proc: block mounts below /proc/<pid>/net and on the links into /proc/self
Subsystems (7)
Christian Brauner
maintainercover proc: block mounts below /proc/<pid>/net and on the links into /proc/self
In Linux 6.12 I stopped allowing mounts on the entries below /proc/<pid>/fd/, /proc/<pid>/fdinfo/ and /proc/<pid>/map_files/ (cf. [1]). This extends that to two more groups of entries:
(1) Everything below /proc/<pid>/net/.
proc_net_d_revalidate() fails on every lookup since 1fde6f21d90f ("proc: fix /proc/net/* after setns(2)"). lookup_fast() answers that with d_invalidate() which detaches every mount on the dentry. That affects every mount namespace and includes locked mounts.
So a mask on /proc/<pid>/net/dev goes away the first time anybody looks the name up. The directory is 0555 so anybody really means anybody. This is reproducible with a container that masks /proc/<pid>/net/dev with a locked bind mount and a stat() by nobody in the initial namespaces makes the real file show up again. So it's pointless.
A mount that can't stick is better refused than taken away. I also have a version that makes proc_net_d_revalidate() check the network namespace instead so mounts there stick. But I think that's bullsh*t. It would make mounts on the net entries of other processes stick as well and the few users that try to mount there fail today anyway (see below).
(2) The links into /proc/self: /proc/self, /proc/thread-self, /proc/mounts, /proc/net and the afs and nfs links below /proc/fs/.
These resolve to a different directory for every process. A mount on the link itself is crossed by every lookup through it in that mount namespace, so it redirects all processes at once. Mounting through the links keeps working and lands on the caller's own entry as before.
What I really wanted was to block mounting on anything below /proc/<pid>/. That was the big hammer I mentioned in [1] years ago. We can't... I did a pretty exhaustive survey of who mounts on the entries below /proc/<pid>/ or on the links into it:
- GitHub code search and regex searches on Sourcegraph
- Debian Code Search over unstable
- the systemd units of every binary package in Debian 11, 12, 13 and sid, Ubuntu 22.04, 24.04 and 26.04, Fedora 43 and 44, CentOS Stream 10, openSUSE Leap 16 and Tumbleweed and Arch
- clones of 46 container runtimes and orchestrators, the sandboxes (systemd, bubblewrap, flatpak, firejail, snapd, nsjail, minijail, Chromium and Firefox), Android and the rooting tools on top of it, and the AppArmor and SELinux policies
- detection rules, issue trackers, lore, LWN and the Q&A sites
The legitimate users all mount on entries of the process that does the mount:
(1) systemd units that list /proc/self/... in ReadOnlyPaths=, ReadWritePaths= or ReadWriteDirectories=. systemd resolves /proc/self in the child of the service and binds the entry onto itself before it execs the service. If the mount fails the unit fails with 226/NAMESPACE and the "-" prefix doesn't help. That's netdata in Debian 11 and 12 and Ubuntu 22.04 and 24.04, opendht in Debian 12 and sid, Ubuntu 24.04 and Arch, accountsservice 22.x in Debian 12 and Ubuntu 22.04, OpenSearch 3, wazuh-indexer, TrueNAS SCALE, OONI's backend and the JupyterHub systemdspawner example. Of the 731 unit files on GitHub that name a /proc path in one of these directives 53 name an entry of the service itself.
(2) systemd's make_filesystem() binds /dev/null over /proc/self/mounts before it execs mkfs.btrfs so that it doesn't refuse a device whose other partitions are mounted. systemd-repart, systemd-homed and systemd-makefs use that.
(3) mkosi binds a fake /proc/self/mountinfo for grub-bios-setup.
(4) runc applies maskedPaths such as /proc/1/environ from the container's init when a config asks for it, bwrap does the same from its pid 1, firejail hides the name of its pid 1 and clampdown masks /proc/1/mem.
(5) Shell scripts that mount on /proc/$$/... and then exec, e.g., luci-app-shadowrt on OpenWrt.
(6) Tests in systemd, elogind, LXC, libpathrs and bun.
Fedora's SELinux policy even grants fsadm_t and init_t the permission to mount over their own entries because of (1) and (2). Sad, very sad...
Everything else I found is malware or offensive tooling:
Hiding a process by mounting over /proc/<pid> is catalogued as MITRE ATT&CK T1564.013 and used in the wild by TeamTNT, Commando Cat, the KONO DIO DA miner, KV-botnet, UNC2891 and kbotne. hackshell and emp3r0r ship it and Elastic's endpoint rule kills any process that does it.
