[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[PATCH 18/25] virtiofsd: move to a new pid namespace
From: |
Dr. David Alan Gilbert (git) |
Subject: |
[PATCH 18/25] virtiofsd: move to a new pid namespace |
Date: |
Thu, 24 Oct 2019 12:27:11 +0100 |
From: Stefan Hajnoczi <address@hidden>
virtiofsd needs access to /proc/self/fd. Let's move to a new pid
namespace so that a compromised process cannot see another other
processes running on the system.
One wrinkle in this approach: unshare(CLONE_NEWPID) affects *child*
processes and not the current process. Therefore we need to fork the
pid 1 process that will actually run virtiofsd and leave a parent in
waitpid(2). This is not the same thing as daemonization and parent
processes should not notice a difference.
Signed-off-by: Stefan Hajnoczi <address@hidden>
---
contrib/virtiofsd/passthrough_ll.c | 76 ++++++++++++++++++++++++------
1 file changed, 62 insertions(+), 14 deletions(-)
diff --git a/contrib/virtiofsd/passthrough_ll.c
b/contrib/virtiofsd/passthrough_ll.c
index 58cf82a89f..c027db64e6 100644
--- a/contrib/virtiofsd/passthrough_ll.c
+++ b/contrib/virtiofsd/passthrough_ll.c
@@ -52,6 +52,8 @@
#include <sys/syscall.h>
#include <sys/xattr.h>
#include <sys/mount.h>
+#include <sys/types.h>
+#include <sys/wait.h>
#include "passthrough_helpers.h"
@@ -1835,6 +1837,63 @@ static void setup_net_namespace(void)
}
}
+/*
+ * Move to a new pid namespace to prevent access to other processes if this
+ * process is compromised.
+ */
+static void setup_pid_namespace(void)
+{
+ pid_t child;
+
+ /*
+ * Create a new pid namespace for *child* processes. We'll have to
+ * fork in order to enter the new pid namespace. A new mount namespace
+ * is also needed so that we can remount /proc for the new pid
+ * namespace.
+ */
+ if (unshare(CLONE_NEWPID | CLONE_NEWNS) != 0) {
+ fuse_log(FUSE_LOG_ERR, "unshare(CLONE_NEWPID | CLONE_NEWNS):
%m\n");
+ exit(1);
+ }
+
+ child = fork();
+ if (child < 0) {
+ fuse_log(FUSE_LOG_ERR, "fork() failed: %m\n");
+ exit(1);
+ }
+ if (child > 0) {
+ pid_t waited;
+ int wstatus;
+
+ /* The parent waits for the child */
+ do {
+ waited = waitpid(child, &wstatus, 0);
+ } while (waited < 0 && errno == EINTR);
+
+ if (WIFEXITED(wstatus)) {
+ exit(WEXITSTATUS(wstatus));
+ }
+
+ exit(1);
+ }
+
+ /*
+ * If the mounts have shared propagation then we want to opt out so our
+ * mount changes don't affect the parent mount namespace.
+ */
+ if (mount(NULL, "/", NULL, MS_REC|MS_SLAVE, NULL) < 0) {
+ fuse_log(FUSE_LOG_ERR, "mount(/, MS_REC|MS_SLAVE): %m\n");
+ exit(1);
+ }
+
+ /* The child must remount /proc to use the new pid namespace */
+ if (mount("proc", "/proc", "proc",
+ MS_NODEV | MS_NOEXEC | MS_NOSUID | MS_RELATIME, NULL) < 0) {
+ fuse_log(FUSE_LOG_ERR, "mount(/proc): %m\n");
+ exit(1);
+ }
+}
+
/* This magic is based on lxc's lxc_pivot_root() */
static void setup_pivot_root(const char *source)
{
@@ -1898,20 +1957,10 @@ static void setup_proc_self_fd(struct lo_data *lo)
/*
* Make the source directory our root so symlinks cannot escape and no other
- * files are accessible.
+ * files are accessible. Assumes unshare(CLONE_NEWNS) was already called.
*/
static void setup_mount_namespace(const char *source)
{
- if (unshare(CLONE_NEWNS) != 0) {
- fuse_log(FUSE_LOG_ERR, "unshare(CLONE_NEWNS): %m\n");
- exit(1);
- }
-
- if (mount(NULL, "/", NULL, MS_REC|MS_SLAVE, NULL) < 0) {
- fuse_log(FUSE_LOG_ERR, "mount(/, MS_REC|MS_PRIVATE): %m\n");
- exit(1);
- }
-
if (mount(source, source, NULL, MS_BIND, NULL) < 0) {
fuse_log(FUSE_LOG_ERR, "mount(%s, %s, MS_BIND): %m\n", source,
source);
exit(1);
@@ -1926,6 +1975,8 @@ static void setup_mount_namespace(const char *source)
*/
static void setup_sandbox(struct lo_data *lo)
{
+ setup_pid_namespace();
+ setup_proc_self_fd(lo);
setup_net_namespace();
setup_mount_namespace(lo->source);
}
@@ -2046,9 +2097,6 @@ int main(int argc, char *argv[])
fuse_daemonize(opts.foreground);
- /* Must be after daemonize to get the right /proc/self/fd */
- setup_proc_self_fd(&lo);
-
setup_sandbox(&lo);
/* Block until ctrl+c or fusermount -u */
--
2.23.0
- [PATCH 08/25] virtiofsd: Plumb fuse_bufvec through to do_write_buf, (continued)
- [PATCH 08/25] virtiofsd: Plumb fuse_bufvec through to do_write_buf, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 09/25] virtiofsd: Pass write iov's all the way through, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 10/25] virtiofsd: add fuse_mbuf_iter API, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 11/25] virtiofsd: validate input buffer sizes in do_write_buf(), Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 12/25] virtiofsd: check input buffer size in fuse_lowlevel.c ops, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 15/25] virtiofsd: use /proc/self/fd/ O_PATH file descriptor, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 13/25] virtiofsd: prevent ".." escape in lo_do_lookup(), Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 14/25] virtiofsd: prevent ".." escape in lo_do_readdir(), Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 16/25] virtiofsd: sandbox mount namespace, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 17/25] virtiofsd: move to an empty network namespace, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 18/25] virtiofsd: move to a new pid namespace,
Dr. David Alan Gilbert (git) <=
- [PATCH 19/25] virtiofsd: add seccomp whitelist, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 20/25] virtiofsd: Parse flag FUSE_WRITE_KILL_PRIV, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 21/25] virtiofsd: Drop CAP_FSETID if client asked for it, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 22/25] virtiofsd: set maximum RLIMIT_NOFILE limit, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 23/25] virtiofsd: add security guide document, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 24/25] virtiofsd: add --syslog command-line option, Dr. David Alan Gilbert (git), 2019/10/24
- [PATCH 25/25] virtiofsd: print log only when priority is high enough, Dr. David Alan Gilbert (git), 2019/10/24