Internet Engineering Task Force (IETF) J. Fields Request for Comments: 8275 A. Gruenbacher Category: Standards Track Red Hat ISSN: 2070-1721 December 2017
Allowing Inheritable NFSv4 Access Control Entries to Override the Umask
Abstract
In many environments, inheritable NFSv4 Access Control Entries (ACEs) can be rendered ineffective by the application of the per-process file mode creation mask (umask). This can be addressed by transmitting the umask and create mode as separate pieces of data, allowing the server to make more intelligent decisions about the permissions to set on new files. This document proposes a protocol extension to accomplish that.
Status of This Memo
This is an Internet Standards Track document.
This document is a product of the Internet Engineering Task Force (IETF). It represents the consensus of the IETF community. It has received public review and has been approved for publication by the Internet Engineering Steering Group (IESG). Further information on Internet Standards is available in Section 2 of RFC 7841.
Information about the current status of this document, any errata, and how to provide feedback on it may be obtained at https://www.rfc-editor.org/info/rfc8275.
Copyright Notice
Copyright (c) 2017 IETF Trust and the persons identified as the document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.
On Unix-like systems, each process is associated with a file mode creation mask (umask). The umask specifies which permissions must be turned off when creating new file system objects.
When applying the mode, Section 6.4.1.1 of [RFC7530] recommends that servers SHOULD restrict permissions granted to any user or group named in the Access Control List (ACL) to be no more than the permissions granted by the MODE4_RGRP, MODE4_WGRP, and MODE4_XGRP bits. Servers aiming to provide clients with Unix-like chmod behavior may also be motivated by the same requirements in [SUSv4]. (See the discussion of additional and alternate access control mechanisms in "File Permissions", Section 4.4 of [SUSv4].)
On many existing installations, all ordinary users use the same effective group ID by default. To prevent granting all users full access to each other's files, such installations usually default to a umask with very restrictive permissions. As a result, inherited ACL entries (inheritable ACEs) describing the permissions to be granted to named users and groups are often ignored. This makes inheritable ACEs useless in some common cases.
Linux solves this problem on local file systems by ignoring the umask whenever a newly created file inherits ACEs from its parent; see [LinuxACL].
The same solution should work for NFS. However, the NFSv4 protocol does not currently give the client a way to transmit the umask of the process opening a file. And clients have no way of atomically checking for inheritable permissions and applying the umask only when necessary. As a result, the server receives an OPEN with a mode attribute that already has the umask applied.
Fields & Gruenbacher Standards Track [Page 2]
RFC 8275 NFSv4 Umask December 2017
This document solves the problem by defining a new attribute that allows the client to transmit umask and the mode specified at file creation separately, allowing the client to ignore the umask in the presence of inheritable ACEs. At least in the Linux case, this allows NFSv4 to provide the same semantics available using local access.
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.
This document presents an extension to minor version 2 of the NFSv4 protocol as described in [RFC8178]. It describes a new OPTIONAL feature. NFSv4.2 servers and clients implemented without knowledge of this extension will continue to interoperate with clients and servers that are aware of the extension (whether or not they support it).
Note that [RFC7862] does not define NFSv4.2 as non-extensible, so [RFC8178] treats it as an extensible minor version. This Standards Track RFC extends NFSv4.2 but does not update [RFC7862] or [RFC7863].
The additional lines of External Data Representation (XDR) [RFC4506] description embedded in this document can be extracted by feeding this document into the following shell script:
<CODE BEGINS>
#!/bin/sh grep '^ *///' $* | sed 's?^ */// ??' | sed 's?^ *///$??'
<CODE ENDS>
That is, if the above script is stored in a file called "extract.sh", and this document is in a file called "umask.txt", then the reader can do:
sh extract.sh < umask.txt > umask.x
Fields & Gruenbacher Standards Track [Page 3]
RFC 8275 NFSv4 Umask December 2017
The effect of the script is to remove leading white space from each line, plus a sentinel sequence of "///".
Once that extraction is done, these added lines need to be inserted into an appropriate base XDR of the generated XDR from [RFC7863] together with XDR from any additional extensions to be recognized by the implementation. This will result in a ready-to-compile XDR file.
+------------+----+-------------+-----+------------+ | Name | Id | Data Type | Acc | Defined in | +------------+----+-------------+-----+------------+ | mode_umask | 81 | mode_umask4 | W | Section 5 | +------------+----+-------------+-----+------------+
Table 1
The NFSv4.2 mode_umask attribute is based on the umask and on the mode bits specified at open time, which together determine the mode of a newly created UNIX file. Only the nine low-order mode4 bits of mu_umask are defined. A server MUST return NFS4ERR_INVAL if bits other than those nine are set.
The mode_umask attribute is only meaningful for operations that create objects (CREATE and OPEN); in other operations that take fattr4 arguments, the server MUST reject it with NFS4ERR_INVAL.
The server MUST return NFS4ERR_INVAL if the client attempts to set both mode and mode_umask in the same operation.
Fields & Gruenbacher Standards Track [Page 4]
RFC 8275 NFSv4 Umask December 2017
When the server supports the mode_umask attribute, a client creating a file should use mode_umask in place of mode, with mu_mode set to the unmodified mode provided by the user and mu_umask set to the umask of the requesting process.
The server then uses mode_umask as follows:
o On a server that supports ACL attributes, if an object inherits any ACEs from its parent directory, mu_mode SHOULD be used and mu_umask ignored.
o Otherwise, mu_umask MUST be used to limit the mode: all bits in the mode that are set in the unmask MUST be turned off; the mode assigned to the new object becomes (mu_mode & ~mu_umask) instead.
The mode_umask attribute shifts to the server the decision about when to apply the umask. Because the server MUST apply the umask if there are no inheritable permissions, the traditional semantics are preserved in the absence of a permission inheritance mechanism. The only relaxation of permissions comes in the case in which servers follow the recommendation that they ignore the umask in the presence of inheritable permissions.
The practice of ignoring the umask when there are inheritable permissions in the form of a "POSIX" default ACL is of long standing and has not given rise to security issues. The "POSIX" default ACL mechanism and the mechanism for permission inheritance in NFSv4 are equivalent from a security perspective.
[SUSv4] The Open Group, "Single UNIX Specification, Version 4", 2013.
Fields & Gruenbacher Standards Track [Page 6]
RFC 8275 NFSv4 Umask December 2017
Acknowledgments
Thanks to Trond Myklebust and Dave Noveck for their review and the suggestion to define this as a (mode, umask) pair rather than just umask. Thanks to Warren Kumari, Adam Roach, Spencer Dawkins, Mike Kupfer, and Thomas Haynes for their review and to Thomas Haynes for help with XDR.