So blocking all of /proc/<pid>/ would break units in released distributions and systemd itself. A rule that only lets a process mount on its own entries, and on a directory only a bind of itself, keeps (1) to (4) working, except for a few bwrap sandboxes that put a tmpfs over directories like /proc/1/task, and stops all of the attacks. I have that as well but it's not really a fix so we can delay it.
The remaining entries of a process only fail pid_revalidate() once the task is gone and proc_flush_pid() invalidates them when it is reaped. Their mounts go away together with the task and nothing gets revealed.
Nothing I found mounts on the links themselves and crun even has a test that expects it to fail. The handful of units that mask entries below /proc/<pid>/net/, e.g., Algo VPN's ReadOnlyPaths=/proc/net/pfkey for strongswan, already fail today.
systemd's own lookup invalidates the dentry before it can bind onto it. With this series they fail the same way. Apart from tests in libpathrs and securejoin that had to be adapted the only fallout from the 6.12 change we found is one small project that handed libmount's new mount api a /proc/self/fd/N target that move_mount() doesn't follow without MOVE_MOUNT_T_SYMLINKS. Ubuntu even backported the 6.12 change to their 22.04 and 24.04 kernels this year.
Nuke this crap.
[1]: https://lore.kernel.org/r/[email protected]
---
Christian Brauner (7):
proc: don't let an entry below /proc/<pid>/net be a mountpoint
proc: don't let /proc/self and /proc/thread-self be mountpoints
proc: don't let /proc/mounts and /proc/net be mountpoints
afs: create /proc/fs/afs with proc_self_symlink()
nfs: create the /proc/fs/nfsfs links with proc_self_symlink()
proc: explain which entries can't be mountpoints selftests/proc: test which entries can't be mountpoints
fs/afs/main.c | 2 +- fs/nfs/client.c | 4 +- fs/proc/generic.c | 31 ++++- fs/proc/internal.h | 18 ++- fs/proc/proc_net.c | 2 +- fs/proc/root.c | 2 +- fs/proc/self.c | 2 + fs/proc/thread_self.c | 2 + include/linux/proc_fs.h | 5 + tools/testing/selftests/proc/.gitignore | 1 + tools/testing/selftests/proc/Makefile | 1 + tools/testing/selftests/proc/proc-unmountable.c | 150 ++++++++++++++++++++++++ 12 files changed, 208 insertions(+), 12 deletions(-)
---
base-commit: 7be4bc52fd9e0e55dc9a2e0232b89427d2c4b614
change-id: 20261024-work-procfs-fixes-8b21a6d17448
Christian Brauner
maintainerPATCH 1/7 proc: don't let an entry below /proc/<pid>/net be a mountpoint
proc_net_d_revalidate() always returns 0 so that a dentry below /proc/<pid>/net is never reused after setns(). So lookup_fast() will end up calling d_invalidate(), which unhashes the dentry and detaches every mount on it, in every mount namespace; including locked mounts.
Then lookup_slow() creates a new dentry for the same entry. So a mount on /proc/<pid>/net/dev succeeds and goes away with the first lookup of the name by anybody with search permission on the directory. Given it's 0555 that's everyone's grandma. A mask on such a name hides nothing as the repro below shows.
Treat those entries like I did fd/, fdinfo/ and map_files/ and refuse to mount on them in the first place. It's nonsensical theater otherwise:
procnet net/dev stat
// SPDX-License-Identifier: GPL-2.0
/*
* procnet: cover an entry of /proc/<pid>/ with a bind mount, lock the mount,
* then look the name up.
*
* The first task stays alive, <pid> is its pid. The "container" task has its
* own mount namespace. There it binds a mask on /proc/<pid>/<name>. It drops to
* uid 1000 and enters a new user namespace with a new mount namespace, so that
* the mask is locked. Then one task looks the name up:
*
* stat the container calls stat() on the name * umount the container calls umount2(MNT_DETACH) on the name * nobody a task with uid 65534 in the first namespaces calls stat()
*
* The container lists its mounts before and after (that does not look the
* name up) and then reads the name. MASK is fine. Anything else means that
* the lookup uncovered the entry.
