[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

deep mounting patch

L.S.

Is it possible to include the attached patch in the debian kernel patch.
We use a lot of debian NFS clients 20 servers and 200 workstations
with home directories mounted from a solaris NFS server via the
automounter. There was a problem with secure authentication between
solaris server and linux client. We delivered a patch for this which  was
included in the kernel around 2.6.16. The standard 2.6.18 kernels for
etch work fine in this respect.

Unfortunately the relevant part of the kernel was rewritten around
2.2.20, reintroducing the problem. We  discovered the problem when
2.6.24-etchnhalf was introduced. Our debian clients on which we
installed 2.6.24-etchnhalf could not mount the home directories anymore.
We made a new patch for it and got this accepted upstream. It will be
included in the 2.7.27 kernels (see attachment).

I understand that lenny will be released with kernel 2.6.26.
This will be a major regression from the etch 2.6.18 kernels for us, unless
this patch can be included in de debian kernel patch.

regards,

Gert Huisman

-- 
Gert Huisman                       Systeem programmeur    IT-groep FEW
email:     gert@few.vu.nl          Vrije Universiteit Amsterdam
telefoon:  +31 20 5987806          De Boelelaan 1081a,    kamer: S4.14
fax:       +31 20 5987653          1081 HV  Amsterdam, the Netherlands
--- Begin Message --- 
Kennelijk staat het ook op de rol voor 2.6.27

Ed
--- Begin Message --- 
Most of the following patches are bugfixes:
  There are a couple of spin locking fixes to follow up the BKL removal.
  There are some attribute update fixups, mainly to make the spin
  locking more efficient
  A fix for the problem of autobind on cloned RPC clients

The one feature that has been added is the lookup revalidation mount
option to allow clients to specify that they might not want to cache
negative dentries, and that they might want strict dentry revalidation.

-- 
Trond Myklebust
Linux NFS client maintainer

NetApp
Trond.Myklebust@netapp.com
www.netapp.com
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

--- End Message ---
--- Begin Message --- 
From: EG Keizer <keie@few.vu.nl>

Allow mount to do authenticated mounts below the root of the exported tree.
The wording in RFC 2623, sec 2.3.2. allows fsinfo with UNIX authentication
on the root of the export. Mounts are not always done on the root
of the exported tree. Especially autoumounts often mount below the root of
the exported tree.
Some server implementations (justly) require full authentication for the
so-called deep mounts. The old code used AUTH_SYS only. This caused deep
mounts to fail on systems requiring stronger authentication..
The client should try both authentication types and use the first one that
succeeds.
This method was already partially implemented. This patch completes
the implementation for NFS2 and NFS3.
This patch was developed to allow Debian systems to automount home directories
on Solaris servers with krb5 authentication.

Tested on kernel 2.6.24-etchnhalf.1

Signed-off-by: E.G. Keizer <keie@few.vu.nl>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---

 fs/nfs/nfs3proc.c |   20 ++++++++++++++++++--
 fs/nfs/proc.c     |   10 ++++++++--
 2 files changed, 26 insertions(+), 4 deletions(-)

diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index 1e750e4..c55be7a 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -699,7 +699,7 @@ nfs3_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle,
 }
 
 static int
-nfs3_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
+do_proc_fsinfo(struct rpc_clnt *client, struct nfs_fh *fhandle,
 		 struct nfs_fsinfo *info)
 {
 	struct rpc_message msg = {
@@ -711,11 +711,27 @@ nfs3_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
 
 	dprintk("NFS call  fsinfo\n");
 	nfs_fattr_init(info->fattr);
-	status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0);
+	status = rpc_call_sync(client, &msg, 0);
 	dprintk("NFS reply fsinfo: %d\n", status);
 	return status;
 }
 
+/*
+ * Bare-bones access to fsinfo: this is for nfs_get_root/nfs_get_sb via
+ * nfs_create_server
+ */
+static int
+nfs3_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
+		   struct nfs_fsinfo *info)
+{
+	int	status;
+
+	status = do_proc_fsinfo(server->client, fhandle, info);
+	if (status && server->nfs_client->cl_rpcclient != server->client)
+		status = do_proc_fsinfo(server->nfs_client->cl_rpcclient, fhandle, info);
+	return status;
+}
+
 static int
 nfs3_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
 		   struct nfs_pathconf *info)
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index 4dbb84d..1934652 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -65,14 +65,20 @@ nfs_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
 
 	dprintk("%s: call getattr\n", __func__);
 	nfs_fattr_init(fattr);
-	status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0);
+	status = rpc_call_sync(server->client, &msg, 0);
+	/* Retry with default authentication if different */
+	if (status && server->nfs_client->cl_rpcclient != server->client)
+		status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0);
 	dprintk("%s: reply getattr: %d\n", __func__, status);
 	if (status)
 		return status;
 	dprintk("%s: call statfs\n", __func__);
 	msg.rpc_proc = &nfs_procedures[NFSPROC_STATFS];
 	msg.rpc_resp = &fsinfo;
-	status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0);
+	status = rpc_call_sync(server->client, &msg, 0);
+	/* Retry with default authentication if different */
+	if (status && server->nfs_client->cl_rpcclient != server->client)
+		status = rpc_call_sync(server->nfs_client->cl_rpcclient, &msg, 0);
 	dprintk("%s: reply statfs: %d\n", __func__, status);
 	if (status)
 		return status;

--- End Message ---

--- End Message ---
Reply to: