| <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> |
| <html><head><meta http-equiv="Content-Type" content="text/html; charset=ANSI_X3.4-1968"><title>POSIX accounts, permission, and security</title><link rel="stylesheet" type="text/css" href="docbook.css"><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot"><link rel="home" href="cygwin-ug-net.html" title="Cygwin User's Guide"><link rel="up" href="using.html" title="Chapter 3. Using Cygwin"><link rel="prev" href="using-cygwinenv.html" title="The CYGWIN environment variable"><link rel="next" href="using-cygserver.html" title="Cygserver"></head><body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF"><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="3" align="center">POSIX accounts, permission, and security</th></tr><tr><td width="20%" align="left"><a accesskey="p" href="using-cygwinenv.html">Prev</a> </td><th width="60%" align="center">Chapter 3. Using Cygwin</th><td width="20%" align="right"> <a accesskey="n" href="using-cygserver.html">Next</a></td></tr></table><hr></div><div class="sect1"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="ntsec"></a>POSIX accounts, permission, and security</h2></div></div></div><p>This section discusses how the Windows security model is |
| utilized in Cygwin to implement POSIX account information, POSIX-like |
| permissions, and how the Windows authentication model is used to allow |
| cygwin applications to switch users in a POSIX-like fashion.</p><p>The setting of POSIX-like file and directory permissions is |
| controlled by the <a class="link" href="using.html#mount-table" title="The Cygwin Mount Table">mount</a> option |
| <code class="literal">(no)acl</code> which is set to <code class="literal">acl</code> by |
| default.</p><p>We start with a short overview. Note that this overview must |
| be necessarily short. If you want to learn more about the Windows security |
| model, see the <a class="ulink" href="http://msdn.microsoft.com/en-us/library/aa374860(VS.85).aspx" target="_top">Access Control</a> article in MSDN documentation.</p><p>POSIX concepts and in particular the POSIX security model are not |
| discussed here, but assumed to be understood by the reader. If you |
| don't know the POSIX security model, search the web for beginner |
| documentation.</p><div class="sect2"><div class="titlepage"><div><div><h3 class="title"><a name="ntsec-common"></a>Brief overview of Windows security</h3></div></div></div><p>In the Windows security model, almost any "object" is securable. |
| "Objects" are files, processes, threads, semaphores, etc.</p><p>Every object has a data structure attached, called a "security |
| descriptor" (SD). The SD contains all information necessary to control |
| who can access an object, and to determine what they are allowed to do |
| to or with it. The SD of an object consists of five parts:</p><div class="itemizedlist"><ul class="itemizedlist compact" style="list-style-type: disc; "><li class="listitem"><p>Flags which control several aspects of this SD. This is |
| not discussed here.</p></li><li class="listitem"><p>The SID of the object owner.</p></li><li class="listitem"><p>The SID of the object owner group.</p></li><li class="listitem"><p>A list of "Access Control Entries" (ACE), called the |
| "Discretionary Access Control List" (DACL).</p></li><li class="listitem"><p>Another list of ACEs, called the "Security Access Control List" |
| (SACL), which doesn't matter for our purpose. We ignore it here.</p></li></ul></div><p>Every ACE contains a so-called "Security IDentifier" (SID) and |
| other stuff which is explained a bit later. Let's talk about the SID first. |
| </p><p>A SID is a unique identifier for users, groups, computers and |
| Active Directory (AD) domains. SIDs are basically comparable to POSIX |
| user ids (UIDs) and group ids (GIDs), but are more complicated because |
| they are unique across multiple machines or domains. A SID is a |
| structure of multiple numerical values. There's a convenient convention |
| to type SIDs, as a string of numerical fields separated by hyphen |
| characters. Here's an example:</p><p>SID of a machine "foo":</p><pre class="screen"> |
| S-1-5-21-165875785-1005667432-441284377 |
| </pre><p>SID of a user "johndoe" of the system "foo":</p><pre class="screen"> |
| S-1-5-21-165875785-1005667432-441284377-1023 |
| </pre><p>The first field is always "S", which is just a notational convention |
| to show that this is a SID. The second field is the version number of |
| the SID structure, So far there exists only one version of SIDs, so this |
| field is always 1. The third and fourth fields represent the "authority" |
| which can be thought of as a type or category of SIDs. There are a |
| couple of builtin accounts and accounts with very special meaning which |
| have certain well known values in these third and fourth fields. |
| However, computer and domain SIDs always start with "S-1-5-21". The |
| next three fields, all 32 bit values, represent the unique 96 bit |
| identifier of the computer system. This is a hopefully unique value all |
| over the world, but in practice it's sufficient if the computer SIDs are |
| unique within a single Windows network.</p><p>As you can see in the above example, SIDs of users (and groups) |
| are identical to the computer SID, except for an additional part, the |
| so-called "relative identifier" (RID). So the SID of a user is always |
| uniquely attached to the system on which the account has been generated.</p><p>It's a bit different in domains. The domain has its own SID, and |
| that SID is identical to the SID of the first domain controller, on |
| which the domain is created. Domain user SIDs look exactly like the |
| computer user SIDs, the leading part is just the domain SID and the RID |
| is created when the user is created.</p><p>Ok, consider you created a new domain "bar" on some new domain |
| controller and you would like to create a domain account "johndoe":</p><p>SID of a domain "bar.local":</p><pre class="screen"> |
| S-1-5-21-186985262-1144665072-740312968 |
| </pre><p>SID of a user "johndoe" in the domain "bar.local":</p><pre class="screen"> |
| S-1-5-21-186985262-1144665072-740312968-1207 |
| </pre><p>So you now have two accounts called johndoe, one account |
| created on the machine "foo", one created in the domain "bar.local". |
| Both have different SIDs and not even the RID is the same. How do |
| the systems know it's the same account? After all, the name is |
| the same, right? The answer is, these accounts are <span class="bold"><strong>not</strong></span> identical. All machines on the network will |
| treat these SIDs as identifying two separate accounts. One is |
| "FOO\johndoe", the other one is "BAR\johndoe" or "johndoe@bar.local". |
| Different SID, different account. Full stop. </p><p>Starting with Cygwin 1.7.34, Cygwin uses an automatic, internal |
| translation from Windows SID to POSIX UID/GID. This mechanism, which is |
| the preferred method for the SID<=>UID/GID mapping, is described in |
| detail in <a class="xref" href="ntsec.html#ntsec-mapping" title="Mapping Windows accounts to POSIX accounts">the section called “Mapping Windows accounts to POSIX accounts”</a>.</p><p>Prior to Cygwin 1.7.34, the last part of the SID, the so called |
| "Relative IDentifier" (RID), was by default used as UID and/or GID |
| when you created the <code class="filename">/etc/passwd</code> and |
| <code class="filename">/etc/group</code> files using the |
| <span class="command"><strong><a class="link" href="mkpasswd.html" title="mkpasswd">mkpasswd</a></strong></span> and |
| <span class="command"><strong><a class="link" href="mkgroup.html" title="mkgroup">mkgroup</a></strong></span> tools. |
| These tools as well as reading accounts from <code class="filename">/etc/passwd</code> |
| and <code class="filename">/etc/group</code> files is still present in recent |
| versions of Cygwin, but you should switch to the aforementioned |
| automatic translation, unless you have very specific needs. Again, |
| see <a class="xref" href="ntsec.html#ntsec-mapping" title="Mapping Windows accounts to POSIX accounts">the section called “Mapping Windows accounts to POSIX accounts”</a> for the details.</p><p>Do you still remember the SIDs with special meaning? In offical |
| notation they are called "well-known SIDs". For example, POSIX has no GID |
| for the group of "all users" or "world" or "others". The last three rwx |
| bits in a unix-style permission value just represent the permissions for |
| "everyone who is not the owner or is member of the owning group". |
| Windows has a SID for these poor souls, the "Everyone" SID. Other |
| well-known SIDs represent circumstances under which a process is |
| running, rather than actual users or groups. Here are a few examples |
| for well-known SIDs:</p><pre class="screen"> |
| Everyone S-1-1-0 Simply everyone... |
| Batch S-1-5-3 Processes started via the task |
| scheduler are member of this group. |
| Interactive S-1-5-4 Only processes of users which are |
| logged in via an interactive |
| session are members here. |
| Authenticated Users S-1-5-11 Users which have gone through |
| the authentication process and |
| survived. Anonymously accessing |
| users are not incuded here. |
| SYSTEM S-1-5-18 A special account which has all |
| kinds of dangerous rights, sort of |
| an uber-root account. |
| </pre><p>For a full list please refer to the MSDN document <a class="ulink" href="http://msdn.microsoft.com/en-us/library/aa379649.aspx" target="_top">Well-known |
| SIDs</a>. The Cygwin package called "csih" provides a tool, |
| /usr/lib/csih/getAccountName.exe, which can be used to print the |
| (possibly localized) name for the various well-known SIDS.</p><p>Naturally, well-known SIDs are the same on each machine, so they are |
| not unique to a machine or domain. They have the same meaning across |
| the Windows network.</p><p>Additionally, there are a couple of well-known builtin groups, |
| which have the same SID on every machine and which have certain user |
| rights by default:</p><pre class="screen"> |
| administrators S-1-5-32-544 |
| users S-1-5-32-545 |
| guests S-1-5-32-546 |
| ... |
| </pre><p>For instance, every account is usually member in the "Users" |
| group. All administrator accounts are member of the "Administrators" |
| group. That's all about it as far as single machines are involved. In |
| a domain environment it's a bit more tricky. Since these SIDs are not |
| unique to a machine, every domain user and every domain group can be a |
| member of these well known groups. Consider the domain group "Domain |
| Admins". This group is by default in the "Administrators" group. Let's |
| assume the above computer called "foo" is a member machine of the domain |
| "bar.local". If you stick the user "BAR\johndoe" into the group "Domain |
| Admins", this guy will automatically be a member of the administrators |
| group on "foo" when logging on to "foo". Neat, isn't it?</p><p>Back to ACE and ACL. POSIX is able to create three different |
| permissions, the permissions for the owner, for the group and for the |
| world. In contrast the Windows ACL has a potentially infinite number of |
| members... as long as they fit into 64K. Every member is an ACE. |
| ACE consist of three parts:</p><div class="itemizedlist"><ul class="itemizedlist compact" style="list-style-type: disc; "><li class="listitem"><p>The type of the ACE (allow ACE or deny ACE).</p></li><li class="listitem"><p>Permission bits, 32 of them.</p></li><li class="listitem"><p>The SID for which the permissions are allowed or denied.</p></li></ul></div><p>The two (for us) important types of ACEs are the "access allowed |
| ACE" and the "access denied ACE". As the names imply, the allow ACE |
| tells the system to allow the given permissions to the SID, the deny ACE |
| results in denying the specific permission bits.</p><p>The possible permissions on objects are more detailed than in |
| POSIX. For example, the permission to delete an object is different |
| from the permission to change object data, and even changing object data |
| can be separated into different permission bits for different kind of |
| data. But there's a problem with the definition of a "correct" ACL |
| which disallows mapping of certain POSIX permissions cleanly. See |
| <a class="xref" href="ntsec.html#ntsec-files" title="File permissions">the section called “File permissions”</a>.</p><p>POSIX is able to create only three different permissions? Not quite. |
| Newer operating systems and file systems on POSIX systems also provide |