*
* usage: procnet <name> stat|umount|nobody
* name "status" is the control, "net/dev" is a file below net, "net/stat" is
* a directory below net
*/
#define _GNU_SOURCE
#include <dirent.h>
#include <errno.h>
#include <fcntl.h>
#include <sched.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <time.h>
#include <unistd.h>
#include <sys/mount.h>
#include <sys/prctl.h>
#include <sys/stat.h>
#include <sys/wait.h>
#define WORK "/run/pn"
static char target[128];
static void die(const char *what)
{
fprintf(stderr, "FATAL %s: %s\n", what, strerror(errno));
exit(2);
}
static void write_str(const char *path, const char *s, int flags)
{
int fd = open(path, O_WRONLY | flags, 0644);
if (fd < 0 || write(fd, s, strlen(s)) < 0)
die(path);
close(fd);
}
/* list the mounts below /proc without a lookup of the target */
static void mounts(const char *when)
{
char line[512];
FILE *f = fopen("/proc/self/mountinfo", "r");
int n = 0;
if (!f)
die("mountinfo");
while (fgets(line, sizeof(line), f)) {
if (!strstr(line, " /proc/"))
continue;
printf("container: %s: mount: %s", when, line);
n++;
}
fclose(f);
if (!n)
printf("container: %s: nothing is mounted below /proc\n", when);
}
static void show(const char *when, int isdir)
{
char buf[48] = "";
int fd;
if (isdir) {
struct dirent *de;
DIR *d = opendir(target);
int hit = 0;
if (!d) {
printf("container: %s: opendir(%s) = %s\n", when, target, strerror(errno));
return;
}
printf("container: %s: readdir(%s) =", when, target);
while ((de = readdir(d))) {
if (de->d_name[0] == '.')
continue;
printf(" %s", de->d_name);
hit |= strcmp(de->d_name, "MASK");
}
closedir(d);
printf("%s\n", hit ? " <-- HIT" : "");
return;
}
fd = open(target, O_RDONLY);
if (fd < 0) {
printf("container: %s: open(%s) = %s\n", when, target, strerror(errno));
return;
}
if (read(fd, buf, sizeof(buf) - 1) < 0)
die("read");
close(fd);
if (strchr(buf, '\n'))
*strchr(buf, '\n') = 0;
printf("container: %s: read(%s) = \"%s\"%s\n", when, target, buf,
strncmp(buf, "MASK", 4) ? " <-- HIT" : "");
}
int main(int argc, char **argv)
{
const char *name = argc > 1 ? argv[1] : "net/dev";
const char *op = argc > 2 ? argv[2] : "stat";
int isdir = !strcmp(name, "net/stat");
int to_host[2], to_cont[2];
struct stat st;
pid_t pid;
char c;
setvbuf(stdout, NULL, _IOLBF, 0);
snprintf(target, sizeof(target), "/proc/%d/%s", getpid(), name);
if (unshare(CLONE_NEWNS) || mount(NULL, "/", NULL, MS_REC | MS_PRIVATE, NULL))
die("unshare");
mkdir(WORK, 0755);
if (mount("W", WORK, "tmpfs", 0, NULL))
die("mount W");
write_str(WORK "/mask", "MASK", O_CREAT);
mkdir(WORK "/maskdir", 0755);
write_str(WORK "/maskdir/MASK", "MASK", O_CREAT);
if (pipe(to_host) || pipe(to_cont))
die("pipe");
pid = fork();
if (pid < 0)
die("fork");
if (!pid) {
if (unshare(CLONE_NEWNS))
die("container unshare");
if (mount(isdir ? WORK "/maskdir" : WORK "/mask", target, NULL, MS_BIND, NULL)) {
printf("container: bind the mask on %s: %s\n", target, strerror(errno));
exit(0);
}
printf("container: bind the mask on %s: ok\n", target);
if (setgid(1000) || setuid(1000) || prctl(PR_SET_DUMPABLE, 1))
die("setuid");
/* lock the mount: a new user namespace with a copy of the tree */
if (unshare(CLONE_NEWUSER | CLONE_NEWNS))
die("container userns");
write_str("/proc/self/setgroups", "deny", 0);
write_str("/proc/self/uid_map", "0 1000 1", 0);
write_str("/proc/self/gid_map", "0 1000 1", 0);
mounts("before");
errno = 0;
if (!strcmp(op, "stat")) {
stat(target, &st);
printf("container: stat(%s): %s\n", target, strerror(errno));
} else if (!strcmp(op, "umount")) {
umount2(target, MNT_DETACH);
printf("container: umount2(%s, MNT_DETACH): %s\n", target, strerror(errno));
} else if (write(to_host[1], "g", 1) != 1 || read(to_cont[0], &c, 1) != 1) {
die("sync");
}
mounts("after");
show("after", isdir);
exit(0);
}
close(to_host[1]);
if (read(to_host[0], &c, 1) == 1) {
pid_t p2 = fork();
if (!p2) {
if (setgid(65534) || setuid(65534))
die("setuid");
errno = 0;
stat(target, &st);
printf("nobody: uid %u: stat(%s): %s\n", getuid(), target, strerror(errno));
exit(0);
}
waitpid(p2, NULL, 0);
if (write(to_cont[1], "g", 1) != 1)
die("sync");
}
waitpid(pid, NULL, 0);
return 0;
}--- fs/proc/generic.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
fs/proc/generic.c +3 -2
@@ -256,9 +256,10 @@ struct dentry *proc_lookup_de(struct inode *dir, struct dentry *dentry, inode =proc_get_inode (dir->i_sb, de);if (!inode)return ERR_PTR (-ENOMEM ); +/* never reused and so never a mountpoint for long */ if (de->flags &PROC_ENTRY_FORCE_LOOKUP ) -return d_splice_alias_ops (inode, dentry, - &proc_net_dentry_ops); +return proc_splice_unmountable (inode, dentry, + &proc_net_dentry_ops);return d_splice_alias_ops (inode, dentry, &proc_misc_dentry_ops); }
Signature · 1 line
2.53.0
Christian Brauner
maintainerPATCH 2/7 proc: don't let /proc/self and /proc/thread-self be mountpoints
In reply to Christian Brauner/proc/self and /proc/thread-self resolve to a different directory for every process that walks through them. A mount on one of the links itself is crossed by every lookup of /proc/self/... in that mount namespace, so it redirects all processes at once, not just the one that created the mount. There's no good reason to ever do that.
