man-pages/man2/ioctl_ns.2

.\" Copyright (c) 2017 by Michael Kerrisk <mtk.manpages@gmail.com>
.\"
.\" %%%LICENSE_START(VERBATIM)
.\" Permission is granted to make and distribute verbatim copies of this
.\" manual provided the copyright notice and this permission notice are
.\" preserved on all copies.
.\"
.\" Permission is granted to copy and distribute modified versions of this
.\" manual under the conditions for verbatim copying, provided that the
.\" entire resulting derived work is distributed under the terms of a
.\" permission notice identical to this one.
.\"
.\" Since the Linux kernel and libraries are constantly changing, this
.\" manual page may be incorrect or out-of-date.  The author(s) assume no
.\" responsibility for errors or omissions, or for damages resulting from
.\" the use of the information contained herein.  The author(s) may not
.\" have taken the same level of care in the production of this manual,
.\" which is licensed free of charge, as they might when working
.\" professionally.
.\"
.\" Formatted or processed versions of this manual, if unaccompanied by
.\" the source, must acknowledge the copyright and authors of this work.
.\" %%%LICENSE_END
.\"
.\"
.TH IOCTL_NS 2 2021-03-22 "Linux" "Linux Programmer's Manual"
.SH NAME
ioctl_ns \- ioctl() operations for Linux namespaces
.SH DESCRIPTION
.\" ============================================================
.\"
.SS Discovering namespace relationships
The following
.BR ioctl (2)
operations are provided to allow discovery of namespace relationships (see
.BR user_namespaces (7)
and
.BR pid_namespaces (7)).
The form of the calls is:
.PP
.in +4n
.EX
new_fd = ioctl(fd, request);
.EE
.in
.PP
In each case,
.I fd
refers to a
.IR /proc/[pid]/ns/*
file.
Both operations return a new file descriptor on success.
.TP
.BR NS_GET_USERNS " (since Linux 4.9)"
.\" commit bcac25a58bfc6bd79191ac5d7afb49bea96da8c9
.\" commit 6786741dbf99e44fb0c0ed85a37582b8a26f1c3b
Returns a file descriptor that refers to the owning user namespace
for the namespace referred to by
.IR fd .
.TP
.BR NS_GET_PARENT " (since Linux 4.9)"
.\" commit a7306ed8d94af729ecef8b6e37506a1c6fc14788
Returns a file descriptor that refers to the parent namespace of
the namespace referred to by
.IR fd .
This operation is valid only for hierarchical namespaces
(i.e., PID and user namespaces).
For user namespaces,
.BR NS_GET_PARENT
is synonymous with
.BR NS_GET_USERNS .
.PP
The new file descriptor returned by these operations is opened with the
.BR O_RDONLY
and
.BR O_CLOEXEC
(close-on-exec; see
.BR fcntl (2))
flags.
.PP
By applying
.BR fstat (2)
to the returned file descriptor, one obtains a
.I stat
structure whose
.I st_dev
(resident device) and
.I st_ino
(inode number) fields together identify the owning/parent namespace.
This inode number can be matched with the inode number of another
.IR /proc/[pid]/ns/{pid,user}
file to determine whether that is the owning/parent namespace.
.PP
Either of these
.BR ioctl (2)
operations can fail with the following errors:
.TP
.B EPERM
The requested namespace is outside of the caller's namespace scope.
This error can occur if, for example, the owning user namespace is an
ancestor of the caller's current user namespace.
It can also occur on attempts to obtain the parent of the initial
user or PID namespace.
.TP
.B ENOTTY
The operation is not supported by this kernel version.
.PP
Additionally, the
.B NS_GET_PARENT
operation can fail with the following error:
.TP
.B EINVAL
.I fd
refers to a nonhierarchical namespace.
.PP
See the EXAMPLES section for an example of the use of these operations.
.\" ============================================================
.\"
.SS Discovering the namespace type
The
.B NS_GET_NSTYPE
.\" commit e5ff5ce6e20ee22511398bb31fb912466cf82a36
operation (available since Linux 4.11) can be used to discover
the type of namespace referred to by the file descriptor
.IR fd :
.PP
.in +4n
.EX
nstype = ioctl(fd, NS_GET_NSTYPE);
.EE
.in
.PP
.I fd
refers to a
.IR /proc/[pid]/ns/*
file.
.PP
The return value is one of the
.BR CLONE_NEW*
values that can be specified to
.BR clone (2)
or
.BR unshare (2)
in order to create a namespace.
.\" ============================================================
.\"
.SS Discovering the owner of a user namespace
The
.B NS_GET_OWNER_UID
.\" commit 015bb305b8ebe8d601a238ab70ebdc394c7a19ba
operation (available since Linux 4.11) can be used to discover
the owner user ID of a user namespace (i.e., the effective user ID
of the process that created the user namespace).
The form of the call is:
.PP
.in +4n
.EX
uid_t uid;
ioctl(fd, NS_GET_OWNER_UID, &uid);
.EE
.in
.PP
.I fd
refers to a
.IR /proc/[pid]/ns/user
file.
.PP
The owner user ID is returned in the
.I uid_t
pointed to by the third argument.
.PP
This operation can fail with the following error:
.TP
.B EINVAL
.I fd
does not refer to a user namespace.
.SH ERRORS
Any of the above
.BR ioctl ()
operations can return the following errors:
.TP
.B ENOTTY
.I fd
does not refer to a
.I /proc/[pid]/ns/*
file.
.SH CONFORMING TO
Namespaces and the operations described on this page are a Linux-specific.
.SH EXAMPLES
The example shown below uses the
.BR ioctl (2)
operations described above to perform simple
discovery of namespace relationships.
The following shell sessions show various examples of the use
of this program.
.PP
Trying to get the parent of the initial user namespace fails,
since it has no parent:
.PP
.in +4n
.EX
$ \fB./ns_show /proc/self/ns/user p\fP
The parent namespace is outside your namespace scope
.EE
.in
.PP
Create a process running
.BR sleep (1)
that resides in new user and UTS namespaces,
and show that the new UTS namespace is associated with the new user namespace:
.PP
.in +4n
.EX
$ \fBunshare \-Uu sleep 1000 &\fP
[1] 23235
$ \fB./ns_show /proc/23235/ns/uts u\fP
Device/Inode of owning user namespace is: [0,3] / 4026532448
$ \fBreadlink /proc/23235/ns/user\fP
user:[4026532448]
.EE
.in
.PP
Then show that the parent of the new user namespace in the preceding
example is the initial user namespace:
.PP
.in +4n
.EX
$ \fBreadlink /proc/self/ns/user\fP
user:[4026531837]
$ \fB./ns_show /proc/23235/ns/user p\fP
Device/Inode of parent namespace is: [0,3] / 4026531837
.EE
.in
.PP
Start a shell in a new user namespace, and show that from within
this shell, the parent user namespace can't be discovered.
Similarly, the UTS namespace
(which is associated with the initial user namespace)
can't be discovered.
.PP
.in +4n
.EX
$ \fBPS1="sh2$ " unshare \-U bash\fP
sh2$ \fB./ns_show /proc/self/ns/user p\fP
The parent namespace is outside your namespace scope
sh2$ \fB./ns_show /proc/self/ns/uts u\fP
The owning user namespace is outside your namespace scope
.EE
.in
.SS Program source
\&
.EX
/* ns_show.c