| access control lists. Two different APIs exist for accessing these |
| ACLs, the Solaris API and the POSIX API. Cygwin implements the original |
| Solaris API to access Windows ACLs in a Unixy way. Online man pages for the |
| Solaris ACL API can be found on |
| <a class="ulink" href="http://docs.oracle.com" target="_top">http://docs.oracle.com</a>. |
| For an overview see <a class="ulink" href="http://docs.oracle.com/cd/E23824_01/html/821-1474/acl-5.html#scrolltoc" target="_top">acl(5)</a>.</p></div><div class="sect2"><div class="titlepage"><div><div><h3 class="title"><a name="ntsec-mapping"></a>Mapping Windows accounts to POSIX accounts</h3></div></div></div><p> |
| For as long as Cygwin has existed, it has stored user and group information in |
| <code class="filename">/etc/passwd</code> and <code class="filename">/etc/group</code> files. |
| Under the assumption that these files would never be too large, the first |
| process in a process tree, as well as every execing process within the tree |
| would parse them into structures in memory. Thus every Cygwin process would |
| contain an expanded copy of the full information from |
| <code class="filename">/etc/passwd</code> and <code class="filename">/etc/group</code>. |
| </p><p> |
| This approach has a few downsides. One of them is that the idea that these |
| files will always be small, is flawed. Another one is that reading the entire |
| file is most of the time entirely useless, since most processes only |
| need information on their own user and the primary group. Last but not |
| least, the passwd and group files have to be maintained separately from |
| the already existing Windows user databases, the local SAM and Active |
| Directory. |
| </p><p> |
| On the other hand, we have to have this mapping between Windows SIDs and |
| POSIX uid/gid values, so we need a mechanism to convert SIDs to uid/gid |
| values and vice versa. |
| </p><p> |
| Microsoft "Services for UNIX" (SFU) (deprecated since Windows 8/Server 2012) |
| never used passwd/group files. Rather, SFU used a fixed, computational mapping |
| between SIDs and POSIX uid/gid which even has Active Directory support. It |
| allows us to generate uid/gid values from SIDs and vice versa. The mechanism is |
| documented, albeit in a confusing way and spread over multiple MSDN articles. |
| </p><p> |
| Starting with Cygwin 1.7.34, Cygwin utilizes an approach inspired by the |
| mapping method as implemented by SFU, with a few differences for backward |
| compatibility and to handle some border cases differently. |
| </p><div class="sect3"><div class="titlepage"><div><div><h4 class="title"><a name="ntsec-mapping-how"></a>Mapping Windows SIDs to POSIX uid/gid values</h4></div></div></div><p> |
| The following description assumes you're comfortable with the concept of |
| Windows SIDs and RIDs. For a brief introduction, see |
| <a class="xref" href="ntsec.html#ntsec-common" title="Brief overview of Windows security">the section called “Brief overview of Windows security”</a>. |
| </p><p> |
| Cygwin's mapping between SIDs and uid/gid values works in two ways. |
| </p><div class="itemizedlist"><ul class="itemizedlist compact" style="list-style-type: disc; "><li class="listitem"><p>Read <code class="filename">/etc/passwd<code class="filename"> and |
| </code>/etc/group</code> files if they exist, just as in the olden |
| days, mainly for backward compatibility.</p></li><li class="listitem"><p>If no files are present, or if an entry is missing in the files, |
| ask Windows.</p></li></ul></div><p> |
| At least, that's the default behaviour now. It will be configurable |
| using a file <code class="filename">/etc/nsswitch.conf</code>, which is discussed in |
| <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch" title="The /etc/nsswitch.conf file">the section called “The <code class="filename">/etc/nsswitch.conf</code> file”</a>. Let's explore the default |
| for now. |
| </p><p> |
| If the passwd or group files are present, they will be scanned on demand as |
| soon as a mapping from SIDs to uid/gid or account names is required. The new |
| mechanism will never read the entire file into memory, but only scan for |
| the requested entry and cache this one in memory. |
| </p><p> |
| If no entry is found, or no passwd or group file was present, Cygwin |
| will ask the OS. |
| </p><div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"><h3 class="title">Note</h3><p> |
| If the first process in a Cygwin process tree determines that no |
| <code class="filename">/etc/passwd</code> or <code class="filename">/etc/group</code> file is |
| present, no other process in the entire process tree will try to read the files |
| later on. This is done for self-preservation. It's rather bad if the uid |
| or gid of a user changes during the lifetime of a process tree. |
| </p><p> |
| For the same reason, if you delete the <code class="filename">/etc/passwd</code> |
| or <code class="filename">/etc/group</code> file, this will be ignored. The passwd |
| and group records read from the files will persist in memory until either a |
| new <code class="filename">/etc/passwd</code> or <code class="filename">/etc/group</code> |
| is created, or you exit all processes in the current process tree. |
| </p><p> |
| See the note in <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch" title="The /etc/nsswitch.conf file">the section called “The <code class="filename">/etc/nsswitch.conf</code> file”</a> for some |
| comprehensive examples. |
| </p></div><p> |
| So if we've drawn a blank reading the files, we're going to ask the OS. |
| First thing, we ask the local machine for the SID or the username. The |
| OS functions |
| <a class="ulink" href="http://msdn.microsoft.com/en-us/library/windows/desktop/aa379166%28v=vs.85%29.aspx" target="_top">LookupAccountSid</a> |
| and |
| <a class="ulink" href="http://msdn.microsoft.com/en-us/library/windows/desktop/aa379159%28v=vs.85%29.aspx" target="_top">LookupAccountName</a> |
| are pretty intelligent. They have all the stuff built in to ask for any |
| account of the local machine, the Active Directory domain of the machine, |
| the Global Catalog of the forest of the domain, as well as any trusted |
| domain of our forest for the information. One OS call and we're |
| practically done... |
| </p><p> |
| Except, the calls only return the mapping between SID, account name and the |
| account's domain. We don't have a mapping to POSIX uid/gid and we're missing |
| information on the user's home dir and login shell. |
| </p><p> |
| Let's discuss the SID<=>uid/gid mapping first. Here's how it works. |
| </p><div class="itemizedlist"><ul class="itemizedlist compact" style="list-style-type: disc; "><li class="listitem"><p> |
| <a class="ulink" href="http://msdn.microsoft.com/en-us/library/aa379649.aspx" target="_top">Well-known |
| SIDs</a> |
| in the NT_AUTHORITY domain of the S-1-5-RID type, or aliases of the |
| S-1-5-32-RID type are mapped to the uid/gid value RID. Examples: |
| </p><pre class="screen"> |
| "SYSTEM" S-1-5-18 <=> uid/gid: 18 |
| "Users" S-1-5-32-545 <=> uid/gid: 545 |
| </pre></li><li class="listitem"><p> |
| Other well-known SIDs in the NT_AUTHORITY domain (S-1-5-X-RID): |
| </p><pre class="screen"> |
| S-1-5-X-RID <=> uid/gid: 0x1000 * X + RID |
| </pre><p>Example:</p><pre class="screen"> |
| "NTLM Authentication" S-1-5-64-10 <=> uid/gid: 0x4000A == 262154 |
| </pre></li><li class="listitem"><p> |
| Other well-known SIDs: |
| </p><pre class="screen"> |
| S-1-X-Y <=> uid/gid: 0x10000 + 0x100 * X + Y |
| </pre><p>Example:</p><pre class="screen"> |
| "LOCAL" S-1-2-0 <=> uid/gid: 0x10200 == 66048 |
| "Creator Group" S-1-3-1 <=> uid/gid: 0x10301 == 66305 |
| </pre></li><li class="listitem"><p> |
| Logon SIDs: The LogonSid of the current user's session is converted to the |
| fixed uid 0xfff == 4095 and named "CurrentSession". Any other LogonSid is |
| converted to the fixed uid 0xffe == 4094 and named "OtherSession". |
| </p></li><li class="listitem"><p> |
| Mandatory Labels: |
| </p><pre class="screen"> |
| S-1-16-RID <=> uid/gid: 0x60000 + RID |
| </pre><p>Example:</p><pre class="screen"> |
| "Medium Mandatory Level" S-1-16-8192 <=> uid/gid: 0x62000 == 401408 |
| </pre></li><li class="listitem"><p> |
| Accounts from the local machine's user DB (SAM): |
| </p><pre class="screen"> |
| S-1-5-21-X-Y-Z-RID <=> uid/gid: 0x30000 + RID |
| </pre><p>Example:</p><pre class="screen"> |
| "Administrator" S-1-5-21-X-Y-Z-500 <=> uid/gid: 0x301f4 == 197108 |
| </pre></li><li class="listitem"><p> |
| Accounts from the machine's primary domain: |
| </p><pre class="screen"> |
| S-1-5-21-X-Y-Z-RID <=> uid/gid: 0x100000 + RID |
| </pre><p>Example:</p><pre class="screen"> |
| "Domain Users" S-1-5-21-X-Y-Z-513 <=> uid/gid: 0x100201 == 1049089 |
| </pre></li><li class="listitem"><p> |
| Accounts from a trusted domain of the machine's primary domain: |
| </p><pre class="screen"> |
| S-1-5-21-X-Y-Z-RID <=> uid/gid: trustPosixOffset(domain) + RID |
| </pre><p> |
| <code class="literal">trustPosixOffset</code>? This needs a bit of explanation. This |
| value exists in Windows domains already since before Active Directory days. |
| What happens is this. If you create a domain trust between two domains, a |
| trustedDomain entry will be added to both databases. It describes how |
| <span class="emphasis"><em>this</em></span> domain trusts the <span class="emphasis"><em>other</em></span> domain. |
| One attribute of a trust is a 32 bit value called |
| <code class="literal">trustPosixOffset</code> For each new trust, |
| <code class="literal">trustPosixOffset</code> will get some automatic value. In recent |
| AD domain implementations, the first trusted domain will get |
| <code class="literal">trustPosixOffset</code> set to 0x80000000. Following domains will |
| get lower values. Unfortunately the domain admins are allowed to set the |
| <code class="literal">trustPosixOffset</code> value for each trusted domain to some |
| arbitrary 32 bit value, no matter what the other |
| <code class="literal">trustPosixOffset</code> are set to, thus allowing any kind of |
| collisions between the <code class="literal">trustPosixOffset</code> values of domains. |
| That's not exactly helpful, but as the user of this value, we have to |
| <span class="emphasis"><em>trust</em></span> the domain admins to set |
| <code class="literal">trustPosixOffset</code> to sensible values, or to keep it at the |
| system chosen defaults. |
| </p><p> |
| So, for the first (or only) trusted domain of your domain, the automatic offset |
| is 0x80000000. An example for a user of that trusted domain is |
| </p><pre class="screen"> |
| S-1-5-21-X-Y-Z-1234 <=> uid/gid 0x800004d2 == 2147484882 |
| </pre><p> |
| There's one problem with this approach. Assuming you're running in the context |
| of a local SAM user on a domain member machine. Local users don't have the |
| right to fetch this kind of domain information from the DC, they'll get |
| permission denied. In this case Cygwin will fake a sensible |
| <code class="literal">trustPosixOffset</code> value. |
| </p><p> |
| Another problem is if the AD administrators chose an unreasonably small |
| <code class="literal">trustPosixOffset</code> value. Anything below the hexadecimal |
| value 0x100000 (the POSIX offset of the primary domain) is bound to produce |
| collisions with system accounts as well as local accounts. The right thing |
| to do in this case is to notify your administrator of the problem and to ask |
| for moving the offset to a more reasonable value. However, to reduce the |
| probability for collisions, Cygwin overrides this offset with a sensible |
| fixed replacement offset. |
| </p></li><li class="listitem"><p> |
| Local accounts from another machine in the network: |
| </p><p> |
| There's no SID<=>uid/gid mapping implemented for this case. The problem |
| is, there's no way to generate a bijective mapping. There's no central place |
| which keeps an analogue of the <code class="literal">trustPosixOffset</code>, and there's |
| the additional problem that the |
| <a class="ulink" href="http://msdn.microsoft.com/en-us/library/windows/desktop/aa379166%28v=vs.85%29.aspx" target="_top">LookupAccountSid</a> |
| and |
| <a class="ulink" href="http://msdn.microsoft.com/en-us/library/windows/desktop/aa379159%28v=vs.85%29.aspx" target="_top">LookupAccountName</a> |
| functions cannnot resolve the SIDs, unless they know the name of the machine |
| this SID comes from. And even then it will probably suffer a |
| <code class="literal">Permission denied</code> error when trying to ask the machine |
| for its local account. |
| </p></li></ul></div><p> |
