Skip to contentSkip to conversation

Fix prctl privilege escalation (CVE-2006-2451)

PATCH Marcel Holtmann 2 participants 3 messages Jul 12, 2006

lore
Share & tools
lore
#0 lore raw

Hi Linus,

attached is the fix with full explanation for CVE-2006-2451. It fixes a possible privilege escalation through the prctl() system call.

I also put Michael Kerrisk on CC, because the manual page of prctl() needs adjustment. The value 2 for the PR_SET_DUMPABLE flag is no longer valid after this patch. The only way to get root-owned core dumps is through /proc/sys/fs/suid_dumpable and the manual page should reflect that.

Regards

Marcel

[PATCH] Fix prctl privilege escalation and suid_dumpable (CVE-2006-2451)

Based on a patch from Ernie Petrides

During security research, Red Hat discovered a behavioral flaw in core dump handling. A local user could create a program that would cause a core file to be dumped into a directory they would not normally have permissions to write to. This could lead to a denial of service (disk consumption), or allow the local user to gain root privileges.

The prctl() system call should never allow to set "dumpable" to the value 2. Especially not for non-privileged users.

This can be split into three cases:

1) running as root -- then core dumps will already be done as root, and so prctl(PR_SET_DUMPABLE, 2) is not useful

2) running as non-root w/setuid-to-root -- this is the debatable case

3) running as non-root w/setuid-to-non-root -- then you definitely do NOT want "dumpable" to get set to 2 because you have the privilege escalation vulnerability

With case #2, the only potential usefulness is for a program that has designed to run with higher privilege (than the user invoking it) that wants to be able to create root-owned root-validated core dumps. This might be useful as a debugging aid, but would only be safe if the program had done a chdir() to a safe directory.

There is no benefit to a production setuid-to-root utility, because it shouldn't be dumping core in the first place. If this is true, then the same debugging aid could also be accomplished with the "suid_dumpable" sysctl.

---
commit a5c53fee00abefc2269ebf3fe05b53c05e9f389f
tree 871b99fbcced6db040c59fe2f3de6d14a9bf09fe
parent 7c3dec0679c66ce177726802adbe2f403942fc27
author Marcel Holtmann <[email protected]> Wed, 12 Jul 2006 12:44:51 +0200 committer Marcel Holtmann <[email protected]> Wed, 12 Jul 2006 12:44:51 +0200

 kernel/sys.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
kernel/sys.c +1 -1
index dbb3b9c..e236f98 100644
@@ -1983,7 +1983,7 @@ asmlinkage long sys_prctl(int option, un
 			error = current->mm->dumpable;
 			break;
 		case PR_SET_DUMPABLE:
-			if (arg2 < 0 || arg2 > 2) {
+			if (arg2 < 0 || arg2 > 1) {
 				error = -EINVAL;
 				break;
 			}
#1 lore raw

Already fixed since 07/06/2006:

http://www.kernel.org/git/?p=linux/kernel/git/stable/linux-2.6.17.y.git;a=commit;h=0af184bb9f80edfbb94de46cb52e9592e5a547b0
http://www.kernel.org/git/?p=linux/kernel/git/stable/linux-2.6.17.y.git;a=commitdiff;h=0af184bb9f80edfbb94de46cb52e9592e5a547b0;hp=52cbb7b78994ea3799f1bbb8c03bce1e2f72a271

Quoted 14 lines · Marcel Holtmann

Hi Linus,

attached is the fix with full explanation for CVE-2006-2451. It fixes a possible privilege escalation through the prctl() system call.

I also put Michael Kerrisk on CC, because the manual page of prctl() needs adjustment. The value 2 for the PR_SET_DUMPABLE flag is no longer valid after this patch. The only way to get root-owned core dumps is through /proc/sys/fs/suid_dumpable and the manual page should reflect that.

Regards

Marcel

Signature · 7 lines
Micskó Gábor
HP APS, AIS, ASE
Szintézis ZRt.
H-9023 Győr, Tihanyi Á. u. 2.
Tel: +36 96 502 216
Fax: +36 96 318 658
E-mail: [email protected]
#2 lore raw

Hi Gabor,

Quoted 4 lines

Already fixed since 07/06/2006:

http://www.kernel.org/git/?p=linux/kernel/git/stable/linux-2.6.17.y.git;a=commit;h=0af184bb9f80edfbb94de46cb52e9592e5a547b0
http://www.kernel.org/git/?p=linux/kernel/git/stable/linux-2.6.17.y.git;a=commitdiff;h=0af184bb9f80edfbb94de46cb52e9592e5a547b0;hp=52cbb7b78994ea3799f1bbb8c03bce1e2f72a271

it is fixed in the stable tree, but it is not fixed upstream. It happens that security fixes go into the stable tree, before the enter upstream.

Regards

Marcel