1LSOF(8) System Manager's Manual LSOF(8)
2
6 lsof - list open files
7
9 lsof [ -?abChlnNOPRtUvVX ] [ -A A ] [ -c c ] [ +c c ] [ +|-d d ] [ +|-D
10 D ] [ +|-f [cfgGn] ] [ -F [f] ] [ -g [s] ] [ -i [i] ] [ -k k ] [ +|-L
11 [l] ] [ +|-m m ] [ +|-M ] [ -o [o] ] [ -p s ] [ +|-r [t[m<fmt>]] ] [ -s
12 [p:s] ] [ -S [t] ] [ -T [t] ] [ -u s ] [ +|-w ] [ -x [fl] ] [ -z [z] ]
13 [ -Z [Z] ] [ -- ] [names]
14
16 Lsof revision 4.83 lists on its standard output file information about
17 files opened by processes for the following UNIX dialects:
18 AIX 5.3
20 Apple Darwin 9 (Mac OS X 10.5)
21 FreeBSD 4.9 for x86-based systems
22 FreeBSD 7.[012], 8.0 and 9.0 for AMD64-based systems
23 Linux 2.1.72 and above for x86-based systems
24 Solaris 9, 10 and 11
25 (See the DISTRIBUTION section of this manual page for information on
27 how to obtain the latest lsof revision.)
28 An open file may be a regular file, a directory, a block special file,
30 a character special file, an executing text reference, a library, a
31 stream or a network file (Internet socket, NFS file or UNIX domain
32 socket.) A specific file or all the files in a file system may be
33 selected by path.
34 Instead of a formatted display, lsof will produce output that can be
36 parsed by other programs. See the -F, option description, and the OUT‐
37 PUT FOR OTHER PROGRAMS section for more information.
38 In addition to producing a single output list, lsof will run in repeat
40 mode. In repeat mode it will produce output, delay, then repeat the
41 output operation until stopped with an interrupt or quit signal. See
42 the +|-r [t[m<fmt>]] option description for more information.
43
45 In the absence of any options, lsof lists all open files belonging to
46 all active processes.
47 If any list request option is specified, other list requests must be
49 specifically requested - e.g., if -U is specified for the listing of
50 UNIX socket files, NFS files won't be listed unless -N is also speci‐
51 fied; or if a user list is specified with the -u option, UNIX domain
52 socket files, belonging to users not in the list, won't be listed
53 unless the -U option is also specified.
54 Normally list options that are specifically stated are ORed - i.e.,
56 specifying the -i option without an address and the -ufoo option pro‐
57 duces a listing of all network files OR files belonging to processes
58 owned by user ``foo''. The exceptions are:
59 1) the `^' (negated) login name or user ID (UID), specified with the -u
61 option;
62 2) the `^' (negated) process ID (PID), specified with the -p option;
64 3) the `^' (negated) process group ID (PGID), specified with the -g
66 option;
67 4) the `^' (negated) command, specified with the -c option;
69 5) the (`^') negated TCP or UDP protocol state names, specified with
71 the -s [p:s] option.
72 Since they represent exclusions, they are applied without ORing or AND‐
74 ing and take effect before any other selection criteria are applied.
75 The -a option may be used to AND the selections. For example, specify‐
77 ing -a, -U, and -ufoo produces a listing of only UNIX socket files that
78 belong to processes owned by user ``foo''.
79 Caution: the -a option causes all list selection options to be ANDed;
81 it can't be used to cause ANDing of selected pairs of selection options
82 by placing it between them, even though its placement there is accept‐
83 able. Wherever -a is placed, it causes the ANDing of all selection
84 options.
85 Items of the same selection set - command names, file descriptors, net‐
87 work addresses, process identifiers, user identifiers, zone names,
88 security contexts - are joined in a single ORed set and applied before
89 the result participates in ANDing. Thus, for example, specifying
90 -i@aaa.bbb, -i@ccc.ddd, -a, and -ufff,ggg will select the listing of
91 files that belong to either login ``fff'' OR ``ggg'' AND have network
92 connections to either host aaa.bbb OR ccc.ddd.