| Now we have a semi-bijective mapping between SIDs and POSIX uid/gid values, |
| but given that we have potentially users and groups in different domains having |
| the same name, how do we uniquely distinguish between them by name? Well, we |
| can do that by making their names unique in a per-machine way. Dependent on |
| the domain membership of the account, and dependent of the machine being a |
| domain member or not, the user and group names will be generated using a domain |
| prefix and a separator character between domain and account name. |
| The separator character is the plus sign, <code class="literal">+</code>. |
| </p><div class="itemizedlist"><ul class="itemizedlist compact" style="list-style-type: disc; "><li class="listitem"><p> |
| Well-known and builtin accounts will be named as in Windows: |
| </p><pre class="screen"> |
| "SYSTEM", "LOCAL", "Medium Mandatory Level", ... |
| </pre></li><li class="listitem"><p> |
| If the machine is not a domain member machine, only local accounts can be resolved |
| into names, so for ease of use, just the account names are used as Cygwin |
| user/group names: |
| </p><pre class="screen"> |
| "corinna", "bigfoot", "None", ... |
| </pre></li><li class="listitem"><p> |
| If the machine is a domain member machine, all accounts from the primary domain |
| of the machine are mapped to Cygwin names without domain prefix: |
| </p><pre class="screen"> |
| "corinna", "bigfoot", "Domain Users", ... |
| </pre><p> |
| while accounts from other domains are prepended by their domain: |
| </p><pre class="screen"> |
| "DOMAIN1+corinna", "DOMAIN2+bigfoot", "DOMAIN3+Domain Users", ... |
| </pre></li><li class="listitem"><p> |
| Local machine accounts of a domain member machine get a Cygwin user name the |
| same way as accounts from another domain: The local machine name gets |
| prepended: |
| </p><pre class="screen"> |
| "MYMACHINE+corinna", "MYMACHINE+bigfoot", "MYMACHINE+None", ... |
| </pre></li><li class="listitem"><p> |
| If LookupAccountSid fails, Cygwin checks the accounts against the known trusted |
| domains. If the account is from one of the trusted domains, an artificial |
| account name is created. It consists of the domain name, and a special name |
| created from the account RID: |
| </p><pre class="screen"> |
| "MY_DOM+User(1234)", "MY_DOM+Group(5678)" |
| </pre><p> |
| Otherwise we know nothing about this SID, so it will be mapped to the |
| fake accounts <code class="literal">Unknown+User</code>/<code class="literal">Unknown+Group</code> |
| with uid/gid -1. |
| </p></li></ul></div></div><div class="sect3"><div class="titlepage"><div><div><h4 class="title"><a name="ntsec-mapping-caching"></a>Caching account information</h4></div></div></div><p> |
| The information fetched from the Windows account database or the |
| <code class="filename">/etc/passwd</code> and <code class="filename">/etc/group</code> files is |
| cached by the process. The cached information is inherited by Cygwin child |
| processes. A Cygwin process invoked from a Windows command, such as CMD.exe, |
| will start a new Cygwin process tree and the caching starts from scratch |
| (unless <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span> is |
| running, but read on). |
| </p><p> |
| While usually working fine, this has some drawbacks. Consider a shell calling |
| <span class="command"><strong>id</strong></span>. <span class="command"><strong>id</strong></span> fetches all group information |
| from the current token and caches them. Unfortunately <span class="command"><strong>id</strong></span> |
| doesn't start any child processes, so the information is lost as soon as |
| <span class="command"><strong>id</strong></span> exits. |
| </p><p> |
| But there's another caching mechanism available. If |
| <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span> is running |
| it will provide passwd and group entry caching for all processes in every Cygwin |
| process tree started after |
| <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span>. So, if |
| you start a Cygwin Terminal and |
| <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span> is running |
| at the time, <span class="command"><strong>mintty</strong></span>, the shell, and all child processes will |
| use <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span> caching. |
| If you start a Cygwin Terminal and |
| <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span> is not |
| running at the time, none of the processes started inside this terminal window |
| will use <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span> |
| caching. |
| </p><p> |
| The advantage of |
| <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span> caching is |
| that it's system-wide and, as long as |
| <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span> is running, |
| unforgetful. Every Cygwin process on the system will have the |
| <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span> cache at |
| its service. Additionally, all information requested from |
| <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span> once, will |
| be cached inside the process itself and, again, propagated to child processes. |
| </p><p> |
| If you automatically start Cygwin processes as Windows services at system |
| startup, you may wish to consider starting |
| <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span> first in |
| order to take advantage of this system-wide caching. To assure that |
| <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span> has started |
| prior to starting <span class="command"><strong>cygsshd</strong></span> or other Cygwin services, you may |
| wish to create service startup dependencies. |
| <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">Cygserver</a></strong></span> should |
| probably wait for Windows TCPIP and AFD services before it starts, and then |
| other Cygwin process should start after |
| <span class="command"><strong><a class="link" href="using-cygserver.html" title="Cygserver">cygserver</a></strong></span>. Example |
| Windows commands to accomplish this (after the services already exist) are |
| shown below. You will need an administrative prompt to run the |
| <span class="command"><strong>sc config</strong></span> commands. |
| </p><pre class="screen"> |
| # Delay Cygserver until TCPIP and AFD have started |
| # Note the (odd) required space character after "depend=" |
|
|
| sc config cygserver depend= tcpip/afd |
|
|
| # Delay sshd until after Cygserver has started |
| # Again note the (odd) required space character after "depend=" |
|
|
| sc config cygsshd depend= cygserver |
|
|
| # View the Cygserver service details |
|
|
| sc qc cygserver |
| </pre><p> |
| Note that this <span class="command"><strong>sc config</strong></span> command |
| <span class="emphasis"><em>replaces</em></span> any existing dependencies. The above changes |
| will not impact the running instance, only future instances. |
| </p><pre class="screen"> |
| # To remove all dependencies from the cygserver service |
|
|
| sc config cygserver depend= / |
| </pre></div><div class="sect3"><div class="titlepage"><div><div><h4 class="title"><a name="ntsec-mapping-passwdinfo"></a>Cygwin user names, home dirs, login shells</h4></div></div></div><p> |
| Obviously, if you don't maintain <code class="filename">passwd</code> and |
| <code class="filename">group</code> files, you need to have a way to maintain the other |
| fields of a passwd entry as well. A couple of things come to mind: |
| </p><div class="itemizedlist"><ul class="itemizedlist compact" style="list-style-type: disc; "><li class="listitem"><p> |
| You want to use a Cygwin username different from your Windows username. |
| </p><div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"><h3 class="title">Note</h3><p> |
| This is only supported via <code class="filename">/etc/passwd</code>. A Cygwin |
| username maintained in the Windows user databases would require very costly |
| (read: slow) search operations. |
| </p></div></li><li class="listitem"><p> |
| You want to change the primary group of a user. For AD accounts this is |
| not supported. The primary group of a user is always the Windows |
| primary group set in Active Directory and can't be changed. For SAM |
| accounts, you can add the primary group to the SAM |
| <code class="literal">description</code> field of the user. See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-desc" title="The desc schema">the section called “The <code class="literal">desc</code> schema”</a> for more info. |
| </p></li><li class="listitem"><p> |
| You want a home dir different from the default |
| <code class="filename">/home/$USERNAME</code>. |
| </p></li><li class="listitem"><p> |
| You want to specify a different login shell than <code class="filename">/bin/bash</code>. |
| </p></li><li class="listitem"><p> |
| You want to add specific content to the pw_gecos field. |
| </p></li></ul></div><p> |
| For simple needs you can create <code class="filename">/etc/passwd</code> and/or |
| <code class="filename">/etc/group</code> files with entries for your account |
| and tweak that. |
| </p><p> |
| For bigger installations, maintaining per-client files is rather troublesome. |
| Also, no two environments are the same, so the needs are pretty different. |
| Therefore Cygwin supports configuring how to fetch home directory, |
| login shell, and gecos information in /etc/nsswitch.conf. See the next |
| section for detailed information how to configure Cygwin's account handling. |
| </p></div><div class="sect3"><div class="titlepage"><div><div><h4 class="title"><a name="ntsec-mapping-nsswitch"></a>The <code class="filename">/etc/nsswitch.conf</code> file</h4></div></div></div><p> |
| On Linux and some other UNIXy OSes, we have a file called |
| <a class="ulink" href="http://linux.die.net/man/5/nsswitch.conf" target="_top">/etc/nsswitch.conf</a>. |
| Among other things, it determines how passwd and group entries are generated. |
| That's what Cygwin now provides as well. |
| </p><p> |
| The <code class="filename">/etc/nsswitch.conf</code> file is optional. If you don't |
| have one, Cygwin uses sensible defaults. |
| </p><div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"><h3 class="title">Note</h3><p> |
| The <code class="filename">/etc/nsswitch.conf</code> file is read exactly once by |
| the first process of a Cygwin process tree. If there was no |
| <code class="filename">/etc/nsswitch.conf</code> file when this first process started, |
| then no other process in the running Cygwin process tree will try to read the |
| file. |
| </p><p> |
| If you create or change <code class="filename">/etc/nsswitch.conf</code>, you have to |
| restart all Cygwin processes that need to see the change. If the process |
| you want to see the change is a child of another process, you need to restart |
| all of that process's parents, too. |
| </p><p> |
| For example, if you run <span class="command"><strong>vim</strong></span> inside the default Cygwin |
| Terminal, <span class="command"><strong>vim</strong></span> is a child of your shell, which is a child |
| of <span class="command"><strong>mintty</strong></span>. If you edit |
| <code class="filename">/etc/nsswitch.conf</code> in that <span class="command"><strong>vim</strong></span> |
| instance, your shell won't immediately see the change, nor will |
| <span class="command"><strong>vim</strong></span> if you restart it from that same shell instance. |
| This is because both are getting their nsswitch information from their |
| ancestor, <span class="command"><strong>mintty</strong></span>. You have to start a fresh terminal |
| window for the change to take effect. |
| </p><p> |
| By contrast, if you leave that Cygwin Terminal window open after making the |
| change to <code class="filename">/etc/nsswitch.conf</code>, then restart a Cygwin |
| service like <span class="command"><strong>cron</strong></span>, <span class="command"><strong>cron</strong></span> will see the |
| change, because it is not a child of <span class="command"><strong>mintty</strong></span> or any other |
| Cygwin process. (Technically, it is a child of <span class="command"><strong>cygrunsrv</strong></span>, |
| but that instance also restarts when you restart the service.) |
| </p><p> |
| The reason we point all this out is that the requirements for restarting |
| things are not quite as stringent as when you replace |
| <code class="filename">cygwin1.dll</code>. If you have three process trees, you have |
| three independent copies of the nsswitch information. If you start a fresh |
