1FANOTIFY_INIT(2) Linux Programmer's Manual FANOTIFY_INIT(2)
2
6 fanotify_init - create and initialize fanotify group
7
9 #include <fcntl.h>
10 #include <sys/fanotify.h>
11 int fanotify_init(unsigned int flags, unsigned int event_f_flags);
13
15 For an overview of the fanotify API, see fanotify(7).
16 fanotify_init() initializes a new fanotify group and returns a file
18 descriptor for the event queue associated with the group.
19 The file descriptor is used in calls to fanotify_mark(2) to specify the
21 files, directories, mounts or filesystems for which fanotify events
22 shall be created. These events are received by reading from the file
23 descriptor. Some events are only informative, indicating that a file
24 has been accessed. Other events can be used to determine whether
25 another application is permitted to access a file or directory. Per‐
26 mission to access filesystem objects is granted by writing to the file
27 descriptor.
28 Multiple programs may be using the fanotify interface at the same time
30 to monitor the same files.
31 In the current implementation, the number of fanotify groups per user
33 is limited to 128. This limit cannot be overridden.
34 Calling fanotify_init() requires the CAP_SYS_ADMIN capability. This
36 constraint might be relaxed in future versions of the API. Therefore,
37 certain additional capability checks have been implemented as indicated
38 below.
39 The flags argument contains a multi-bit field defining the notification
41 class of the listening application and further single bit fields speci‐
42 fying the behavior of the file descriptor.
43 If multiple listeners for permission events exist, the notification
45 class is used to establish the sequence in which the listeners receive
46 the events.
47 Only one of the following notification classes may be specified in
49 flags:
50 FAN_CLASS_PRE_CONTENT
52 This value allows the receipt of events notifying that a file
53 has been accessed and events for permission decisions if a file
54 may be accessed. It is intended for event listeners that need
55 to access files before they contain their final data. This
56 notification class might be used by hierarchical storage man‐
57 agers, for example.
58 FAN_CLASS_CONTENT
60 This value allows the receipt of events notifying that a file
61 has been accessed and events for permission decisions if a file
62 may be accessed. It is intended for event listeners that need
63 to access files when they already contain their final content.
64 This notification class might be used by malware detection pro‐
65 grams, for example.
66 FAN_CLASS_NOTIF
68 This is the default value. It does not need to be specified.
69 This value only allows the receipt of events notifying that a
70 file has been accessed. Permission decisions before the file is
71 accessed are not possible.
72 Listeners with different notification classes will receive events in
74 the order FAN_CLASS_PRE_CONTENT, FAN_CLASS_CONTENT, FAN_CLASS_NOTIF.
75 The order of notification for listeners in the same notification class
76 is undefined.
77 The following bits can additionally be set in flags:
79 FAN_CLOEXEC
81 Set the close-on-exec flag (FD_CLOEXEC) on the new file descrip‐
82 tor. See the description of the O_CLOEXEC flag in open(2).
83 FAN_NONBLOCK
85 Enable the nonblocking flag (O_NONBLOCK) for the file descrip‐
86 tor. Reading from the file descriptor will not block. Instead,
87 if no data is available, read(2) fails with the error EAGAIN.
88 FAN_UNLIMITED_QUEUE
90 Remove the limit of 16384 events for the event queue. Use of
91 this flag requires the CAP_SYS_ADMIN capability.
92 FAN_UNLIMITED_MARKS
94 Remove the limit of 8192 marks. Use of this flag requires the
95 CAP_SYS_ADMIN capability.
96 FAN_REPORT_TID (since Linux 4.20)
98 Report thread ID (TID) instead of process ID (PID) in the pid
99 field of the struct fanotify_event_metadata supplied to read(2)
100 (see fanotify(7)).
101 FAN_REPORT_FID (since Linux 5.1)
103 This value allows the receipt of events which contain additional
104 information about the underlying filesystem object correlated to
105 an event. An additional structure encapsulates the information
106 about the object and is included alongside the generic event
107 metadata structure. The file descriptor that is used to repre‐
108 sent the object correlated to an event is instead substituted
109 with a file handle. It is intended for applications that may
110 find the use of a file handle to identify an object more suit‐
111 able than a file descriptor. Additionally, it may be used for
112 applications that are interested in directory entry events, such
113 as FAN_CREATE, FAN_ATTRIB, FAN_MOVE, and FAN_DELETE for example.
114 Note that the use of directory modification events are not sup‐
115 ported when monitoring a mount point. The use of FAN_CLASS_CON‐
116 TENT or FAN_CLASS_PRE_CONTENT is not permitted with this flag
117 and will result in the error EINVAL. See fanotify(7) for addi‐
118 tional information.
119 The event_f_flags argument defines the file status flags that will be
121 set on the open file descriptions that are created for fanotify events.
122 For details of these flags, see the description of the flags values in
123 open(2). event_f_flags includes a multi-bit field for the access mode.
124 This field can take the following values:
125 O_RDONLY
127 This value allows only read access.
128 O_WRONLY
130 This value allows only write access.
131 O_RDWR This value allows read and write access.
133 Additional bits can be set in event_f_flags. The most useful values
135 are:
136 O_LARGEFILE
138 Enable support for files exceeding 2 GB. Failing to set this
139 flag will result in an EOVERFLOW error when trying to open a
140 large file which is monitored by an fanotify group on a 32-bit
141 system.
142 O_CLOEXEC (since Linux 3.18)
144 Enable the close-on-exec flag for the file descriptor. See the
145 description of the O_CLOEXEC flag in open(2) for reasons why
146 this may be useful.
147 The following are also allowable: O_APPEND, O_DSYNC, O_NOATIME, O_NON‐
149 BLOCK, and O_SYNC. Specifying any other flag in event_f_flags yields
150 the error EINVAL (but see BUGS).
151
153 On success, fanotify_init() returns a new file descriptor. On error,
154 -1 is returned, and errno is set to indicate the error.
155
157 EINVAL An invalid value was passed in flags or event_f_flags.
158 FAN_ALL_INIT_FLAGS (deprecated since Linux kernel version 4.20)
159 defines all allowable bits for flags.
160 EMFILE The number of fanotify groups for this user exceeds 128.
162 EMFILE The per-process limit on the number of open file descriptors has
164 been reached.
165 ENOMEM The allocation of memory for the notification group failed.
167 ENOSYS This kernel does not implement fanotify_init(). The fanotify
169 API is available only if the kernel was configured with CON‐
170 FIG_FANOTIFY.
171 EPERM The operation is not permitted because the caller lacks the
173 CAP_SYS_ADMIN capability.
174
176 fanotify_init() was introduced in version 2.6.36 of the Linux kernel
177 and enabled in version 2.6.37.
178
180 This system call is Linux-specific.
181
183 The following bug was present in Linux kernels before version 3.18:
184 * The O_CLOEXEC is ignored when passed in event_f_flags.
186 The following bug was present in Linux kernels before version 3.14:
188 * The event_f_flags argument is not checked for invalid flags. Flags
190 that are intended only for internal use, such as FMODE_EXEC, can be
191 set, and will consequently be set for the file descriptors returned
192 when reading from the fanotify file descriptor.
193
195 fanotify_mark(2), fanotify(7)
196
198 This page is part of release 5.07 of the Linux man-pages project. A
199 description of the project, information about reporting bugs, and the
200 latest version of this page, can be found at
201 https://www.kernel.org/doc/man-pages/.
202Linux 2020-06-09 FANOTIFY_INIT(2)