Mark both links with dont_mount() before they become visible.
--- fs/proc/self.c | 2 ++ fs/proc/thread_self.c | 2 ++ 2 files changed, 4 insertions(+)
fs/proc/self.c +2 -0
@@ -48,6 +48,8 @@ int proc_setup_self(struct super_block *s) inode->i_uid =GLOBAL_ROOT_UID ; inode->i_gid =GLOBAL_ROOT_GID ; inode->i_op = &proc_self_inode_operations; +/* a mount here would redirect every process */ +dont_mount (self);d_make_persistent (self, inode); ret =0 ; }
fs/proc/thread_self.c +2 -0
@@ -48,6 +48,8 @@ int proc_setup_thread_self(struct super_block *s) inode->i_uid =GLOBAL_ROOT_UID ; inode->i_gid =GLOBAL_ROOT_GID ; inode->i_op = &proc_thread_self_inode_operations; +/* a mount here would redirect every thread */ +dont_mount (thread_self);d_make_persistent (thread_self, inode); ret =0 ; }
Signature · 1 line
2.53.0
Christian Brauner
maintainerPATCH 3/7 proc: don't let /proc/mounts and /proc/net be mountpoints
In reply to Christian Brauner/proc/mounts and /proc/net are links into /proc/self. Like /proc/self itself they resolve differently for every process. A mount on one of the links redirects the lookup for all processes in that mount namespace.
Add proc_self_symlink() for links that point into /proc/self. It creates the link with PROC_ENTRY_CANT_MOUNT and proc_lookup_de() splices such an entry with proc_splice_unmountable(). Use it for /proc/mounts and /proc/net.
Mounting through the links keeps working. Such a mount lands on the caller's own /proc/<pid>/mounts or /proc/<pid>/net, not on the link.
--- fs/proc/generic.c | 26 ++++++++++++++++++++++++-- fs/proc/proc_net.c | 2 +- fs/proc/root.c | 2 +- include/linux/proc_fs.h | 5 +++++ 4 files changed, 31 insertions(+), 4 deletions(-)
fs/proc/generic.c +24 -2
@@ -260,6 +260,10 @@ struct dentry *proc_lookup_de(struct inode *dir, struct dentry *dentry,if (de->flags &PROC_ENTRY_FORCE_LOOKUP )return proc_splice_unmountable (inode, dentry, &proc_net_dentry_ops); +/* a mount on a link into /proc/self redirects every process */ +if (de->flags &PROC_ENTRY_CANT_MOUNT ) +return proc_splice_unmountable (inode, dentry, + &proc_misc_dentry_ops);return d_splice_alias_ops (inode, dentry, &proc_misc_dentry_ops); }
@@ -487,8 +491,8 @@ static struct proc_dir_entry *__proc_create(struct proc_dir_entry **parent,return ent; } -struct proc_dir_entry *proc_symlink (const char *name, -struct proc_dir_entry *parent,const char *dest) +static struct proc_dir_entry *__proc_symlink (const char *name, +struct proc_dir_entry *parent,const char *dest,u8 flags) {struct proc_dir_entry *ent;
@@ -496,6 +500,7 @@ struct proc_dir_entry *proc_symlink(const char *name, (S_IFLNK | S_IRUGO | S_IWUGO | S_IXUGO),1);if (ent) { + ent->flags |= flags; ent->size =strlen (dest); ent->data =kmemdup (dest, ent->size +1 ,GFP_KERNEL );if (ent->data) {
@@ -508,8 +513,25 @@ struct proc_dir_entry *proc_symlink(const char *name, }return ent; } + +struct proc_dir_entry *proc_symlink (const char *name, +struct proc_dir_entry *parent,const char *dest) +{ +return __proc_symlink (name, parent, dest,0 ); +}EXPORT_SYMBOL (proc_symlink); +/* +* A link into /proc/self resolves differently for every process, so a +* mount on the link itself would redirect all of them. +*/ +struct proc_dir_entry *proc_self_symlink (const char *name, +struct proc_dir_entry *parent,const char *dest) +{ +return __proc_symlink (name, parent, dest,PROC_ENTRY_CANT_MOUNT ); +} +EXPORT_SYMBOL_GPL (proc_self_symlink); +struct proc_dir_entry *_proc_mkdir (const char *name,umode_t mode,struct proc_dir_entry *parent,void *data,bool force_lookup) {