| process tree, it will see the changes. As long as any process in an existing |
| process tree remains running, all processes in that tree will continue to use |
| the old information. |
| </p></div><p> |
| So, what settings can we perform with <code class="filename">/etc/nsswitch.conf</code>? |
| Let's start with an example <code class="filename">/etc/nsswitch.conf</code> file |
| set up to all default values: |
| </p><pre class="screen"> |
| # /etc/nsswitch.conf |
| passwd: files db |
| group: files db |
| db_enum: cache builtin |
| db_home: /home/%U |
| db_shell: /bin/bash |
| db_gecos: <empty> |
| </pre><div class="sect4"><div class="titlepage"><div><div><h5 class="title"><a name="ntsec-mapping-nsswitch-syntax"></a>The <code class="filename">/etc/nsswitch.conf</code> syntax</h5></div></div></div><p> |
| The first line, starting with a hash <code class="literal">#</code> is a comment. |
| The hash character starts a comment, just as in shell scripts. Everything |
| up to the end of the line is ignored. So this: |
| </p><pre class="screen"> |
| foo: bar # baz |
| </pre><p> |
| means, set "foo" to value "bar", ignore everything after the hash. |
| </p><p> |
| The other lines define the available settings. The first word up to a |
| colon is a keyword. Note that the colon <span class="emphasis"><em>must</em></span> follow |
| immediately after the keyword. This is a valid line: |
| </p><pre class="screen"> |
| foo: bar |
| </pre><p> |
| This is not valid: |
| </p><pre class="screen"> |
| foo : bar |
| </pre><p> |
| Apart from this restriction, the remainder of the line can have as |
| many spaces and TABs as you like. |
| </p><p> |
| When the same keyword occurs multiple times, the last one wins, as if the |
| previous ones were ignored. |
| </p></div><div class="sect4"><div class="titlepage"><div><div><h5 class="title"><a name="ntsec-mapping-nsswitch-pwdgrp"></a>The <code class="literal">passwd:</code> and <code class="literal">group:</code> settings</h5></div></div></div><p> |
| The two lines starting with the keywords <code class="literal">passwd:</code> and |
| <code class="literal">group:</code> define where Cygwin gets its passwd and group |
| information from. <code class="literal">files</code> means, fetch the information |
| from the corresponding file in the /etc directory. <code class="literal">db</code> |
| means, fetch the information from the Windows account databases, the SAM |
| for local accounts, Active Directory for domain account. Examples: |
| </p><pre class="screen"> |
| passwd: files |
| </pre><p> |
| Read passwd entries only from /etc/passwd. |
| </p><pre class="screen"> |
| group: db |
| </pre><p> |
| Read group entries only from SAM/AD. |
| </p><pre class="screen"> |
| group: files # db |
| </pre><p> |
| Read group entries only from <code class="filename">/etc/group</code> |
| (<code class="literal">db</code> is only a comment). |
| </p><pre class="screen"> |
| passwd: files db |
| </pre><p> |
| Read passwd entries from <code class="filename">/etc/passwd</code>. If a user account |
| isn't found, try to find it in SAM or AD. This is the default for both, |
| passwd and group information. |
| </p><pre class="screen"> |
| group: db files |
| </pre><p> |
| This is a valid entry, but the order will be ignored by Cygwin. If both |
| settings, <code class="literal">files</code> and <code class="literal">db</code> are specified, |
| Cygwin will always try the files first, then the db. |
| </p><p> |
| <code class="literal">passwd:</code> and <code class="literal">group:</code> are the two basic |
| settings defining where to get the account information from. The following |
| settings starting with <code class="literal">db_</code> define certain aspects of the |
| Windows account database search and how to generate <code class="literal">passwd</code> |
| and <code class="literal">group</code> information from the database. |
| </p></div><div class="sect4"><div class="titlepage"><div><div><h5 class="title"><a name="ntsec-mapping-nsswitch-enum"></a>The <code class="literal">db_enum:</code> setting</h5></div></div></div><p> |
| <code class="literal">db_enum:</code> defines the depth of a database search, if an |
| application calls one of the enumeration functions |
| <a class="ulink" href="http://linux.die.net/man/3/getpwent" target="_top">getpwent</a> |
| or <a class="ulink" href="http://linux.die.net/man/3/getgrent" target="_top">getgrent</a>. |
| The problem with these functions is, they neither allow to define how many |
| entries will be enumerated when calling them in a loop, nor do they |
| allow to add some filter criteria. They were designed back in the days, |
| when only <code class="filename">/etc/passwd</code> and <code class="filename">/etc/group</code> |
| files existed and the number of user accounts on a typical UNIX system was |
| seldomly a three-digit number. |
| </p><p> |
| These days, with user and group databases sometimes going in the |
| six-digit range, they are a potential burden. For that reason, Cygwin |
| does not enumerate all user or group accounts by default, but rather |
| just a very small list, consisting only of the accounts cached in memory |
| by the current process, as well as a handful of predefined builtin |
| accounts. |
| </p><p> |
| <code class="literal">db_enum:</code> allows to specify the accounts to enumerate in a |
| fine-grained manner. It takes a list of sources as argument: |
| </p><pre class="screen"> |
| db_enum: source1 source2 ... |
| </pre><p> |
| The recognized sources are the following: |
| </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">none</code></span></dt><dd>No output from |
| <code class="function">getpwent</code>/<code class="function">getgrent</code> |
| at all.</dd><dt><span class="term"><code class="literal">all</code></span></dt><dd>The opposite. Enumerates accounts from all known sources, |
| including all trusted domains.</dd><dt><span class="term"><code class="literal">cache</code></span></dt><dd>Enumerate all accounts currently cached in memory.</dd><dt><span class="term"><code class="literal">builtin</code></span></dt><dd>Enumerate the predefined builtin accounts for backward |
| compatibility. These are five passwd accounts (SYSTEM, |
| LocalService, NetworkService, Administrators, TrustedInstaller) |
| and two group accounts (SYSTEM and TrustedInstaller).</dd><dt><span class="term"><code class="literal">files</code></span></dt><dd>Enumerate the accounts from <code class="filename">/etc/passwd</code> or |
| <code class="filename">/etc/group</code>.</dd><dt><span class="term"><code class="literal">local</code></span></dt><dd>Enumerate all accounts from the local SAM.</dd><dt><span class="term"><code class="literal">primary</code></span></dt><dd>Enumerate all accounts from the primary domain.</dd><dt><span class="term"><code class="literal">alltrusted</code></span></dt><dd>Enumerate all accounts from all trusted domains.</dd><dt><span class="term"><code class="literal">some.domain</code></span></dt><dd>Enumerate all accounts from the trusted domain some.domain. The |
| trusted domain can be given as Netbios flat name (MY_DOMAIN) or |
| as dns domain name (my_domain.corp). In contrast to the |
| aforementioned fixed source keywords, distinct domain names are |
| caseinsensitive. Only domains which are actually trusted domains |
| are enumerated. Unknown domains are simply ignored.</dd></dl></div><p> |
| Please note that <code class="function">getpwent</code>/<code class="function">getgrent</code> |
| do <span class="emphasis"><em>not</em></span> test if an account was already listed from another |
| source, so an account can easily show up twice or three times. Such a test |
| would be rather tricky, nor does the Linux implementation perform such test. |
| Here are a few examples for <code class="filename">/etc/nsswitch.conf</code>: |
| </p><pre class="screen"> |
| db_enum: none |
| </pre><p> |
| No output from <code class="function">getpwent</code>/<code class="function">getgrent</code> |
| at all. The first call to the function immediately returns a NULL pointer. |
| </p><pre class="screen"> |
| db_enum: cache files |
| </pre><p> |
| Enumerate all accounts cached by the current process, plus all entries |
| from either the /etc/passwd or /etc/group file. |
| </p><pre class="screen"> |
| db_enum: cache local primary |
| </pre><p> |
| Enumerate all accounts cached by the current process, all accounts from the SAM |
| of the local machine, and all accounts from the primary domain of the machine. |
| </p><pre class="screen"> |
| db_enum: local primary alltrusted |
| </pre><p> |
| Enumerate the accounts from the machine's SAM, from the primary domain of the |
| machine, and from all trusted domains. |
| </p><pre class="screen"> |
| db_enum: primary domain1.corp sub.domain.corp domain2.net |
| </pre><p> |
| Enumerate the accounts from the primary domain and from the domains |
| domain1.corp, sub.domain.corp and domain2.net. |
| </p><pre class="screen"> |
| db_enum: all |
| </pre><p> |
| Enumerate everything and the kitchen sink. |
| </p></div><div class="sect4"><div class="titlepage"><div><div><h5 class="title"><a name="ntsec-mapping-nsswitch-passwd"></a>Settings defining how to create the <code class="literal">passwd</code> entry</h5></div></div></div><p> |
| <code class="filename">/etc/nsswitch.conf</code> supports three settings to configure |
| where to get the pw_dir, pw_shell, and pw_gecos content of a |
| <code class="literal">passwd</code> entry from: |
| </p><pre class="screen"> |
| db_home: schema... # Define how to fetch the pw_dir entry. |
| db_shell: schema... # Define how to fetch the pw_shell entry. |
| db_gecos: schema... # Define how to fetch the pw_gecos entry. |
| </pre> |
|
|
| "schema..." is a list of up to four space-separated schemata: |
|
|
| <pre class="screen"> |
| db_FOO: schema1 schema2 ... |
| </pre><p> |
| When generating a passwd entry, Cygwin tries the schemata in order. If |
| the first schema returns an empty string, it skips to the second, and so |
| on. Schemata only supported on AD are silently skipped for SAM accounts |
| and on non-AD machines. |
| </p><p> |
| Five schemata are predefined, two schemata are variable. The predefined |
| schemata are the following: |
| </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">windows</code></span></dt><dd>Utilizes typical Windows settings. Supported for AD and SAM |
| accounts.</dd><dt><span class="term"><code class="literal">cygwin</code></span></dt><dd>Utilizes the cygwinUser AD schema extension. This schema |
| extension is available via a schema extension file |
| <code class="filename">/usr/share/cygwin/cygwin.ldif</code>. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-cygwin" title="The cygwin schema">the section called “The <code class="literal">cygwin</code> schema”</a> for |
| more information.</dd><dt><span class="term"><code class="literal">unix</code></span></dt><dd>Utilizes the posixAccount schema attributes per |
| <a class="ulink" href="https://tools.ietf.org/html/rfc2307" target="_top">RFC 2307</a>. |
| The posixAccount schema is available by default in Windows |
| Server, but typically only utilized when installing the |
| Active Directory "Server for NIS" feature (which is deprecated |
| since Server 2012 R2). |
| See also <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-posix" title="The unix schema">the section called “The <code class="literal">unix</code> schema”</a>. |
| </dd><dt><span class="term"><code class="literal">desc</code></span></dt><dd>Utilizes XML-style attributes in the description attribute. |
| Supported for AD and SAM accounts. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-desc" title="The desc schema">the section called “The <code class="literal">desc</code> schema”</a> |
| for a more detailed description.</dd><dt><span class="term"><code class="literal">env</code></span></dt><dd>Utilizes the user's environment. This schema is only supported |
| for setting the home directory yet. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-home" title="The db_home setting">the section called “The <code class="literal">db_home</code> setting”</a> for |
| the description.</dd></dl></div><p> |
| The variable schemata are as follows. Note that the leading characters |
| (<code class="literal">@</code> and <code class="literal">/</code>) are an integral part of the |
| definition. |
| </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">@ad_attribute</code></span></dt><dd><code class="literal">ad_attribute</code> is any arbitrary AD attribute |
| name which should (ideally) be available in the User class or |
| in any attached auxiliary class. It's always treated as a |
