Skip to contentSkip to conversation

ceph: give up on paths longer than PATH_MAX

PATCH In review Ilya Dryomov 3 participants 3 messages Jan 7, 2025 – Jan 9, 2025 Stable branch

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 ceph: give up on paths longer than PATH_MAX

From: Max Kellermann <[email protected]>

commit 550f7ca98ee028a606aa75705a7e77b1bd11720f upstream.

If the full path to be built by ceph_mdsc_build_path() happens to be longer than PATH_MAX, then this function will enter an endless (retry) loop, effectively blocking the whole task. Most of the machine becomes unusable, making this a very simple and effective DoS vulnerability.

I cannot imagine why this retry was ever implemented, but it seems rather useless and harmful to me. Let's remove it and fail with ENAMETOOLONG instead.

---
 fs/ceph/mds_client.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)
fs/ceph/mds_client.c +4 -5
index 11289ce8a8cc..dfa1b3c82b53 100644
@@ -2713,12 +2713,11 @@ char *ceph_mdsc_build_path(struct ceph_mds_client *mdsc, struct dentry *dentry,
 
 	if (pos < 0) {
 		/*
-		 * A rename didn't occur, but somehow we didn't end up where
-		 * we thought we would. Throw a warning and try again.
+		 * The path is longer than PATH_MAX and this function
+		 * cannot ever succeed.  Creating paths that long is
+		 * possible with Ceph, but Linux cannot use them.
 		 */
-		pr_warn("build_path did not end path lookup where expected (pos = %d)\n",
-			pos);
-		goto retry;
+		return ERR_PTR(-ENAMETOOLONG);
 	}
 
 	*pbase = base;
Signature · 1 line
2.46.1
maintainer Reviewed-by Alex Markuze on v1 1/1
#1 lore raw

[ Sasha's backport helper bot ]

Hi,

The upstream commit SHA1 provided is correct: 550f7ca98ee028a606aa75705a7e77b1bd11720f

WARNING: Author mismatch between patch and upstream commit:
Backport author: Ilya Dryomov<[email protected]> Commit author: Max Kellermann<[email protected]>

Status in newer kernel trees:
6.12.y | Present (different SHA1: 99a37ab76a31)
6.6.y | Present (different SHA1: 82dfe5074a06)

Note: The patch differs from the upstream commit:
---
1: 550f7ca98ee0 ! 1: 092e42aac21a ceph: give up on paths longer than PATH_MAX
@@ Metadata
## Commit message ##
ceph: give up on paths longer than PATH_MAX

+ commit 550f7ca98ee028a606aa75705a7e77b1bd11720f upstream.
+
If the full path to be built by ceph_mdsc_build_path() happens to be longer than PATH_MAX, then this function will enter an endless (retry) loop, effectively blocking the whole task. Most of the machine
@@ Commit message

+ [[email protected]: backport to 6.6: pr_warn() is still in use]

## fs/ceph/mds_client.c ##
@@ fs/ceph/mds_client.c: char *ceph_mdsc_build_path(struct ceph_mds_client *mdsc, struct dentry *dentry,
@@ fs/ceph/mds_client.c: char *ceph_mdsc_build_path(struct ceph_mds_client *mdsc, s
+ * cannot ever succeed. Creating paths that long is
+ * possible with Ceph, but Linux cannot use them.
*/
-- pr_warn_client(cl, "did not end path lookup where expected (pos = %d)\n",
-- pos);
+- pr_warn("build_path did not end path lookup where expected (pos = %d)\n",
+- pos);
- goto retry;
+ return ERR_PTR(-ENAMETOOLONG);
}
---

Results of testing on various branches:

| Branch                    | Patch Apply | Build Test |
|---------------------------|-------------|------------|
| stable/linux-6.6.y        |  Success    |  Success   |
#2 lore raw
In reply to Ilya Dryomov
Quoted 20 lines · Ilya Dryomov

From: Max Kellermann <[email protected]>

commit 550f7ca98ee028a606aa75705a7e77b1bd11720f upstream.

If the full path to be built by ceph_mdsc_build_path() happens to be longer than PATH_MAX, then this function will enter an endless (retry) loop, effectively blocking the whole task. Most of the machine becomes unusable, making this a very simple and effective DoS vulnerability.

I cannot imagine why this retry was ever implemented, but it seems rather useless and harmful to me. Let's remove it and fail with ENAMETOOLONG instead.

Quoted 3 lines
---
 fs/ceph/mds_client.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

Thank you, I've dropped the "large" ceph patches from 6.6.y now and added this one instead.

greg k-h