93 Options may be grouped together following a single prefix -- e.g., the
95 option set ``-a -b -C'' may be stated as -abC. However, since values
96 are optional following +|-f, -F, -g, -i, +|-L, -o, +|-r, -s, -S, -T, -x
97 and -z. when you have no values for them be careful that the following
98 character isn't ambiguous. For example, -Fn might represent the -F and
99 -n options, or it might represent the n field identifier character fol‐
100 lowing the -F option. When ambiguity is possible, start a new option
101 with a `-' character - e.g., ``-F -n''. If the next option is a file
102 name, follow the possibly ambiguous option with ``--'' - e.g., ``-F --
103 name''.
104 Either the `+' or the `-' prefix may be applied to a group of options.
106 Options that don't take on separate meanings for each prefix - e.g., -i
107 - may be grouped under either prefix. Thus, for example, ``+M -i'' may
108 be stated as ``+Mi'' and the group means the same as the separate
109 options. Be careful of prefix grouping when one or more options in the
110 group does take on separate meanings under different prefixes - e.g.,
111 +|-M; ``-iM'' is not the same request as ``-i +M''. When in doubt, use
112 separate options with appropriate prefixes.
113 -? -h These two equivalent options select a usage (help) output
115 list. Lsof displays a shortened form of this output when it
116 detects an error in the options supplied to it, after it has
117 displayed messages explaining each error. (Escape the `?'
118 character as your shell requires.)
119 -a This option causes list selection options to be ANDed, as
121 described above.
122 -A A This option is available on systems configured for AFS whose
124 AFS kernel code is implemented via dynamic modules. It allows
125 the lsof user to specify A as an alternate name list file
126 where the kernel addresses of the dynamic modules might be
127 found. See the lsof FAQ (The FAQ section gives its location.)
128 for more information about dynamic modules, their symbols, and
129 how they affect lsof.
130 -b This option causes lsof to avoid kernel functions that might
132 block - lstat(2), readlink(2), and stat(2).
133 See the BLOCKS AND TIMEOUTS and AVOIDING KERNEL BLOCKS sec‐
135 tions for information on using this option.
136 -c c This option selects the listing of files for processes execut‐
138 ing the command that begins with the characters of c. Multi‐
139 ple commands may be specified, using multiple -c options.
140 They are joined in a single ORed set before participating in
141 AND option selection.
142 If c begins with a `^', then the following characters specify
144 a command name whose processes are to be ignored (excluded.)
145 If c begins and ends with a slash ('/'), the characters
147 between the slashes are interpreted as a regular expression.
148 Shell meta-characters in the regular expression must be quoted
149 to prevent their interpretation by the shell. The closing
150 slash may be followed by these modifiers:
151 b the regular expression is a basic one.
153 i ignore the case of letters.
154 x the regular expression is an extended one
155 (default).
156 See the lsof FAQ (The FAQ section gives its location.) for
158 more information on basic and extended regular expressions.
159 The simple command specification is tested first. If that
161 test fails, the command regular expression is applied. If the
162 simple command test succeeds, the command regular expression
163 test isn't made. This may result in ``no command found for
164 regex:'' messages when lsof's -V option is specified.
165 +c w This option defines the maximum number of initial characters
167 of the name, supplied by the UNIX dialect, of the UNIX command
168 associated with a process to be printed in the COMMAND column.
169 (The lsof default is nine.)
170 Note that many UNIX dialects do not supply all command name
172 characters to lsof in the files and structures from which lsof
173 obtains command name. Often dialects limit the number of
174 characters supplied in those sources. For example, Linux
175 2.4.27 and Solaris 9 both limit command name length to 16
176 characters.
177 If w is zero ('0'), all command characters supplied to lsof by
179 the UNIX dialect will be printed.
180 If w is less than the length of the column title, ``COMMAND'',
182 it will be raised to that length.