| single string argument. Only the first string of a multi-string |
| attributes will be read.</dd><dt><span class="term"><code class="literal">/path</code></span></dt><dd>An arbitrary string, typically a path. The leading slash is |
| required. Given that a single, fixed path for all users |
| only makes marginal sense, the /path schema supports wildcards. |
| A wildcard is a per-cent (<code class="literal">%</code>) character, |
| followed by another character giving the meaning. The supported |
| wildcard characters are: |
| |
| <div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">%u</code></span></dt><dd>The Cygwin username (that's lowercase |
| <code class="literal">u</code>).</dd><dt><span class="term"><code class="literal">%U</code></span></dt><dd>The Windows username (that's uppercase |
| <code class="literal">U</code>).</dd><dt><span class="term"><code class="literal">%D</code></span></dt><dd>Windows domain in NetBIOS style.</dd><dt><span class="term"><code class="literal">%H</code></span></dt><dd>Windows home directory in POSIX style. Note that, |
| for the <code class="literal">db_home:</code> setting, this |
| only makes sense right after the preceeding slash, |
| as in |
| <pre class="screen"> |
| db_home: /%H/cygwin |
| </pre></dd><dt><span class="term"><code class="literal">%_</code></span></dt><dd>Since space and TAB characters are used to separate |
| the schemata, a space in the filename has to be |
| given as <code class="literal">%_</code> (that's an |
| underscore).</dd><dt><span class="term"><code class="literal">%%</code></span></dt><dd>A per-cent character.</dd></dl></div><p>Any other <code class="literal">%X</code> expression is treated as if |
| the character <code class="literal">X</code> has been given alone.</p></dd></dl></div><p> |
| The exact meaning of a schema depends on the setting it's used for. The |
| following sections explain the settings in detail. |
| </p></div><div class="sect4"><div class="titlepage"><div><div><h5 class="title"><a name="ntsec-mapping-nsswitch-home"></a>The <code class="literal">db_home</code> setting</h5></div></div></div><p> |
| The <code class="literal">db_home:</code> setting defines how Cygwin fetches the user's |
| home directory, or, more precise, the content of the <code class="literal">pw_dir</code> |
| member of the user's passwd entry. The following list describes the meaning |
| of each schema when used with <code class="literal">db_home:</code> |
| </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">windows</code></span></dt><dd>The user's home directory is set to the same directory which is |
| used as Windows home directory. This is the |
| <code class="literal">homeDirectory</code> AD attribute. |
| For SAM accounts, this is equivalent to the "Home folder" setting |
| in SAM. If both attributes are unset, Cygwin falls back to the |
| user's local profile directory, typically something along the |
| lines of <code class="filename">C:\Users\$USERNAME</code>. Of course, the |
| Windows directory is converted to POSIX-style by Cygwin. |
| </dd><dt><span class="term"><code class="literal">cygwin</code></span></dt><dd>AD only: The user's home directory is set to the POSIX path given |
| in the <code class="literal">cygwinHome</code> attribute from the |
| <code class="literal">cygwinUser</code> auxiliary class. |
| See also <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-cygwin" title="The cygwin schema">the section called “The <code class="literal">cygwin</code> schema”</a>. |
| </dd><dt><span class="term"><code class="literal">unix</code></span></dt><dd>AD only: The user's home directory is set to the POSIX path given |
| in the <code class="literal">unixHomeDirectory</code> attribute from the |
| <code class="literal">posixAccount</code> auxiliary class. |
| See also <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-posix" title="The unix schema">the section called “The <code class="literal">unix</code> schema”</a>. |
| </dd><dt><span class="term"><code class="literal">desc</code></span></dt><dd>The user's home directory is set to the POSIX path given in the |
| home="..." XML-alike setting in the user's |
| <code class="literal">description</code> attribute in SAM or AD. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-desc" title="The desc schema">the section called “The <code class="literal">desc</code> schema”</a> |
| for a detailed description.</dd><dt><span class="term"><code class="literal">env</code></span></dt><dd>Derives the home directory of the current user from the |
| environment variable <code class="literal">HOME</code> (falling back to |
| <code class="literal">HOMEDRIVE\HOMEPATH</code> and |
| <code class="literal">USERPROFILE</code>, in that order). This is faster |
| than the <code class="literal">windows</code> schema at the |
| expense of determining only the current user's home directory |
| correctly. This schema is skipped for any other account. |
| </dd><dt><span class="term"><code class="literal">@ad_attribute</code></span></dt><dd>AD only: The user's home directory is set to the path given |
| in the <code class="literal">ad_attribute</code> attribute. The path |
| can be given as Windows or POSIX path.</dd><dt><span class="term"><code class="literal">/path</code></span></dt><dd>The user's home directory is set to the given POSIX path. |
| Remember the wildcards described in |
| <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-passwd" title="Settings defining how to create the passwd entry">the section called “Settings defining how to create the <code class="literal">passwd</code> entry”</a>.</dd><dt><span class="term">Fallback</span></dt><dd>If none of the schemes given for <code class="literal">db_home:</code> |
| define a non-empty directory, the user's home directory is set to |
| <code class="filename">/home/$USERNAME</code>.</dd></dl></div><p> |
| <code class="literal">db_home:</code> defines no default schemata. If this setting is not |
| present in <code class="filename">/etc/nsswitch.conf</code>, the aforementioned fallback |
| takes over, which is equivalent to a <code class="filename">/etc/nsswitch.conf</code> |
| settting of |
| </p><pre class="screen"> |
| db_home: /home/%U |
| </pre></div><div class="sect4"><div class="titlepage"><div><div><h5 class="title"><a name="ntsec-mapping-nsswitch-shell"></a>The <code class="literal">db_shell</code> setting</h5></div></div></div><p> |
| The <code class="literal">db_shell:</code> setting defines how Cygwin fetches the user's |
| login shell, the content of the <code class="literal">pw_shell</code> member of the |
| user's passwd entry. The following list describes the meaning of each schema |
| when used with <code class="literal">db_shell:</code> |
| </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">windows</code></span></dt><dd>The <code class="literal">windows</code> schema is ignored for now. |
| The logical choice would be CMD, but that introduces some |
| problems, for instance the fact that CMD has no concept of |
| running as <code class="literal">login shell</code>. This may change |
| in future.</dd><dt><span class="term"><code class="literal">cygwin</code></span></dt><dd>AD only: The user's home directory is set to the POSIX path given |
| in the <code class="literal">cygwinShell</code> attribute from the |
| <code class="literal">cygwinUser</code> auxiliary class. |
| See also <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-cygwin" title="The cygwin schema">the section called “The <code class="literal">cygwin</code> schema”</a>. |
| </dd><dt><span class="term"><code class="literal">unix</code></span></dt><dd>AD only: The user's login shell is set to the POSIX path given |
| in the <code class="literal">loginShell</code> attribute from the |
| <code class="literal">posixAccount</code> auxiliary class. |
| See also <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-posix" title="The unix schema">the section called “The <code class="literal">unix</code> schema”</a>. |
| </dd><dt><span class="term"><code class="literal">desc</code></span></dt><dd>The user's login shell is set to the POSIX path given in the |
| shell="..." XML-alike setting in the user's |
| <code class="literal">description</code> attribute in SAM or AD. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-desc" title="The desc schema">the section called “The <code class="literal">desc</code> schema”</a> |
| for a detailed description.</dd><dt><span class="term"><code class="literal">@ad_attribute</code></span></dt><dd>AD only: The user's login shell is set to the path given |
| in the <code class="literal">ad_attribute</code> attribute. The path |
| can be given as Windows or POSIX path.</dd><dt><span class="term"><code class="literal">/path</code></span></dt><dd>The user's login shell is set to the given POSIX path. |
| Albeit not being as important here, the wildcards described in |
| <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-passwd" title="Settings defining how to create the passwd entry">the section called “Settings defining how to create the <code class="literal">passwd</code> entry”</a> |
| are also available for specifying a login shell path.</dd><dt><span class="term">Fallback</span></dt><dd>If none of the schemes given for <code class="literal">db_shell:</code> |
| define a non-empty pathname, the user's login shell is set to |
| <code class="filename">/bin/bash</code>.</dd></dl></div><p> |
| <code class="literal">db_shell:</code> defines no default schemata. If this setting is |
| not present in <code class="filename">/etc/nsswitch.conf</code>, the aforementioned |
| fallback takes over, which is equivalent to a |
| <code class="filename">/etc/nsswitch.conf</code> settting of |
| </p><pre class="screen"> |
| db_shell: /bin/bash |
| </pre></div><div class="sect4"><div class="titlepage"><div><div><h5 class="title"><a name="ntsec-mapping-nsswitch-gecos"></a>The <code class="literal">db_gecos</code> setting</h5></div></div></div><p> |
| The <code class="literal">db_gecos:</code> setting defines how to fetch additional |
| content for the <code class="literal">pw_gecos</code> member of the user's passwd entry. |
| There's always a fixed, Cygwin-specific part in the <code class="literal">pw_gecos</code> |
| field for identifying the account. However, an administrator might want to |
| add informative content like, for instance, the user's full name. That's |
| what the <code class="literal">db_gecos:</code> setting is for. |
| The following list describes the meaning of each schema when used with |
| <code class="literal">db_gecos:</code> |
| </p><div class="variablelist"><dl class="variablelist"><dt><span class="term"><code class="literal">windows</code></span></dt><dd>Add the AD <code class="literal">displayName</code> attribute or, for |
| SAM accounts, the "Full name" entry to the |
| <code class="literal">pw_gecos</code> field.</dd><dt><span class="term"><code class="literal">cygwin</code></span></dt><dd>AD only: The content of the <code class="literal">cygwinGecos</code> |
| attribute from the <code class="literal">cygwinUser</code> auxiliary class |
| is added to <code class="literal">pw_gecos</code>. |
| See also <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-cygwin" title="The cygwin schema">the section called “The <code class="literal">cygwin</code> schema”</a>. |
| </dd><dt><span class="term"><code class="literal">unix</code></span></dt><dd>AD only: The content of the <code class="literal">gecos</code> attribute |
| from the <code class="literal">posixAccount</code> auxiliary class |
| is added to <code class="literal">pw_gecos</code>. |
| See also <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-posix" title="The unix schema">the section called “The <code class="literal">unix</code> schema”</a>. |
| </dd><dt><span class="term"><code class="literal">desc</code></span></dt><dd>The content of the gecos="..." XML-alike setting in the user's |
| <code class="literal">description</code> attribute in SAM or AD is added |
| to <code class="literal">pw_gecos</code>. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-desc" title="The desc schema">the section called “The <code class="literal">desc</code> schema”</a> |
| for a detailed description.</dd><dt><span class="term"><code class="literal">@ad_attribute</code></span></dt><dd>AD only: The content of the <code class="literal">ad_attribute</code> |
| attribute is added to <code class="literal">pw_gecos</code>.</dd><dt><span class="term"><code class="literal">/path</code></span></dt><dd>The string following the slash is added to |
| <code class="literal">pw_gecos</code>. Here, the wildcards described in |