   Licensed under the GNU General Public License v2 or later.
*/
#include <stdint.h>
#include <stdlib.h>
#include <unistd.h>
#include <stdio.h>
#include <fcntl.h>
#include <string.h>
#include <sys/stat.h>
#include <sys/ioctl.h>
#include <errno.h>
#include <sys/sysmacros.h>

#ifndef NS_GET_USERNS
#define NSIO    0xb7
#define NS_GET_USERNS   _IO(NSIO, 0x1)
#define NS_GET_PARENT   _IO(NSIO, 0x2)
#endif

int
main(int argc, char *argv[])
{
    int fd, userns_fd, parent_fd;
    struct stat sb;

    if (argc < 2) {
        fprintf(stderr, "Usage: %s /proc/[pid]/ns/[file] [p|u]\en",
                argv[0]);
        fprintf(stderr, "\enDisplay the result of one or both "
                "of NS_GET_USERNS (u) or NS_GET_PARENT (p)\en"
                "for the specified /proc/[pid]/ns/[file]. If neither "
                "\(aqp\(aq nor \(aqu\(aq is specified,\en"
                "NS_GET_USERNS is the default.\en");
        exit(EXIT_FAILURE);
    }

    /* Obtain a file descriptor for the \(aqns\(aq file specified
       in argv[1]. */

    fd = open(argv[1], O_RDONLY);
    if (fd == \-1) {
        perror("open");
        exit(EXIT_FAILURE);
    }

    /* Obtain a file descriptor for the owning user namespace and
       then obtain and display the inode number of that namespace. */

    if (argc < 3 || strchr(argv[2], \(aqu\(aq)) {
        userns_fd = ioctl(fd, NS_GET_USERNS);

        if (userns_fd == \-1) {
            if (errno == EPERM)
                printf("The owning user namespace is outside "
                        "your namespace scope\en");
            else
               perror("ioctl\-NS_GET_USERNS");
            exit(EXIT_FAILURE);
         }

        if (fstat(userns_fd, &sb) == \-1) {
            perror("fstat\-userns");
            exit(EXIT_FAILURE);
        }
        printf("Device/Inode of owning user namespace is: "
                "[%jx,%jx] / %ju\en",
                (uintmax_t) major(sb.st_dev),
                (uintmax_t) minor(sb.st_dev),
                (uintmax_t) sb.st_ino);

        close(userns_fd);
    }

    /* Obtain a file descriptor for the parent namespace and
       then obtain and display the inode number of that namespace. */

    if (argc > 2 && strchr(argv[2], \(aqp\(aq)) {
        parent_fd = ioctl(fd, NS_GET_PARENT);

        if (parent_fd == \-1) {
            if (errno == EINVAL)
                printf("Can\(aq get parent namespace of a "
                        "nonhierarchical namespace\en");
            else if (errno == EPERM)
                printf("The parent namespace is outside "
                        "your namespace scope\en");
            else
                perror("ioctl\-NS_GET_PARENT");
            exit(EXIT_FAILURE);
        }

        if (fstat(parent_fd, &sb) == \-1) {
            perror("fstat\-parentns");
            exit(EXIT_FAILURE);
        }
        printf("Device/Inode of parent namespace is: [%jx,%jx] / %ju\en",
                (uintmax_t) major(sb.st_dev),
                (uintmax_t) minor(sb.st_dev),
                (uintmax_t) sb.st_ino);

        close(parent_fd);
    }

    exit(EXIT_SUCCESS);
}
.EE
.SH SEE ALSO
.BR fstat (2),
.BR ioctl (2),
.BR proc (5),
.BR namespaces (7)