183 -C This option disables the reporting of any path name components
185 from the kernel's name cache. See the KERNEL NAME CACHE sec‐
186 tion for more information.
187 +d s This option causes lsof to search for all open instances of
189 directory s and the files and directories it contains at its
190 top level. This option does NOT descend the directory tree,
191 rooted at s. The +D D option may be used to request a
192 full-descent directory tree search, rooted at directory D.
193 Processing of the +d option does not follow symbolic links
195 within s unless the -x or -x l option is also specified. Nor
196 does it search for open files on file system mount points on
197 subdirectories of s unless the -x or -x f option is also
198 specified.
199 Note: the authority of the user of this option limits it to
201 searching for files that the user has permission to examine
202 with the system stat(2) function.
203 -d s This option specifies a list of file descriptors (FDs) to
205 exclude from or include in the output listing. The file
206 descriptors are specified in the comma-separated set s - e.g.,
207 ``cwd,1,3'', ``^6,^2''. (There should be no spaces in the
208 set.)
209 The list is an exclusion list if all entries of the set begin
211 with `^'. It is an inclusion list if no entry begins with
212 `^'. Mixed lists are not permitted.
213 A file descriptor number range may be in the set as long as
215 neither member is empty, both members are numbers, and the
216 ending member is larger than the starting one - e.g., ``0-7''
217 or ``3-10''. Ranges may be specified for exclusion if they
218 have the `^' prefix - e.g., ``^0-7'' excludes all file
219 descriptors 0 through 7.
220 Multiple file descriptor numbers are joined in a single ORed
222 set before participating in AND option selection.
223 When there are exclusion and inclusion members in the set,
225 lsof reports them as errors and exits with a non-zero return
226 code.
227 See the description of File Descriptor (FD) output values in
229 the OUTPUT section for more information on file descriptor
230 names.
231 +D D This option causes lsof to search for all open instances of
233 directory D and all the files and directories it contains to
234 its complete depth.
235 Processing of the +D option does not follow symbolic links
237 within D unless the -x or -x l option is also specified. Nor
238 does it search for open files on file system mount points on
239 subdirectories of D unless the -x or -x f option is also
240 specified.
241 Note: the authority of the user of this option limits it to
243 searching for files that the user has permission to examine
244 with the system stat(2) function.
245 Further note: lsof may process this option slowly and require
247 a large amount of dynamic memory to do it. This is because it
248 must descend the entire directory tree, rooted at D, calling
249 stat(2) for each file and directory, building a list of all
250 the files it finds, and searching that list for a match with
251 every open file. When directory D is large, these steps can
252 take a long time, so use this option prudently.
253 -D D This option directs lsof's use of the device cache file. The
255 use of this option is sometimes restricted. See the DEVICE
256 CACHE FILE section and the sections that follow it for more
257 information on this option.
258 -D must be followed by a function letter; the function letter
260 may optionally be followed by a path name. Lsof recognizes
261 these function letters:
262 ? - report device cache file paths
264 b - build the device cache file
265 i - ignore the device cache file
266 r - read the device cache file
267 u - read and update the device cache file
268 The b, r, and u functions, accompanied by a path name, are
270 sometimes restricted. When these functions are restricted,
271 they will not appear in the description of the -D option that
272 accompanies -h or -? option output. See the DEVICE CACHE
273 FILE section and the sections that follow it for more informa‐
274 tion on these functions and when they're restricted.
275 The ? function reports the read-only and write paths that
277 lsof can use for the device cache file, the names of any envi‐
278 ronment variables whose values lsof will examine when forming
279 the device cache file path, and the format for the personal
280 device cache file path. (Escape the `?' character as your
281 shell requires.)
282 When available, the b, r, and u functions may be followed by
284 the device cache file's path. The standard default is
285 .lsof_hostname in the home directory of the real user ID that
286 executes lsof, but this could have been changed when lsof was
287 configured and compiled. (The output of the -h and -?
288 options show the current default prefix - e.g., ``.lsof''.)