| <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-passwd" title="Settings defining how to create the passwd entry">the section called “Settings defining how to create the <code class="literal">passwd</code> entry”</a> |
| may come in handy.</dd><dt><span class="term">Fallback</span></dt><dd>If none of the schemes given for <code class="literal">db_gecos:</code> |
| define a non-empty string, nothing is added to |
| <code class="literal">pw_gecos</code>.</dd></dl></div><p> |
| <code class="literal">db_gecos:</code> defines no default schemata. |
| </p></div><div class="sect4"><div class="titlepage"><div><div><h5 class="title"><a name="ntsec-mapping-nsswitch-cygwin"></a>The <code class="literal">cygwin</code> schema</h5></div></div></div><p> |
| The <code class="literal">cygwin</code> schema is based on a Cygwin-specific Active |
| Directory schema extension. Using this schema extension allows to maintain |
| Cygwin-specific settings entirely within AD, without colliding with any other |
| schema. |
| </p><p> |
| The cygwin schema extension is available in a default Cygwin installation |
| in the file <code class="filename">/usr/share/cygwin/cygwin.ldif</code>. To install |
| the schema extension, you have to be schema admin, and you have to run the |
| <span class="command"><strong>ldifde</strong></span> command on the schema master. The installation |
| itself is rather simple. Assuming you're schema admin and running a shell |
| with administrative privileges: |
| </p><pre class="screen"> |
| $ cd /usr/share/cygwin |
| $ ldifde -i -f cygwin.ldif -k -c "CN=schema,CN=Configuration,DC=X" #schemaNamingContext |
| </pre><p> |
| Afterwards, the auxiliary class <code class="literal">cygwinUser</code> is attached to |
| the class <code class="literal">User</code>, and the auxiliary class |
| <code class="literal">cygwinGroup</code> is attached to the class |
| <code class="literal">Group</code>. The new attributes can be immediately edited |
| using <span class="command"><strong>ADSI Edit</strong></span>. |
| </p><p> |
| At the time of writing the following attributes are utilized by Cygwin: |
| </p><div class="segmentedlist"><table border="0"><thead><tr class="segtitle"></tr></thead><tbody><tr class="seglistitem"><td class="seg"><code class="literal">cygwinHome</code></td><td class="seg">Used as Cygwin home directory with <code class="literal">db_home: cygwin</code>. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-home" title="The db_home setting">the section called “The <code class="literal">db_home</code> setting”</a>.</td></tr><tr class="seglistitem"><td class="seg"><code class="literal">cygwinShell</code></td><td class="seg">Used as Cygwin login shell with <code class="literal">db_shell: cygwin</code>. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-shell" title="The db_shell setting">the section called “The <code class="literal">db_shell</code> setting”</a>.</td></tr><tr class="seglistitem"><td class="seg"><code class="literal">cygwinGecos</code></td><td class="seg">Content will be added to the pw_gecos field with |
| <code class="literal">db_gecos: cygwin</code>. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-gecos" title="The db_gecos setting">the section called “The <code class="literal">db_gecos</code> setting”</a>.</td></tr></tbody></table></div></div><div class="sect4"><div class="titlepage"><div><div><h5 class="title"><a name="ntsec-mapping-nsswitch-posix"></a>The <code class="literal">unix</code> schema</h5></div></div></div><p> |
| The <code class="literal">unix</code> schema utilizes the |
| <code class="literal">posixAccount</code> attribute extension. This is one of two |
| schema extensions which are connected to AD accounts, available by default. |
| They are usually <span class="bold"><strong>not set</strong></span>, unless used by |
| the Active Directory <code class="literal">Server for NIS</code> feature (deprecated |
| since Server 2012 R2). |
|
|
| Two schemata are interesting for Cygwin, <code class="literal">posixAccount</code>, |
| connected to user accounts, and <code class="literal">posixGroup</code>, connected |
| to group accounts. Both follow the description of RFC 2307, |
| <a class="ulink" href="https://tools.ietf.org/html/rfc2307" target="_top">an Approach for Using LDAP as |
| a Network Information Service</a>. |
| The user attributes utilized by Cygwin are: |
| </p><div class="segmentedlist"><table border="0"><thead><tr class="segtitle"></tr></thead><tbody><tr class="seglistitem"><td class="seg"><code class="literal">unixHomeDirectory</code></td><td class="seg">Used as Cygwin home directory with <code class="literal">db_home: unix</code>. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-home" title="The db_home setting">the section called “The <code class="literal">db_home</code> setting”</a>.</td></tr><tr class="seglistitem"><td class="seg"><code class="literal">loginShell</code></td><td class="seg">Used as Cygwin login shell with <code class="literal">db_shell: unix</code>. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-shell" title="The db_shell setting">the section called “The <code class="literal">db_shell</code> setting”</a>.</td></tr><tr class="seglistitem"><td class="seg"><code class="literal">gecos</code></td><td class="seg">Content will be added to the pw_gecos field with |
| <code class="literal">db_gecos: unix</code>. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-gecos" title="The db_gecos setting">the section called “The <code class="literal">db_gecos</code> setting”</a>.</td></tr><tr class="seglistitem"><td class="seg"><code class="literal">uidNumber</code></td><td class="seg">See <a class="xref" href="ntsec.html#ntsec-mapping-nfs" title="NFS account mapping">the section called “NFS account mapping”</a> and |
| <a class="xref" href="ntsec.html#ntsec-mapping-samba" title="Samba account mapping">the section called “Samba account mapping”</a>.</td></tr></tbody></table></div><p> |
| The group attributes utilized by Cygwin are: |
| </p><div class="segmentedlist"><table border="0"><thead><tr class="segtitle"></tr></thead><tbody><tr class="seglistitem"><td class="seg"><code class="literal">gidNumber</code></td><td class="seg">See <a class="xref" href="ntsec.html#ntsec-mapping-nfs" title="NFS account mapping">the section called “NFS account mapping”</a> and |
| <a class="xref" href="ntsec.html#ntsec-mapping-samba" title="Samba account mapping">the section called “Samba account mapping”</a>.</td></tr></tbody></table></div><p> |
| Apart from power shell scripting or inventing new CLI tools, these attributes |
| can be changed using the <code class="literal">Attribute Editor</code> tab in the user |
| properties dialog of the <code class="literal">Active Directory Users and Computers</code> |
| MMC snap-in. Alternatively, if the <code class="literal">Server for NIS</code> |
| administration feature has been installed, there will be a |
| <code class="literal">UNIX Attributes</code> tab which contains the required fields, |
| except for the gecos field. Last resort is <span class="command"><strong>ADSI Edit</strong></span>. |
| </p></div><div class="sect4"><div class="titlepage"><div><div><h5 class="title"><a name="ntsec-mapping-nsswitch-desc"></a>The <code class="literal">desc</code> schema</h5></div></div></div><p> |
| When using user accounts from the local account database, the SAM, there |
| are only a very limited number of settings available. In contrast to |
| Active Directory there's no way to add fields to a user's entry. You have |
| to make do with the fields available. The method to utilize the |
| <code class="literal">description</code> field has been mainly introduced for those |
| accounts, usually the only ones a home user has. However, for symmetry, |
| and because there may be a reason to use this in an AD environment, this |
| schema is also supported for AD users. |
| </p><div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"><h3 class="title">Note</h3><p> |
| The presentation of local user account settings on Windows is confusing, |
| to say the least. The <code class="literal">description</code> field is not visible at |
| all in the user settings available via the <code class="literal">User Accounts</code> |
| control settings. And while it's called <code class="literal">Description</code> in the |
| <code class="literal">Local Users and Groups</code> MMC snap-in (available, for instance, |
| via the <code class="literal">Computer Management</code> GUI), in the command |
| line tool <span class="command"><strong>net user</strong></span> the same field is called |
| <code class="literal">comment</code>. The latter is especially confusing for |
| AD admins, because the <code class="literal">comment</code> attribute in AD is called |
| <code class="literal">usercomment</code> on the command line. Confused? Never mind, |
| you're not the only one... |
| </p></div><p> |
| Fortunately you can utilize the <code class="literal">description</code> field even if |
| you're running a "home edition" of Windows, by using the command line. The |
| <span class="command"><strong>net user</strong></span> command allows to set all values in the SAM, even |
| if the GUI is crippled. |
| </p><p> |
| A Cygwin SAM comment entry looks like this: |
| </p><pre class="screen"> |
| <cygwin key="value" key="value" [...] /> |
| </pre><p> |
| The supported keys are: |
| </p><div class="segmentedlist"><table border="0"><thead><tr class="segtitle"></tr></thead><tbody><tr class="seglistitem"><td class="seg"><code class="literal">home="value"</code></td><td class="seg">Sets the Cygwin home dir to value.</td></tr><tr class="seglistitem"><td class="seg"><code class="literal">shell="value"</code></td><td class="seg">Sets the Cygwin login shell to value.</td></tr><tr class="seglistitem"><td class="seg"><code class="literal">gecos="value"</code></td><td class="seg">Adds the string value to the user's gecos field.</td></tr></tbody></table></div><p> |
| The next two settings are only supported for SAM accounts. |
| </p><div class="segmentedlist"><table border="0"><thead><tr class="segtitle"></tr></thead><tbody><tr class="seglistitem"><td class="seg"><code class="literal">group="value"</code></td><td class="seg">Sets the Cygwin primary group of the account to value, provided that |
| the user <span class="emphasis"><em>is</em></span> already a member of that group. |
| This allows to override the default <code class="literal">None</code> primary |
| group for local accounts. One nice idea here is, for instance, |
| group="Users".</td></tr><tr class="seglistitem"><td class="seg"><code class="literal">unix="value"</code></td><td class="seg">Sets the NFS/Samba uid of the user to the decimal value. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nfs" title="NFS account mapping">the section called “NFS account mapping”</a> and |
| <a class="xref" href="ntsec.html#ntsec-mapping-samba" title="Samba account mapping">the section called “Samba account mapping”</a>.</td></tr></tbody></table></div><p> |
| The <cygwin .../> string can start at any point in the comment, but |
| you have to follow the rules: |
| </p><div class="itemizedlist"><ul class="itemizedlist compact" style="list-style-type: disc; "><li class="listitem"> |
| It starts with "<cygwin " and ends with "/>". |
| </li><li class="listitem"> |
| The "cygwin" string and the key names have to be lowercase. |
| </li><li class="listitem"> |
| No spaces between key and "value", just the equal sign. |
| </li><li class="listitem"> |
| The value must be placed within double quotes and it must not contain a double |
| quote itself. The double quotes are required for the decimal values as well! |
| </li></ul></div><div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"><h3 class="title">Note</h3><p> |
| There's also a length restriction imposed by Windows. The |
| <code class="literal">description</code> entry has a maximum length of 1023 characters. |
| </p></div><p> |
| CMD example: |
| </p><pre class="screen"> |
| net user corinna /comment:"<cygwin home=\"/home/foo\"/>" |
| </pre><p> |
| Bash example (use single quotes): |
| </p><pre class="screen"> |
| net user corinna /comment:'<cygwin home="/home/foo"/>' |
| </pre><p> |
| For changing group comments, use the `net localgroup' command. The supported |
| key/value pair for SAM groups are: |
| </p><div class="segmentedlist"><table border="0"><thead><tr class="segtitle"></tr></thead><tbody><tr class="seglistitem"><td class="seg"><code class="literal">unix="value"</code></td><td class="seg">Sets the NFS/Samba gid of the group to the decimal value. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nfs" title="NFS account mapping">the section called “NFS account mapping”</a> and |