fs/proc/proc_net.c +1 -1
@@ -420,7 +420,7 @@ static struct pernet_operations __net_initdata proc_net_ns_ops = {int __init proc_net_init (void ) { -proc_symlink ("net" ,NULL ,"self/net" ); +proc_self_symlink ("net" ,NULL ,"self/net" );return register_pernet_subsys (&proc_net_ns_ops); }
fs/proc/root.c +1 -1
@@ -380,7 +380,7 @@ void __init proc_root_init(void)set_proc_pid_nlink ();proc_self_init ();proc_thread_self_init (); -proc_symlink ("mounts" ,NULL ,"self/mounts" ); +proc_self_symlink ("mounts" ,NULL ,"self/mounts" );proc_net_init ();proc_mkdir ("fs" ,NULL );
include/linux/proc_fs.h +5 -0
@@ -29,6 +29,7 @@ enum { PROC_ENTRY_proc_compat_ioctl =1U <<2 , PROC_ENTRY_proc_lseek =1U <<3 , +PROC_ENTRY_CANT_MOUNT =1U <<6 , PROC_ENTRY_FORCE_LOOKUP = 1U << 7, };
@@ -87,6 +88,8 @@ extern void proc_flush_pid(struct pid *);extern struct proc_dir_entry *proc_symlink (const char *,struct proc_dir_entry *,const char *); +struct proc_dir_entry *proc_self_symlink (const char *,struct proc_dir_entry *, +const char *);struct proc_dir_entry *_proc_mkdir (const char *,umode_t ,struct proc_dir_entry *,void *,bool );extern struct proc_dir_entry *proc_mkdir (const char *,struct proc_dir_entry *); extern struct proc_dir_entry *proc_mkdir_data(const char *, umode_t,
@@ -179,6 +182,8 @@ static inline void proc_flush_pid(struct pid *pid)static inline struct proc_dir_entry *proc_symlink (const char *name,struct proc_dir_entry *parent,const char *dest) {return NULL ;} +static inline struct proc_dir_entry *proc_self_symlink (const char *name, +struct proc_dir_entry *parent,const char *dest) {return NULL ; }static inline struct proc_dir_entry *proc_mkdir (const char *name,struct proc_dir_entry *parent) {return NULL ;}static inline struct proc_dir_entry *proc_create_mount_point (const char *name) {return NULL ; }
Signature · 1 line
2.53.0
Christian Brauner
maintainerPATCH 4/7 afs: create /proc/fs/afs with proc_self_symlink()
In reply to Christian Brauner/proc/fs/afs points into /proc/self/net. Create it with proc_self_symlink() so nothing can be mounted on the link itself.
--- fs/afs/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/afs/main.c +1 -1
@@ -188,7 +188,7 @@ static int __init afs_init(void)if (ret <0 ) goto error_fs; - afs_proc_symlink =proc_symlink ("fs/afs" ,NULL ,"../self/net/afs" ); + afs_proc_symlink =proc_self_symlink ("fs/afs" ,NULL ,"../self/net/afs" );if (!afs_proc_symlink) { ret = -ENOMEM ; goto error_proc;
Signature · 1 line
2.53.0
Christian Brauner
maintainerPATCH 5/7 nfs: create the /proc/fs/nfsfs links with proc_self_symlink()
In reply to Christian Brauner/proc/fs/nfsfs/servers and /proc/fs/nfsfs/volumes resolve through /proc/net into /proc/self/net. Create them with proc_self_symlink() so nothing can be mounted on the links themselves.