289 The suffix, hostname, is the first component of the host's
290 name returned by gethostname(2).
291 When available, the b function directs lsof to build a new
293 device cache file at the default or specified path.
294 The i function directs lsof to ignore the default device cache
296 file and obtain its information about devices via direct calls
297 to the kernel.
298 The r function directs lsof to read the device cache at the
300 default or specified path, but prevents it from creating a new
301 device cache file when none exists or the existing one is
302 improperly structured. The r function, when specified without
303 a path name, prevents lsof from updating an incorrect or out‐
304 dated device cache file, or creating a new one in its place.
305 The r function is always available when it is specified with‐
306 out a path name argument; it may be restricted by the permis‐
307 sions of the lsof process.
308 When available, the u function directs lsof to read the device
310 cache file at the default or specified path, if possible, and
311 to rebuild it, if necessary. This is the default device cache
312 file function when no -D option has been specified.
313 +|-f [cfgGn]
315 f by itself clarifies how path name arguments are to be inter‐
316 preted. When followed by c, f, g, G, or n in any combination
317 it specifies that the listing of kernel file structure infor‐
318 mation is to be enabled (`+') or inhibited (`-').
319 Normally a path name argument is taken to be a file system
321 name if it matches a mounted-on directory name reported by
322 mount(8), or if it represents a block device, named in the
323 mount output and associated with a mounted directory name.
324 When +f is specified, all path name arguments will be taken to
325 be file system names, and lsof will complain if any are not.
326 This can be useful, for example, when the file system name
327 (mounted-on device) isn't a block device. This happens for
328 some CD-ROM file systems.
329 When -f is specified by itself, all path name arguments will
331 be taken to be simple files. Thus, for example, the ``-f --
332 /'' arguments direct lsof to search for open files with a `/'
333 path name, not all open files in the `/' (root) file system.
334 Be careful to make sure +f and -f are properly terminated and
336 aren't followed by a character (e.g., of the file or file sys‐
337 tem name) that might be taken as a parameter. For example,
338 use ``--'' after +f and -f as in these examples.
339 $ lsof +f -- /file/system/name
341 $ lsof -f -- /file/name
342 The listing of information from kernel file structures,
344 requested with the +f [cfgGn] option form, is normally inhib‐
345 ited, and is not available in whole or part for some dialects
346 - e.g., /proc-based Linux kernels below 2.6.22. When the pre‐
347 fix to f is a plus sign (`+'), these characters request file
348 structure information:
349 c file structure use count (not Linux)
351 f file structure address (not Linux)
352 g file flag abbreviations (Linux 2.6.22 and up)
353 G file flags in hexadecimal (Linux 2.6.22 and up)
354 n file structure node address (not Linux)
355 When the prefix is minus (`-') the same characters disable the
357 listing of the indicated values.
358 File structure addresses, use counts, flags, and node
360 addresses may be used to detect more readily identical files
361 inherited by child processes and identical files in use by
362 different processes. Lsof column output can be sorted by out‐
363 put columns holding the values and listed to identify identi‐
364 cal file use, or lsof field output can be parsed by an AWK or
365 Perl post-filter script, or by a C program.
366 -F f This option specifies a character list, f, that selects the
368 fields to be output for processing by another program, and the
369 character that terminates each output field. Each field to be
370 output is specified with a single character in f. The field
371 terminator defaults to NL, but may be changed to NUL (000).
372 See the OUTPUT FOR OTHER PROGRAMS section for a description of
373 the field identification characters and the field output
374 process.
375 When the field selection character list is empty, all standard
377 fields are selected (except the raw device field, security
378 context and zone field for compatibility reasons) and the NL
379 field terminator is used.
380 When the field selection character list contains only a zero
382 (`0'), all fields are selected (except the raw device field
383 for compatibility reasons) and the NUL terminator character is
384 used.
385 Other combinations of fields and their associated field termi‐
387 nator character must be set with explicit entries in f, as
388 described in the OUTPUT FOR OTHER PROGRAMS section.
