Skip to contentSkip to conversation

nfsd patches for 2.6.29

PATCH New J. Bruce Fields 4 participants 7 messages Mar 18, 2009

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

The following bugfixes are also available from the for-2.6.29 git repository at:

git://linux-nfs.org/~bfields/linux.git for-2.6.29

The CAP_MKNOD change should really go into CAP_FS_MASK, but that will be a user-visible change, and some more deliberation may be required to be sure we have CAP_FS_MASK exactly right; so I prefer to go ahead with the one obvious nfsd-specific change now.

--b.

Benny Halevy (1):
NFSD: provide encode routine for OP_OPENATTR

J. Bruce Fields (1):
nfsd: nfsd should drop CAP_MKNOD for non-root

 fs/nfsd/nfs4xdr.c          |    1 +
 include/linux/capability.h |    6 ++++--
 2 files changed, 5 insertions(+), 2 deletions(-)
#1 lore raw

PATCH 1/1 NFSD: provide encode routine for OP_OPENATTR

From: Benny Halevy <[email protected]>

Although this operation is unsupported by our implementation we still need to provide an encode routine for it to merely encode its (error) status back in the compound reply.

Thanks for Bill Baker at sun.com for testing with the Sun OpenSolaris' client, finding, and reporting this bug at Connectathon 2009.

This bug was introduced in 2.6.27

---
 fs/nfsd/nfs4xdr.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)
fs/nfsd/nfs4xdr.c +1 -0
index f65953b..9250067 100644
@@ -2596,6 +2596,7 @@ static nfsd4_enc nfsd4_enc_ops[] = {
 	[OP_LOOKUPP]		= (nfsd4_enc)nfsd4_encode_noop,
 	[OP_NVERIFY]		= (nfsd4_enc)nfsd4_encode_noop,
 	[OP_OPEN]		= (nfsd4_enc)nfsd4_encode_open,
+	[OP_OPENATTR]		= (nfsd4_enc)nfsd4_encode_noop,
 	[OP_OPEN_CONFIRM]	= (nfsd4_enc)nfsd4_encode_open_confirm,
 	[OP_OPEN_DOWNGRADE]	= (nfsd4_enc)nfsd4_encode_open_downgrade,
 	[OP_PUTFH]		= (nfsd4_enc)nfsd4_encode_noop,
Signature · 1 line
1.6.0.4
#2 lore raw

PATCH 1/1 nfsd: nfsd should drop CAP_MKNOD for non-root

From: J. Bruce Fields <[email protected]>

Since creating a device node is normally an operation requiring special privilege, Igor Zhbanov points out that it is surprising (to say the least) that a client can, for example, create a device node on a filesystem exported with root_squash.

So, make sure CAP_MKNOD is among the capabilities dropped when an nfsd thread handles a request from a non-root user.

---
 include/linux/capability.h |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)
include/linux/capability.h +4 -2
index 1b98725..4864a43 100644
@@ -393,8 +393,10 @@ struct cpu_vfs_cap_data {
 # define CAP_FULL_SET     ((kernel_cap_t){{ ~0, ~0 }})
 # define CAP_INIT_EFF_SET ((kernel_cap_t){{ ~CAP_TO_MASK(CAP_SETPCAP), ~0 }})
 # define CAP_FS_SET       ((kernel_cap_t){{ CAP_FS_MASK_B0, CAP_FS_MASK_B1 } })
-# define CAP_NFSD_SET     ((kernel_cap_t){{ CAP_FS_MASK_B0|CAP_TO_MASK(CAP_SYS_RESOURCE), \
-					CAP_FS_MASK_B1 } })
+# define CAP_NFSD_SET     ((kernel_cap_t){{ CAP_FS_MASK_B0 \
+					    | CAP_TO_MASK(CAP_SYS_RESOURCE) \
+					    | CAP_TO_MASK(CAP_MKNOD), \
+					    CAP_FS_MASK_B1 } })
 
 #endif /* _KERNEL_CAPABILITY_U32S != 2 */
 
Signature · 1 line
1.6.0.4
Acked-by on v1 1/1
#3 lore raw

Re: [PATCH] nfsd: nfsd should drop CAP_MKNOD for non-root

Quoted 7 lines · J. Bruce Fields

From: J. Bruce Fields <[email protected]>

Since creating a device node is normally an operation requiring special privilege, Igor Zhbanov points out that it is surprising (to say the least) that a client can, for example, create a device node on a filesystem exported with root_squash.

Quoted 6 lines

So, make sure CAP_MKNOD is among the capabilities dropped when an nfsd thread handles a request from a non-root user.

I assume CAP_LINUX_IMMUTABLE simply does not apply to nfs?

And, you're adding CAP_FS_MASK_B1 in anticipation of labeled nfs?

Though, I was going to send a patch later today or tomorrow (figure I should do some ltp testing) adding CAP_MKNOD to the whole CAP_FS_MASK_B0 (and CAP_LINUX_IMMUTABLE and CAP_FS_MASK_B1 to CAP_FS_SET). That will conflict with this one.