--- fs/nfs/client.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
fs/nfs/client.c +2 -2
@@ -1528,11 +1528,11 @@ int __init nfs_fs_proc_init(void) goto error_0;/* a file of servers with which we're dealing */ -if (!proc_symlink ("fs/nfsfs/servers" ,NULL ,"../../net/nfsfs/servers" )) +if (!proc_self_symlink ("fs/nfsfs/servers" ,NULL ,"../../net/nfsfs/servers" )) goto error_1;/* a file of volumes that we have mounted */ -if (!proc_symlink ("fs/nfsfs/volumes" ,NULL ,"../../net/nfsfs/volumes" )) +if (!proc_self_symlink ("fs/nfsfs/volumes" ,NULL ,"../../net/nfsfs/volumes" )) goto error_1;return 0 ;
Signature · 1 line
2.53.0
Re: [PATCH 5/7] nfs: create the /proc/fs/nfsfs links with proc_self_symlink()
Quoted 3 lines · Christian Brauner
/proc/fs/nfsfs/serversand/proc/fs/nfsfs/volumesresolve through/proc/netinto/proc/self/net. Create them withproc_self_symlink()so nothing can be mounted on the links themselves.
Looks good to me.
Quoted 27 lines
Signed-off-by Christian Brauner (Amutable) <[email protected]>--- fs/nfs/client.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)fs/nfs/client.c +2
-2@@ -1528,11 +1528,11 @@ int __init nfs_fs_proc_init(void) goto error_0;/* a file of servers with which we're dealing */ -if (!proc_symlink ("fs/nfsfs/servers" ,NULL ,"../../net/nfsfs/servers" )) +if (!proc_self_symlink ("fs/nfsfs/servers" ,NULL ,"../../net/nfsfs/servers" )) goto error_1;/* a file of volumes that we have mounted */ -if (!proc_symlink ("fs/nfsfs/volumes" ,NULL ,"../../net/nfsfs/volumes" )) +if (!proc_self_symlink ("fs/nfsfs/volumes" ,NULL ,"../../net/nfsfs/volumes" )) goto error_1;return 0 ;Signature · 2 lines
-- 2.53.0
Christian Brauner
maintainerPATCH 6/7 proc: explain which entries can't be mountpoints
In reply to Christian BraunerDocument how a failing ->d_revalidate() takes the mounts off a procfs entry, which entries can't be mountpoints because of that and why the other entries of a process are fine.
--- fs/proc/internal.h | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-)
fs/proc/internal.h +15 -3
@@ -425,9 +425,21 @@ static inline void pde_force_lookup(struct proc_dir_entry *pde) }/* -* Add a new procfs dentry that can't serve as a mountpoint. That should -* encompass anything that is ephemeral and can just disappear while the -* process is still around. +* Add a new procfs dentry that can't serve as a mountpoint. +* +* If ->d_revalidate() fails the VFS calls d_invalidate() on the dentry. That +* detaches every mount on it and below it, in every mount namespace, locked +* mounts included, and the next lookup creates a new dentry without them. So +* nothing whose ->d_revalidate() can fail while the process is still around +* may be a mountpoint: the entries below fd/ and fdinfo/ once the fd is +* closed, the entries below map_files/ once the mapping is gone and +* everything below net/ where proc_net_d_revalidate() fails on every lookup. +* The links into /proc/self can't be mountpoints either. A mount on one of +* them would redirect every process that walks through it. +* +* The other entries of a process fail pid_revalidate() only once the task +* is gone and proc_flush_pid() invalidates them when it is reaped. Mounts +* on them go away together with the task. */ static inline struct dentry *proc_splice_unmountable (struct inode *inode,struct dentry *dentry,const struct dentry_operations *d_ops)
Signature · 1 line
2.53.0
Christian Brauner
maintainerPATCH 7/7 selftests/proc: test which entries can't be mountpoints
In reply to Christian BraunerCheck that nothing can be mounted on:
- /proc/self, /proc/thread-self, /proc/mounts and /proc/net themselves
- the entries below /proc/<pid>/net
- the entries below /proc/<pid>/fd and /proc/<pid>/fdinfo
Mounting through /proc/mounts still works and lands on the entry of the caller.
--- tools/testing/selftests/proc/.gitignore | 1 + tools/testing/selftests/proc/Makefile | 1 + tools/testing/selftests/proc/proc-unmountable.c | 150 ++++++++++++++++++++++++ 3 files changed, 152 insertions(+)
tools/testing/selftests/proc/.gitignore +1 -0
@@ -20,6 +20,7 @@ /proc-uptime-001 /proc-uptime-002 /proc-pidns +/proc-unmountable /read /self /setns-dcache
tools/testing/selftests/proc/Makefile +1 -0
@@ -30,5 +30,6 @@ TEST_GEN_PROGS += thread-self TEST_GEN_PROGS += proc-multiple-procfs TEST_GEN_PROGS += proc-fsconfig-hidepid TEST_GEN_PROGS += proc-pidns +TEST_GEN_PROGS += proc-unmountable include ../lib.mk
tools/testing/selftests/proc/proc-unmountable.c new file +150 -0
150 lines not on this page · raw message
Signature · 1 line
2.53.0
Quoted 116 lines · Christian Brauner
In Linux 6.12 I stopped allowing mounts on the entries below /proc/<pid>/fd/, /proc/<pid>/fdinfo/ and /proc/<pid>/map_files/ (cf. [1]). This extends that to two more groups of entries:
(1) Everything below /proc/<pid>/net/.