389 When a field selection character identifies an item lsof does
391 not normally list - e.g., PPID, selected with -R - specifica‐
392 tion of the field character - e.g., ``-FR'' - also selects the
393 listing of the item.
394 When the field selection character list contains the single
396 character `?', lsof will display a help list of the field
397 identification characters. (Escape the `?' character as your
398 shell requires.)
399 -g [s] This option excludes or selects the listing of files for the
401 processes whose optional process group IDentification (PGID)
402 numbers are in the comma-separated set s - e.g., ``123'' or
403 ``123,^456''. (There should be no spaces in the set.)
404 PGID numbers that begin with `^' (negation) represent exclu‐
406 sions.
407 Multiple PGID numbers are joined in a single ORed set before
409 participating in AND option selection. However, PGID exclu‐
410 sions are applied without ORing or ANDing and take effect
411 before other selection criteria are applied.
412 The -g option also enables the output display of PGID numbers.
414 When specified without a PGID set that's all it does.
415 -i [i] This option selects the listing of files any of whose Internet
417 address matches the address specified in i. If no address is
418 specified, this option selects the listing of all Internet and
419 x.25 (HP-UX) network files.
420 If -i4 or -i6 is specified with no following address, only
422 files of the indicated IP version, IPv4 or IPv6, are dis‐
423 played. (An IPv6 specification may be used only if the
424 dialects supports IPv6, as indicated by ``[46]'' and
425 ``IPv[46]'' in lsof's -h or -? output.) Sequentially speci‐
426 fying -i4, followed by -i6 is the same as specifying -i, and
427 vice-versa. Specifying -i4, or -i6 after -i is the same as
428 specifying -i4 or -i6 by itself.
429 Multiple addresses (up to a limit of 100) may be specified
431 with multiple -i options. (A port number or service name
432 range is counted as one address.) They are joined in a single
433 ORed set before participating in AND option selection.
434 An Internet address is specified in the form (Items in square
436 brackets are optional.):
437 [46][protocol][@hostname|hostaddr][:service|port]
439 where:
441 46 specifies the IP version, IPv4 or IPv6
442 that applies to the following address.
443 '6' may be be specified only if the UNIX
444 dialect supports IPv6. If neither '4' nor
445 '6' is specified, the following address
446 applies to all IP versions.
447 protocol is a protocol name - TCP, UDP
448 hostname is an Internet host name. Unless a
449 specific IP version is specified, open
450 network files associated with host names
451 of all versions will be selected.
452 hostaddr is a numeric Internet IPv4 address in
453 dot form; or an IPv6 numeric address in
454 colon form, enclosed in brackets, if the
455 UNIX dialect supports IPv6. When an IP
456 version is selected, only its numeric
457 addresses may be specified.
458 service is an /etc/services name - e.g., smtp -
459 or a list of them.
460 port is a port number, or a list of them.
461 IPv6 options may be used only if the UNIX dialect supports
463 IPv6. To see if the dialect supports IPv6, run lsof and spec‐
464 ify the -h or -? (help) option. If the displayed description
465 of the -i option contains ``[46]'' and ``IPv[46]'', IPv6 is
466 supported.
467 IPv4 host names and addresses may not be specified if network
469 file selection is limited to IPv6 with -i 6. IPv6 host names
470 and addresses may not be specified if network file selection
471 is limited to IPv4 with -i 4. When an open IPv4 network
472 file's address is mapped in an IPv6 address, the open file's
473 type will be IPv6, not IPv4, and its display will be selected
474 by '6', not '4'.
475 At least one address component - 4, 6, protocol, ,IR hostname
477 , hostaddr, or service - must be supplied. The `@' character,
478 leading the host specification, is always required; as is the
479 `:', leading the port specification. Specify either hostname
480 or hostaddr. Specify either service name list or port number
481 list. If a service name list is specified, the protocol may
482 also need to be specified if the TCP, UDP and UDPLITE port
483 numbers for the service name are different. Use any case -
484 lower or upper - for protocol.