| <a class="xref" href="ntsec.html#ntsec-mapping-samba" title="Samba account mapping">the section called “Samba account mapping”</a>.</td></tr></tbody></table></div></div></div><div class="sect3"><div class="titlepage"><div><div><h4 class="title"><a name="ntsec-mapping-nfs"></a>NFS account mapping</h4></div></div></div><p> |
| Microsoft's NFS client does not map the uid/gid values on the NFS shares |
| to SIDs. There's no such thing as a (fake) security descriptor returned |
| to the application. Rather, via an undocumented API an application can |
| fetch <a class="ulink" href="https://tools.ietf.org/html/rfc1813" target="_top">RFC 1813</a> |
| compatible NFSv3 stat information from the share. This is what Cygwin is |
| using to show stat information for files on NFS shares. |
| </p><p> |
| The problem is, while all other information in this stat record, like |
| timestamps, file size etc., can be used by Cygwin, Cygwin had no way to |
| map the values of the st_uid and st_gid members to a Windows SID for a |
| long time. So it just faked the file owner info and claimed that it's |
| you. |
| </p><p> |
| However, SFU has, over time, developed multiple methods to map UNIX |
| uid/gid values on NFS shares to Windows SIDs. You'll find the full |
| documentation of the mapping methods in |
| <a class="ulink" href="http://blogs.technet.com/b/filecab/archive/2012/10/09/nfs-identity-mapping-in-windows-server-2012.aspx" target="_top">NFS Identity Mapping in Windows Server 2012</a> |
| </p><p> |
| Cygwin now utilizes the |
| <a class="ulink" href="https://tools.ietf.org/html/rfc2307" target="_top">RFC 2307</a> |
| mapping for this purpose. This is most of the time provided by an AD domain, |
| but it could also be a standalone LDAP mapping server. Per |
| <a class="ulink" href="https://tools.ietf.org/html/rfc2307" target="_top">RFC 2307</a>, the uid is |
| in the attribute <code class="literal">uidNumber</code>. For groups, the gid is in the |
| <code class="literal">gidNumber</code> attribute. |
| See <a class="xref" href="ntsec.html#ntsec-mapping-nsswitch-posix" title="The unix schema">the section called “The <code class="literal">unix</code> schema”</a>. |
| </p><p> |
| When Cygwin stat()s files on an NFS share, it asks the mapping server via |
| LDAP in two different ways, depending on the role of the mapping server. |
| </p><div class="itemizedlist"><ul class="itemizedlist compact" style="list-style-type: disc; "><li class="listitem"> |
| If the server is an AD domain controller, it asks for an account with |
| <code class="literal">uidNumber</code> attribute == <code class="literal">st_uid</code> field of |
| the stat record returned by NFS. If an account matches, AD returns the |
| Windows SID, so we have an immediate mapping from UNIX uid to a Windows SID, |
| if the user account has a valid <code class="literal">uidNumber</code> attribute. For |
| groups, the method is the same, just that Cygwin asks for a group with |
| <code class="literal">gidNumber</code> attribute == <code class="literal">st_gid</code> field of the |
| stat record. |
| </li><li class="listitem"> |
| If the server is a standalone LDAP mapping server Cygwin asks for the |
| same <code class="literal">uidNumber</code>/<code class="literal">gidNumber</code> attributes, but |
| it can't expect that the LDAP server knows anything about Windows SIDs. |
| Rather, the mapping server returns the account name. Cygwin then asks the |
| DC for an account with this name, and if that succeeds, we have a mapping |
| between UNIX uid/gid and Windows SIDs. |
| </li></ul></div><p> |
| The mapping will be cached for the lifetime of the process, and inherited |
| by child processes. |
| </p></div><div class="sect3"><div class="titlepage"><div><div><h4 class="title"><a name="ntsec-mapping-samba"></a>Samba account mapping</h4></div></div></div><p> |
| A fully set up Samba file server with domain integration is running winbindd to |
| map Windows SIDs to artificially created UNIX uids and gids, and this mapping is |
| transparent within the domain, so Cygwin doesn't have to do anything special. |
| </p><p> |
| However, setting up winbindd isn't for everybody, and it fails to map |
| Windows accounts to already existing UNIX users or groups. In contrast |
| to NFS, Samba returns security descriptors, but unmapped UNIX accounts |
| get special SIDs: |
| </p><div class="itemizedlist"><ul class="itemizedlist compact" style="list-style-type: disc; "><li class="listitem"> |
| A UNIX user account with uid X is mapped to the Windows SID S-1-22-1-X. |
| </li><li class="listitem"> |
| A UNIX group account with gid X is mapped to SID S-1-22-2-X. |
| </li></ul></div><p> |
| As you can see, even though we have SIDs, they just reflect the actual |
| uid/gid values on the UNIX box in the RID value. It's only marginally |
| different from the NFS method, so why not just use the same method as |
| for NFS? |
| </p><p> |
| That's what Cygwin will do. If it encounters a S-1-22-x-y SID, it |
| will perform the same |
| <a class="ulink" href="https://tools.ietf.org/html/rfc2307" target="_top">RFC 2307</a> |
| mapping as for NFS shares. |
| </p><p> |
| For home users without any Windows domain or LDAP server per |
| <a class="ulink" href="https://tools.ietf.org/html/rfc2307" target="_top">RFC 2307</a>, |
| but with a Linux machine running Samba, just add this information to |
| your SAM account. Assuming the uid of your Linux user account is 505 |
| and the gid of your primary group is, say, 100, just add the values to |
| your SAM user and group accounts. The following example assumes you |
| didn't already add something else to the comment field. |
| </p><p> |
| To your user's SAM comment (remember: called <code class="literal">Description</code> |
| in the GUI), |
| add: |
| </p><pre class="screen"> |
| <cygwin group="Users" unix="505"/> |
| </pre><p> |
| To the <code class="literal">Users</code> group SAM comment add: |
| </p><pre class="screen"> |
| <cygwin unix="100"/> |
| </pre><p> |
| This should be sufficient to work on your Samba share and to see |
| all files owned by your Linux user account as your files. |
| </p></div></div><div class="sect2"><div class="titlepage"><div><div><h3 class="title"><a name="ntsec-files"></a>File permissions</h3></div></div></div><p>On NTFS and if the <code class="literal">noacl</code> mount option is not |
| specified for a mount point, Cygwin sets file permissions as on POSIX |
| systems. Basically this is done by defining a Security Descriptor with the |
| matching owner and group SIDs, and a DACL which contains ACEs for the owner, |
| the group and for "Everyone", which represents what POSIX calls "others".</p><p>There's just one problem when trying to map the POSIX permission model |
| onto the Windows permission model.</p><p>There's a leak in the definition of a "correct" ACL which disallows a |
| certain POSIX permission setting. The official documentation explains in short |
| the following:</p><div class="itemizedlist"><ul class="itemizedlist compact" style="list-style-type: disc; "><li class="listitem"><p>The requested permissions are checked against all |
| ACEs of the user as well as all groups the user is member of. The |
| permissions given in these user and groups access allowed ACEs are |
| accumulated and the resulting set is the set of permissions of that |
| user given for that object.</p></li><li class="listitem"><p>The order of ACEs is important. The system reads them in |
| sequence until either any single requested permission is denied or all |
| requested permissions are granted. Reading stops when this condition is |
| met. Later ACEs are not taken into account.</p></li><li class="listitem"><p>All access denied ACEs <span class="bold"><strong>should</strong></span> precede any access allowed ACE. ACLs |
| following this rule are called "canonical".</p></li></ul></div><p>Note that the last rule is a preference or a definition of |
| correctness. It's not an absolute requirement. All Windows kernels |
| will correctly deal with the ACL regardless of the order of allow and |
| deny ACEs. The second rule is not modified to get the ACEs in the |
| preferred order.</p><p>Unfortunately, the security tab in the file properties dialog of |
| the Windows Explorer will pop up a warning stating "The permissions on |
| ... are incorrectly ordered, which may cause some entries to be |
| ineffective." Pressing the Cancel button of the properties dialog |
| fortunately leaves the sort order unchanged, but pressing OK will cause |
| Explorer to canonicalize the order of the ACEs, thereby invalidating |
| POSIX compatibility.</p><p>Canonical ACLs are unable to reflect each possible combination |
| of POSIX permissions. Example:</p><pre class="screen"> |
| rw-r-xrw- |
| </pre><p>Ok, so here's the first try to create a matching ACL, assuming |
| the Windows permissions only have three bits, as their POSIX counterpart: |
| </p><pre class="screen"> |
| UserAllow: 110 |
| GroupAllow: 101 |
| OthersAllow: 110 |
| </pre><p>Hmm, because of the accumulation of allow rights the user may |
| execute because the group may execute.</p><p>Second try:</p><pre class="screen"> |
| UserDeny: 001 |
| GroupAllow: 101 |
| OthersAllow: 110 |
| </pre><p>Now the user may read and write but not execute. Better? No! |
| Unfortunately the group may write now because others may write.</p><p>Third try:</p><pre class="screen"> |
| UserDeny: 001 |
| GroupDeny: 010 |
| GroupAllow: 001 |
| OthersAllow: 110 |
| </pre><p>Now the group may not write as intended but unfortunately the user may |
| not write anymore, either. How should this problem be solved? According to |
| the canonical order a UserAllow has to follow the GroupDeny but it's |
| easy to see that this can never be solved that way.</p><p>The only chance:</p><pre class="screen"> |
| UserDeny: 001 |
| UserAllow: 010 |
| GroupDeny: 010 |
| GroupAllow: 001 |
| OthersAllow: 110 |
| </pre><p>Again: This works on all supported versions of Windows. Only the GUIs |
| aren't able (or willing) to deal with that order.</p></div><div class="sect2"><div class="titlepage"><div><div><h3 class="title"><a name="ntsec-setuid-overview"></a>Switching the user context</h3></div></div></div><p>Windows users have been accustomed to the "Switch User" feature, which |
| switches the entire desktop to another user while leaving the original user's |
| desktop "suspended". Another Windows feature is the "Run as..." context menu |
| entry, which allows you to start an application using another user account |
| when right-clicking on applications and shortcuts.</p><p>On POSIX systems, this operation can be performed by processes |
| running under the privileged user accounts (usually the "root" user |
| account) on a per-process basis. This is called "switching the user |
| context" for that process, and is performed using the POSIX |
| <span class="command"><strong>setuid</strong></span> and <span class="command"><strong>seteuid</strong></span> system |
| calls.</p><p>While this sort of feature is available on Windows as well, |
| Windows does not support the concept of these calls in a simple fashion. |
| Switching the user context in Windows is generally a tricky process with |
| lots of "behind the scenes" magic involved.</p><p>Windows uses so-called `access tokens' to identify a user and its |
| permissions. Usually the access token is created at logon time and then |
| it's attached to the starting process. Every new process within a session |
| inherits the access token from its parent process. Every thread can |
| get its own access token, which allows, for instance, to define threads |
| with restricted permissions.</p><div class="sect3"><div class="titlepage"><div><div><h4 class="title"><a name="ntsec-logonuser"></a>Switching the user context with password authentication</h4></div></div></div><p>To switch the user context, the process has to request such an access |
| token for the new user. This is typically done by calling the Win32 API |
| function <span class="command"><strong>LogonUser</strong></span> with the user name and the user's |
| cleartext password as arguments. If the user exists and the password was |
| specified correctly, the access token is returned and either used in |
| <span class="command"><strong>ImpersonateLoggedOnUser</strong></span> to change the user context of |
| the current thread, or in <span class="command"><strong>CreateProcessAsUser</strong></span> to |
| change the user context of a spawned child process.</p><p>Later versions of Windows define new functions in this context and |