thanks,
-serge

Quoted 23 lines
---
 include/linux/capability.h |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)
include/linux/capability.h +4 -2
index 1b98725..4864a43 100644
@@ -393,8 +393,10 @@ struct cpu_vfs_cap_data {
 # define CAP_FULL_SET     ((kernel_cap_t){{ ~0, ~0 }})
 # define CAP_INIT_EFF_SET ((kernel_cap_t){{ ~CAP_TO_MASK(CAP_SETPCAP), ~0 }})
 # define CAP_FS_SET       ((kernel_cap_t){{ CAP_FS_MASK_B0, CAP_FS_MASK_B1 } })
-# define CAP_NFSD_SET     ((kernel_cap_t){{ CAP_FS_MASK_B0|CAP_TO_MASK(CAP_SYS_RESOURCE), \
-					CAP_FS_MASK_B1 } })
+# define CAP_NFSD_SET     ((kernel_cap_t){{ CAP_FS_MASK_B0 \
+					    | CAP_TO_MASK(CAP_SYS_RESOURCE) \
+					    | CAP_TO_MASK(CAP_MKNOD), \
+					    CAP_FS_MASK_B1 } })
 
 #endif /* _KERNEL_CAPABILITY_U32S != 2 */
 
Signature · 1 line
1.6.0.4
#4 lore raw

Re: [PATCH] nfsd: nfsd should drop CAP_MKNOD for non-root

Quoted 11 lines · Serge E. Hallyn
Quoted 10 lines · J. Bruce Fields

From: J. Bruce Fields <[email protected]>

Since creating a device node is normally an operation requiring special privilege, Igor Zhbanov points out that it is surprising (to say the least) that a client can, for example, create a device node on a filesystem exported with root_squash.

So, make sure CAP_MKNOD is among the capabilities dropped when an nfsd thread handles a request from a non-root user.

Quoted 7 lines
Quoted 3 lines

I assume CAP_LINUX_IMMUTABLE simply does not apply to nfs?

Right. We shouldn't care how it's set.

Quoted 1 line

And, you're adding CAP_FS_MASK_B1 in anticipation of labeled nfs?

That's unchanged (would have been clearer if I hadn't re-line-wrapped in the same patch).

Quoted 4 lines

Though, I was going to send a patch later today or tomorrow (figure I should do some ltp testing) adding CAP_MKNOD to the whole CAP_FS_MASK_B0 (and CAP_LINUX_IMMUTABLE and CAP_FS_MASK_B1 to CAP_FS_SET). That will conflict with this one.

OK, feel free to revert this at that point if necessary.

--b.

Quoted 27 lines

thanks,
-serge

Quoted 23 lines
---
 include/linux/capability.h |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)
include/linux/capability.h +4 -2
index 1b98725..4864a43 100644
@@ -393,8 +393,10 @@ struct cpu_vfs_cap_data {
 # define CAP_FULL_SET     ((kernel_cap_t){{ ~0, ~0 }})
 # define CAP_INIT_EFF_SET ((kernel_cap_t){{ ~CAP_TO_MASK(CAP_SETPCAP), ~0 }})
 # define CAP_FS_SET       ((kernel_cap_t){{ CAP_FS_MASK_B0, CAP_FS_MASK_B1 } })
-# define CAP_NFSD_SET     ((kernel_cap_t){{ CAP_FS_MASK_B0|CAP_TO_MASK(CAP_SYS_RESOURCE), \
-					CAP_FS_MASK_B1 } })
+# define CAP_NFSD_SET     ((kernel_cap_t){{ CAP_FS_MASK_B0 \
+					    | CAP_TO_MASK(CAP_SYS_RESOURCE) \
+					    | CAP_TO_MASK(CAP_MKNOD), \
+					    CAP_FS_MASK_B1 } })
 
 #endif /* _KERNEL_CAPABILITY_U32S != 2 */
 
Signature · 1 line
1.6.0.4
#5 lore raw

Re: [PATCH] nfsd: nfsd should drop CAP_MKNOD for non-root

That's good and I'm glad to see patch for CAP_NFSD_MASK in git. Thanks. :-) Waiting for CAP_FS_MASK to be fixed too.

By the way, I don't see git repository for 2.4.x kernel. Could you fix 2.4.x too?

Acked-by on v1 1/1
#6 lore raw

Re: [PATCH] nfsd: nfsd should drop CAP_MKNOD for non-root

In reply to J. Bruce Fields
Quoted 7 lines · J. Bruce Fields

From: J. Bruce Fields <[email protected]>

Since creating a device node is normally an operation requiring special privilege, Igor Zhbanov points out that it is surprising (to say the least) that a client can, for example, create a device node on a filesystem exported with root_squash.

Quoted 6 lines

So, make sure CAP_MKNOD is among the capabilities dropped when an nfsd thread handles a request from a non-root user.

Quoted 29 lines
---
 include/linux/capability.h |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)
include/linux/capability.h +4 -2
index 1b98725..4864a43 100644
@@ -393,8 +393,10 @@ struct cpu_vfs_cap_data {
 # define CAP_FULL_SET     ((kernel_cap_t){{ ~0, ~0 }})
 # define CAP_INIT_EFF_SET ((kernel_cap_t){{ ~CAP_TO_MASK(CAP_SETPCAP), ~0 }})
 # define CAP_FS_SET       ((kernel_cap_t){{ CAP_FS_MASK_B0, CAP_FS_MASK_B1 } })
-# define CAP_NFSD_SET     ((kernel_cap_t){{ CAP_FS_MASK_B0|CAP_TO_MASK(CAP_SYS_RESOURCE), \
-					CAP_FS_MASK_B1 } })
+# define CAP_NFSD_SET     ((kernel_cap_t){{ CAP_FS_MASK_B0 \
+					    | CAP_TO_MASK(CAP_SYS_RESOURCE) \
+					    | CAP_TO_MASK(CAP_MKNOD), \
+					    CAP_FS_MASK_B1 } })
 
 #endif /* _KERNEL_CAPABILITY_U32S != 2 */
 
Signature · 6 lines
1.6.0.4

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Signature · 2 lines
James Morris
<[email protected]>