485 Service names and port numbers may be combined in a list whose
487 entries are separated by commas and whose numeric range
488 entries are separated by minus signs. There may be no embed‐
489 ded spaces, and all service names must belong to the specified
490 protocol. Since service names may contain embedded minus
491 signs, the starting entry of a range can't be a service name;
492 it can be a port number, however.
493 Here are some sample addresses:
495 -i6 - IPv6 only
497 TCP:25 - TCP and port 25
498 @1.2.3.4 - Internet IPv4 host address 1.2.3.4
499 @[3ffe:1ebc::1]:1234 - Internet IPv6 host address
500 3ffe:1ebc::1, port 1234
501 UDP:who - UDP who service port
502 TCP@lsof.itap:513 - TCP, port 513 and host name lsof.itap
503 tcp@foo:1-10,smtp,99 - TCP, ports 1 through 10,
504 service name smtp, port 99, host name foo
505 tcp@bar:1-smtp - TCP, ports 1 through smtp, host bar
506 :time - either TCP, UDP or UDPLITE time service port
507 -k k This option specifies a kernel name list file, k, in place of
509 /vmunix, /mach, etc. This option is not available under AIX
510 on the IBM RISC/System 6000.
511 -l This option inhibits the conversion of user ID numbers to
513 login names. It is also useful when login name lookup is
514 working improperly or slowly.
515 +|-L [l] This option enables (`+') or disables (`-') the listing of
517 file link counts, where they are available - e.g., they aren't
518 available for sockets, or most FIFOs and pipes.
519 When +L is specified without a following number, all link
521 counts will be listed. When -L is specified (the default), no
522 link counts will be listed.
523 When +L is followed by a number, only files having a link
525 count less than that number will be listed. (No number may
526 follow -L.) A specification of the form ``+L1'' will select
527 open files that have been unlinked. A specification of the
528 form ``+aL1 <file_system>'' will select unlinked open files on
529 the specified file system.
530 For other link count comparisons, use field output (-F) and a
532 post-processing script or program.
533 +|-m m This option specifies an alternate kernel memory file or acti‐
535 vates mount table supplement processing.
536 The option form -m m specifies a kernel memory file, m, in
538 place of /dev/kmem or /dev/mem - e.g., a crash dump file.
539 The option form +m requests that a mount supplement file be
541 written to the standard output file. All other options are
542 silently ignored.
543 There will be a line in the mount supplement file for each
545 mounted file system, containing the mounted file system direc‐
546 tory, followed by a single space, followed by the device num‐
547 ber in hexadecimal "0x" format - e.g.,
548 / 0x801
550 Lsof can use the mount supplement file to get device numbers
552 for file systems when it can't get them via stat(2) or
553 lstat(2).
554 The option form +m m identifies m as a mount supplement file.
556 Note: the +m and +m m options are not available for all sup‐
558 ported dialects. Check the output of lsof's -h or -? options
559 to see if the +m and +m m options are available.
560 +|-M Enables (+) or disables (-) the reporting of portmapper regis‐
562 trations for local TCP, UDP and UDPLITE ports. The default
563 reporting mode is set by the lsof builder with the HASPMAPEN‐
564 ABLED #define in the dialect's machine.h header file; lsof is
565 distributed with the HASPMAPENABLED #define deactivated, so
566 portmapper reporting is disabled by default and must be
567 requested with +M. Specifying lsof's -h or -? option will
568 report the default mode. Disabling portmapper registration
569 when it is already disabled or enabling it when already
570 enabled is acceptable.
571 When portmapper registration reporting is enabled, lsof dis‐
573 plays the portmapper registration (if any) for local TCP, UDP
574 or UDPLITE ports in square brackets immediately following the
575 port numbers or service names - e.g., ``:1234[name]'' or
576 ``:name[100083]''. The registration information may be a name
577 or number, depending on what the registering program supplied
578 to the portmapper when it registered the port.