| there are also functions to manipulate existing access tokens (usually |
| only to restrict them). Windows Vista also adds subtokens which are |
| attached to other access tokens which plays an important role in the UAC |
| (User Access Control) facility of Vista and later. However, none of |
| these extensions to the original concept are important for this |
| documentation.</p><p>Back to this logon with password, how can this be used to |
| implement <span class="command"><strong>set(e)uid</strong></span>? Well, it requires modification |
| of the calling application. Two Cygwin functions have been introduced |
| to support porting <span class="command"><strong>setuid</strong></span> applications which only |
| require login with passwords. You only give Cygwin the right access |
| token and then you can call <span class="command"><strong>seteuid</strong></span> or |
| <span class="command"><strong>setuid</strong></span> as usual in POSIX applications. Porting such |
| a <span class="command"><strong>setuid</strong></span> application is illustrated by a short |
| example:</p><pre class="screen"> |
|
|
| /* First include all needed cygwin stuff. */ |
| #ifdef __CYGWIN__ |
| #include <windows.h> |
| #include <sys/cygwin.h> |
| #endif |
|
|
| [...] |
|
|
| struct passwd *user_pwd_entry = getpwnam (username); |
| char *cleartext_password = getpass ("Password:"); |
|
|
| [...] |
|
|
| #ifdef __CYGWIN__ |
| /* Patch the typical password test. */ |
| { |
| HANDLE token; |
|
|
| /* Try to get the access token from Windows. */ |
| token = cygwin_logon_user (user_pwd_entry, cleartext_password); |
| if (token == INVALID_HANDLE_VALUE) |
| error_exit; |
| /* Inform Cygwin about the new impersonation token. */ |
| cygwin_set_impersonation_token (token); |
| /* Cygwin is now able, to switch to that user context by setuid or seteuid calls. */ |
| } |
| #else |
| /* Use standard method on non-Cygwin systems. */ |
| hashed_password = crypt (cleartext_password, salt); |
| if (!user_pwd_entry || |
| strcmp (hashed_password, user_pwd_entry->pw_passwd)) |
| error_exit; |
| #endif /* CYGWIN */ |
|
|
| [...] |
|
|
| /* Everything else remains the same! */ |
|
|
| setegid (user_pwd_entry->pw_gid); |
| seteuid (user_pwd_entry->pw_uid); |
| execl ("/bin/sh", ...); |
|
|
|
|
| </pre></div><div class="sect3"><div class="titlepage"><div><div><h4 class="title"><a name="ntsec-nopasswd1"></a>Switching the user context without password, Method 1: Kerberos/MsV1_0 S4U authentication</h4></div></div></div><p>An unfortunate aspect of the implementation of |
| <span class="command"><strong>set(e)uid</strong></span> is the fact that the calling process |
| requires the password of the user to switch to. Applications such as |
| <span class="command"><strong>sshd</strong></span> wishing to switch the user context after a |
| successful public key authentication, or the <span class="command"><strong>cron</strong></span> |
| application which, again, wants to switch the user without any authentication |
| are stuck here. But there are other ways to get new user tokens.</p><p>Starting with Cygwin 3.0, Cygwin tries to create a token by using |
| <code class="literal">Windows S4U authentication</code> by default. For a quick |
| description, see |
| <a class="ulink" href="https://blogs.msdn.microsoft.com/winsdk/2015/08/28/logon-as-a-user-without-a-password/" target="_top">this blog posting</a>. Cygwin versions prior |
| to 3.0 tried to creat a user token from scratch using an officially |
| undocumented function <span class="command"><strong>NtCreateToken</strong></span> which |
| is now disabled.</p><p>So we just start the services which have to switch the user context |
| (<span class="command"><strong>cygsshd</strong></span>, <span class="command"><strong>inetd</strong></span>, <span class="command"><strong>cron</strong></span>, |
| ...) as Windows services under the SYSTEM (or LocalSystem in the GUI) |
| account and everything just works. Unfortunately that's too simple. |
| Using <code class="literal">S4U</code> has a drawback.</p><p>Annoyingly, you don't have the usual comfortable access |
| to network shares. The reason is that the token has been created |
| without knowing the password. The password are your credentials |
| necessary for network access. Thus, if you logon with a password, the |
| password is stored hidden as "token credentials" within the access token |
| and used as default logon to access network resources. Since these |
| credentials are missing from the token created with <code class="literal">S4U</code> |
| or <span class="command"><strong>NtCreateToken</strong></span>, you only can access network shares |
| from the new user's process tree by using explicit authentication, on |
| the command line for instance:</p><pre class="screen"> |
| bash$ net use '\\server\share' /user:DOMAIN\my_user my_users_password |
| </pre><p>Note that, on some systems, you can't even define a drive letter |
| to access the share, and under some circumstances the drive letter you |
| choose collides with a drive letter already used in another session. |
| Therefore it's better to get used to accessing these shares using the UNC |
| path as in</p><pre class="screen"> |
| bash$ grep foo //server/share/foofile |
| </pre></div><div class="sect3"><div class="titlepage"><div><div><h4 class="title"><a name="ntsec-nopasswd3"></a>Switching the user context without password, Method 2: With password</h4></div></div></div><p>Not being able to access network shares without having to specify |
| a cleartext password on the command line or in a script is a harsh problem |
| for automated logons for testing purposes and similar stuff.</p><p>Fortunately there is a solution, but it has its own drawbacks. |
| But, first things first, how does it work? The title of this section |
| says it all. Instead of trying to logon without password, we just logon |
| with password. The password gets stored two-way encrypted in a hidden, |
| obfuscated area of the registry, the LSA private registry area. This |
| part of the registry contains, for instance, the passwords of the Windows |
| services which run under some non-default user account.</p><p>So what we do is to utilize this registry area for the purpose of |
| <span class="command"><strong>set(e)uid</strong></span>. The Cygwin command <span class="command"><strong><a class="link" href="passwd.html" title="passwd">passwd</a> -R</strong></span> allows a user to specify |
| his/her password for storage in this registry area. When this user |
| tries to login using ssh with public key authentication, Cygwin's |
| <span class="command"><strong>set(e)uid</strong></span> examines the LSA private registry area and |
| searches for a Cygwin specific key which contains the password. If it |
| finds it, it calls <span class="command"><strong>LogonUser</strong></span> under the hood, using |
| this password. If that works, <span class="command"><strong>LogonUser</strong></span> returns an |
| access token with all credentials necessary for network access.</p><p>For good measure, and since this way to implement |
| <span class="command"><strong>set(e)uid</strong></span> is not only used by Cygwin but also by |
| Microsoft's SFU (Services for Unix), we also look for a key stored by |
| SFU (using the SFU command <span class="command"><strong>regpwd</strong></span>) and use that if it's |
| available.</p><p>We got it. A full access token with its own logon session, with |
| all network credentials. Hmm, that's heaven...</p><p>Back on earth, what about the drawbacks?</p><p>First, adding a password to the LSA private registry area |
| requires administrative access. So calling <span class="command"><strong>passwd -R</strong></span> |
| as a normal user will fail! Cygwin provides a workaround for |
| this. If <span class="command"><strong>cygserver</strong></span> is started as a service running |
| under the SYSTEM account (which is the default way to run |
| <span class="command"><strong>cygserver</strong></span>) you can use <span class="command"><strong>passwd -R</strong></span> |
| as normal, non-privileged user as well.</p><p>Second, as aforementioned, the password is two-way encrypted in a |
| hidden, obfuscated registry area. Only SYSTEM has access to this area |
| for listing purposes, so, even as an administrator, you can't examine |
| this area with regedit. Right? No. Every administrator can start |
| regedit as SYSTEM user, the Internet is your friend here.</p><p>Additionally, if an administrator knows under which name |
| the private key is stored (which is well-known since the algorithms |
| used to create the Cygwin and SFU keys are no secret), every administrator |
| can access the password of all keys stored this way in the registry.</p><p>Conclusion: If your system is used exclusively by you, and if |
| you're also the only administrator of your system, and if your system is |
| adequately locked down to prevent malicious access, you can safely use |
| this method. If your machine is part of a network which has |
| dedicated administrators, and you're not one of these administrators, |
| but you (think you) can trust your administrators, you can probably |
| safely use this method.</p><p>In all other cases, don't use this method. You have been warned.</p></div><div class="sect3"><div class="titlepage"><div><div><h4 class="title"><a name="ntsec-setuid-impl"></a>Switching the user context, how does it all fit together?</h4></div></div></div><p>Now we learned about three different ways to switch the user |
| context using the <span class="command"><strong>set(e)uid</strong></span> system call, but |
| how does <span class="command"><strong>set(e)uid</strong></span> really work? Which method does it |
| use now?</p><p>The answer is, all three of them. So here's a brief overview |
| what <span class="command"><strong>set(e)uid</strong></span> does under the hood:</p><div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem"><p>When <span class="command"><strong>set(e)uid</strong></span> is called, it tests if the |
| user context had been switched by an earlier call already, and if the |
| new user account is the privileged user account under which the process |
| had been started originally. If so, it just switches to the original |
| access token of the process it had been started with.</p></li><li class="listitem"><p> |
| Next, it tests if an access token has been stored by an earlier call |
| to <span class="command"><strong>cygwin_set_impersonation_token</strong></span>. If so, it tests |
| if that token matches the requested user account. If so, the stored |
| token is used for the user context switch.</p><p> |
| If not, there's no predefined token which can just be used for |
| the user context switch, so we have to create a new token. The order |
| is as follows.</p></li><li class="listitem"><p>Check if the user has stored the logon password in the LSA |
| private registry area, either under a Cygwin key, or under a SFU key. |
| If so, use this to call <span class="command"><strong>LogonUser</strong></span>. If this |
| succeeds, we use the resulting token for the user context switch.</p></li><li class="listitem"><p>Otherwise, use the default <code class="literal">S4U</code> authentication |
| to create a token.</p></li><li class="listitem"><p>Older systems, like WOW64 under Windows 7 64 bit, don't support |
| <code class="literal">S4U</code> authentication for local machine accounts. On |
| these systems Cygwin falls back to an old and otherwise deprecated |
| method to create a user token from scratch. The underlying system call |
| is undocumented and has an unfortunate requirement: We have to create a |
| special account with dangerous permissions to perform this action. |
| Therefore this is only enabled on affected systems.</p></li><li class="listitem"><p>If all of the above fails, our process has insufficient privileges |
| to switch the user context at all, so <span class="command"><strong>set(e)uid</strong></span> |
| fails and returns -1, setting errno to EPERM.</p></li></ul></div></div></div></div><div class="navfooter"><hr><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="using-cygwinenv.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="using.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="using-cygserver.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">The <code class="envar">CYGWIN</code> environment |
| variable </td><td width="20%" align="center"><a accesskey="h" href="cygwin-ug-net.html">Home</a></td><td width="40%" align="right" valign="top"> Cygserver</td></tr></table></div></body></html> |
|
|
