nfs_umount - Notify a server that we have unmounted this export
> + * @info: pointer to umount request arguments
> + *
> + * MOUNTPROC_UMNT is advisory, so we set a short timeout.
> + */
> +int nfs_umount(const struct nfs_mount_request *info)
> +{
> + static const struct rpc_timeout nfs_umnt_timeout = {
> + .to_initval = 1 * HZ,
> + .to_maxval = 3 * HZ,
> + .to_retries = 2,
> + };
These values will never work with a TCP mount. Shouldn't we just enforce
this as UDP only?
> + struct mountres result;
> + struct rpc_message msg = {
> + .rpc_argp = info->dirpath,
> + .rpc_resp = &result,
> + };
> + struct rpc_clnt *clnt;
> + int status;
> +
> + clnt = nfs_mount_rpc_create(info, &nfs_umnt_timeout);
> + if (IS_ERR(clnt))
> + return PTR_ERR(clnt);
Why are we returning a status value here? I see no reason why we should
care if a umount call works or failed. It is certainly not going to
affect anything the client does.
> +
> + dprintk("NFS: sending UMNT request for %s:%s\n",
> + (info->hostname ? info->hostname : "server"),
> + info->dirpath);
> +
> + if (info->version == NFS_MNT3_VERSION)
> + msg.rpc_proc = &clnt->cl_procinfo[MOUNTPROC3_UMNT];
> + else
> + msg.rpc_proc = &clnt->cl_procinfo[MOUNTPROC_UMNT];
> +
> + status = rpc_call_sync(clnt, &msg, 0);
> + rpc_shutdown_client(clnt);
> +
> + if (status < 0) {
> + dprintk("NFS: UMNT request failed, status=%d\n", status);
> + return status;
> + }
> +
> + dprintk("NFS: UMNT request succeeded\n");
> + return 0;
> +}
ditto concerning return values. Please just make this a void function.