579 When portmapper registration reporting is enabled, lsof may
581 run a little more slowly or even become blocked when access to
582 the portmapper becomes congested or stopped. Reverse the
583 reporting mode to determine if portmapper registration report‐
584 ing is slowing or blocking lsof.
585 For purposes of portmapper registration reporting lsof consid‐
587 ers a TCP, UDP or UDPLITE port local if: it is found in the
588 local part of its containing kernel structure; or if it is
589 located in the foreign part of its containing kernel structure
590 and the local and foreign Internet addresses are the same; or
591 if it is located in the foreign part of its containing kernel
592 structure and the foreign Internet address is INADDR_LOOPBACK
593 (127.0.0.1). This rule may make lsof ignore some foreign
594 ports on machines with multiple interfaces when the foreign
595 Internet address is on a different interface from the local
596 one.
597 See the lsof FAQ (The FAQ section gives its location.) for
599 further discussion of portmapper registration reporting
600 issues.
601 -n This option inhibits the conversion of network numbers to host
603 names for network files. Inhibiting conversion may make lsof
604 run faster. It is also useful when host name lookup is not
605 working properly.
606 -N This option selects the listing of NFS files.
608 -o This option directs lsof to display file offset at all times.
610 It causes the SIZE/OFF output column title to be changed to
611 OFFSET. Note: on some UNIX dialects lsof can't obtain accu‐
612 rate or consistent file offset information from its kernel
613 data sources, sometimes just for particular kinds of files
614 (e.g., socket files.) Consult the lsof FAQ (The FAQ section
615 gives its location.) for more information.
616 The -o and -s options are mutually exclusive; they can't both
618 be specified. When neither is specified, lsof displays what‐
619 ever value - size or offset - is appropriate and available for
620 the type of the file.
621 -o o This option defines the number of decimal digits (o) to be
623 printed after the ``0t'' for a file offset before the form is
624 switched to ``0x...''. An o value of zero (unlimited) directs
625 lsof to use the ``0t'' form for all offset output.
626 This option does NOT direct lsof to display offset at all
628 times; specify -o (without a trailing number) to do that.
629 This option only specifies the number of digits after ``0t''
630 in either mixed size and offset or offset-only output. Thus,
631 for example, to direct lsof to display offset at all times
632 with a decimal digit count of 10, use:
633 -o -o 10
635 or
636 -oo10
637 The default number of digits allowed after ``0t'' is normally
639 8, but may have been changed by the lsof builder. Consult the
640 description of the -o o option in the output of the -h or -?
641 option to determine the default that is in effect.
642 -O This option directs lsof to bypass the strategy it uses to
644 avoid being blocked by some kernel operations - i.e., doing
645 them in forked child processes. See the BLOCKS AND TIMEOUTS
646 and AVOIDING KERNEL BLOCKS sections for more information on
647 kernel operations that may block lsof.
648 While use of this option will reduce lsof startup overhead, it
650 may also cause lsof to hang when the kernel doesn't respond to
651 a function. Use this option cautiously.
652 -p s This option excludes or selects the listing of files for the
654 processes whose optional process IDentification (PID) numbers
655 are in the comma-separated set s - e.g., ``123'' or
656 ``123,^456''. (There should be no spaces in the set.)
657 PID numbers that begin with `^' (negation) represent exclu‐
659 sions.
660 Multiple process ID numbers are joined in a single ORed set
662 before participating in AND option selection. However, PID
663 exclusions are applied without ORing or ANDing and take effect
664 before other selection criteria are applied.
665 -P This option inhibits the conversion of port numbers to port
667 names for network files. Inhibiting the conversion may make
668 lsof run a little faster. It is also useful when port name
669 lookup is not working properly.
670 +|-r [t[m<fmt>]]
672 This option puts lsof in repeat mode. There lsof lists open
673 files as selected by other options, delays t seconds (default
674 fifteen), then repeats the listing, delaying and listing
675 repetitively until stopped by a condition defined by the pre‐
676 fix to the option.
677 If the prefix is a `-', repeat mode is endless. Lsof must be
679 terminated with an interrupt or quit signal.