proc_net_d_revalidate()fails on every lookup since1fde6f21d90f("proc: fix/proc/net/*after setns(2)").lookup_fast()answers that withd_invalidate()which detaches every mount on the dentry. That affects every mount namespace and includes locked mounts.So a mask on /proc/<pid>/net/dev goes away the first time anybody looks the name up. The directory is 0555 so anybody really means anybody. This is reproducible with a container that masks /proc/<pid>/net/dev with a locked bind mount and a
stat()by nobody in the initial namespaces makes the real file show up again. So it's pointless.A mount that can't stick is better refused than taken away. I also have a version that makes
proc_net_d_revalidate()check the network namespace instead so mounts there stick. But I think that's bullsh*t. It would make mounts on the net entries of other processes stick as well and the few users that try to mount there fail today anyway (see below).(2) The links into
/proc/self:/proc/self,/proc/thread-self,/proc/mounts,/proc/netand the afs and nfs links below/proc/fs/.These resolve to a different directory for every process. A mount on the link itself is crossed by every lookup through it in that mount namespace, so it redirects all processes at once. Mounting through the links keeps working and lands on the caller's own entry as before.
What I really wanted was to block mounting on anything below /proc/<pid>/. That was the big hammer I mentioned in [1] years ago. We can't... I did a pretty exhaustive survey of who mounts on the entries below /proc/<pid>/ or on the links into it:
- GitHub code search and regex searches on Sourcegraph
- Debian Code Search over unstable
- the systemd units of every binary package in Debian 11, 12, 13 and sid, Ubuntu 22.04, 24.04 and 26.04, Fedora 43 and 44, CentOS Stream 10, openSUSE Leap 16 and Tumbleweed and Arch
- clones of 46 container runtimes and orchestrators, the sandboxes (systemd, bubblewrap, flatpak, firejail, snapd, nsjail, minijail, Chromium and Firefox), Android and the rooting tools on top of it, and the AppArmor and SELinux policies
- detection rules, issue trackers, lore, LWN and the Q&A sitesThe legitimate users all mount on entries of the process that does the mount:
(1) systemd units that list
/proc/self/... in ReadOnlyPaths=, ReadWritePaths= or ReadWriteDirectories=. systemd resolves/proc/selfin the child of the service and binds the entry onto itself before it execs the service. If the mount fails the unit fails with 226/NAMESPACE and the "-" prefix doesn't help. That's netdata in Debian 11 and 12 and Ubuntu 22.04 and 24.04, opendht in Debian 12 and sid, Ubuntu 24.04 and Arch, accountsservice 22.x in Debian 12 and Ubuntu 22.04, OpenSearch 3, wazuh-indexer, TrueNAS SCALE, OONI's backend and the JupyterHub systemdspawner example. Of the 731 unit files on GitHub that name a /proc path in one of these directives 53 name an entry of the service itself.(2) systemd's
make_filesystem()binds/dev/nullover/proc/self/mountsbefore it execs mkfs.btrfs so that it doesn't refuse a device whose other partitions are mounted. systemd-repart, systemd-homed and systemd-makefs use that.(3) mkosi binds a fake
/proc/self/mountinfofor grub-bios-setup.(4) runc applies maskedPaths such as
/proc/1/environfrom the container's init when a config asks for it, bwrap does the same from its pid 1, firejail hides the name of its pid 1 and clampdown masks/proc/1/mem.(5) Shell scripts that mount on
/proc/$$/... and then exec, e.g., luci-app-shadowrt on OpenWrt.(6) Tests in systemd, elogind, LXC, libpathrs and bun.
Fedora's SELinux policy even grants
fsadm_tandinit_tthe permission to mount over their own entries because of (1) and (2). Sad, very sad...Everything else I found is malware or offensive tooling:
Hiding a process by mounting over /proc/<pid> is catalogued as MITRE ATT&CK T1564.013 and used in the wild by TeamTNT, Commando Cat, the KONO DIO DA miner, KV-botnet, UNC2891 and kbotne. hackshell and emp3r0r ship it and Elastic's endpoint rule kills any process that does it.
So blocking all of /proc/<pid>/ would break units in released distributions and systemd itself. A rule that only lets a process mount on its own entries, and on a directory only a bind of itself, keeps (1) to (4) working, except for a few bwrap sandboxes that put a tmpfs over directories like
/proc/1/task, and stops all of the attacks. I have that as well but it's not really a fix so we can delay it.The remaining entries of a process only fail
pid_revalidate()once the task is gone andproc_flush_pid()invalidates them when it is reaped. Their mounts go away together with the task and nothing gets revealed.Nothing I found mounts on the links themselves and crun even has a test that expects it to fail. The handful of units that mask entries below /proc/<pid>/net/, e.g., Algo VPN's ReadOnlyPaths=/proc/net/pfkey for strongswan, already fail today.
systemd's own lookup invalidates the dentry before it can bind onto it. With this series they fail the same way. Apart from tests in libpathrs and securejoin that had to be adapted the only fallout from the 6.12 change we found is one small project that handed libmount's new mount api a
/proc/self/fd/Ntarget thatmove_mount()doesn't follow withoutMOVE_MOUNT_T_SYMLINKS. Ubuntu even backported the 6.12 change to their 22.04 and 24.04 kernels this year.
Quoted 5 lines
Nuke this crap.
[1]: https://lore.kernel.org/r/[email protected]
Signed-off-by Christian Brauner (Amutable) <[email protected]>
Makes sense to me although I don't dare to judge all possible consequences (I don't have much practical experience with container runtimes that play these games). Anyway, feel free to add:
Honza
Quoted 27 lines
---
Christian Brauner (7):
proc: don't let an entry below /proc/<pid>/net be a mountpoint
proc: don't let/proc/selfand/proc/thread-selfbe mountpoints
proc: don't let/proc/mountsand/proc/netbe mountpoints
afs: create/proc/fs/afswithproc_self_symlink()
nfs: create the/proc/fs/nfsfslinks withproc_self_symlink()
proc: explain which entries can't be mountpoints selftests/proc: test which entries can't be mountpointsfs/afs/main.c | 2 +- fs/nfs/client.c | 4 +- fs/proc/generic.c | 31 ++++- fs/proc/internal.h | 18 ++- fs/proc/proc_net.c | 2 +- fs/proc/root.c | 2 +- fs/proc/self.c | 2 + fs/proc/thread_self.c | 2 + include/linux/proc_fs.h | 5 + tools/testing/selftests/proc/.gitignore | 1 + tools/testing/selftests/proc/Makefile | 1 + tools/testing/selftests/proc/proc-unmountable.c | 150 ++++++++++++++++++++++++ 12 files changed, 208 insertions(+), 12 deletions(-)---
base-commit:7be4bc52fd9e0e55dc9a2e0232b89427d2c4b614
change-id: 20261024-work-procfs-fixes-8b21a6d17448
Signature · 2 lines
Jan Kara <[email protected]> SUSE Labs, CR
Quoted 5 lines · Christian Brauner
Nuke this crap.
[1]: https://lore.kernel.org/r/[email protected]
Signed-off-by Christian Brauner (Amutable) <[email protected]>
Quoted 9 lines
---
Christian Brauner (7):
proc: don't let an entry below /proc/<pid>/net be a mountpoint
proc: don't let/proc/selfand/proc/thread-selfbe mountpoints
proc: don't let/proc/mountsand/proc/netbe mountpoints
afs: create/proc/fs/afswithproc_self_symlink()
nfs: create the/proc/fs/nfsfslinks withproc_self_symlink()
proc: explain which entries can't be mountpoints selftests/proc: test which entries can't be mountpoints
I love it, nuke it from orbit.
The proc_net_d_revalidate() thing is particularly wild, I'm surprised it hadn't come up to us in bug reports in the past ~10 years. I guess nobody cared enough to try to mask it?
And yeah, it's a shame we couldn't kill this entirely but I'm looking forward to the patch you mentioned that restricts who and what can be over-mounted.
Quoted 16 lines
fs/afs/main.c | 2 +- fs/nfs/client.c | 4 +- fs/proc/generic.c | 31 ++++- fs/proc/internal.h | 18 ++- fs/proc/proc_net.c | 2 +- fs/proc/root.c | 2 +- fs/proc/self.c | 2 + fs/proc/thread_self.c | 2 + include/linux/proc_fs.h | 5 + tools/testing/selftests/proc/.gitignore | 1 + tools/testing/selftests/proc/Makefile | 1 + tools/testing/selftests/proc/proc-unmountable.c | 150 ++++++++++++++++++++++++ 12 files changed, 208 insertions(+), 12 deletions(-)---
base-commit:7be4bc52fd9e0e55dc9a2e0232b89427d2c4b614
change-id: 20261024-work-procfs-fixes-8b21a6d17448
Signature · 3 lines
Aleksa Sarai Founding Engineer at Amutable https://www.cyphar.com/