680 If the prefix is `+', repeat mode will end the first cycle no
682 open files are listed - and of course when lsof is stopped
683 with an interrupt or quit signal. When repeat mode ends
684 because no files are listed, the process exit code will be
685 zero if any open files were ever listed; one, if none were
686 ever listed.
687 Lsof marks the end of each listing: if field output is in
689 progress (the -F, option has been specified), the default
690 marker is `m'; otherwise the default marker is ``========''.
691 The marker is followed by a NL character.
692 The optional "m<fmt>" argument specifies a format for the
694 marker line. The <fmt> characters following `m' are inter‐
695 preted as a format specification to the strftime(3) function,
696 when both it and the localtime(3) function are available in
697 the dialect's C library. Consult the strftime(3) documenta‐
698 tion for what may appear in its format specification. Note
699 that when field output is requested with the -F option, <fmt>
700 cannot contain the NL format, ``%n''. Note also that when
701 <fmt> contains spaces or other characters that affect the
702 shell's interpretation of arguments, <fmt> must be quoted
703 appropriately.
704 Repeat mode reduces lsof startup overhead, so it is more effi‐
706 cient to use this mode than to call lsof repetitively from a
707 shell script, for example.
708 To use repeat mode most efficiently, accompany +|-r with spec‐
710 ification of other lsof selection options, so the amount of
711 kernel memory access lsof does will be kept to a minimum.
712 Options that filter at the process level - e.g., -c, -g, -p,
713 -u - are the most efficient selectors.
714 Repeat mode is useful when coupled with field output (see the
716 -F, option description) and a supervising awk or Perl script,
717 or a C program.
718 -R This option directs lsof to list the Parent Process IDentifi‐
720 cation number in the PPID column.
721 -s [p:s] s alone directs lsof to display file size at all times. It
723 causes the SIZE/OFF output column title to be changed to SIZE.
724 If the file does not have a size, nothing is displayed.
725 When followed by a protocol name (p), either TCP or UDP, a
727 colon (`:') and a comma-separated protocol state name list,
728 the option causes open TCP and UDP files to be excluded if
729 their state name(s) are in the list (s) preceded by a `^'; or
730 included if their name(s) are not preceded by a `^'.
731 When an inclusion list is defined, only network files with
733 state names in the list will be present in the lsof output.
734 Thus, specifying one state name means that only network files
735 with that lone state name will be listed.
736 Case is unimportant in the protocol or state names, but there
738 may be no spaces and the colon (`:') separating the protocol
739 name (p) and the state name list (s) is required.
740 If only TCP and UDP files are to be listed, as controlled by
742 the specified exclusions and inclusions, the -i option must be
743 specified, too. If only a single protocol's files are to be
744 listed, add its name as an argument to the -i option.
745 For example, to list only network files with TCP state LISTEN,
747 use:
748 -iTCP -sTCP:LISTEN
750 Or, for example, to list network files with all UDP states
752 except Idle, use:
753 -iUDP -sUDP:Idle
755 State names vary with UNIX dialects, so it's not possible to
757 provide a complete list. Some common TCP state names are:
758 CLOSED, IDLE, BOUND, LISTEN, ESTABLISHED, SYN_SENT, SYN_RCDV,
759 ESTABLISHED, CLOSE_WAIT, FIN_WAIT1, CLOSING, LAST_ACK,
760 FIN_WAIT_2, and TIME_WAIT. Two common UDP state names are
761 Unbound and Idle.
762 See the lsof FAQ (The FAQ section gives its location.) for
764 more information on how to use protocol state exclusion and
765 inclusion, including examples.
766 The -o (without a following decimal digit count) and -s option
768 (without a following protocol and state name list) are mutu‐
769 ally exclusive; they can't both be specified. When neither is
770 specified, lsof displays whatever value - size or offset - is
771 appropriate and available for the type of file.
772 Since some types of files don't have true sizes - sockets,
774 FIFOs, pipes, etc. - lsof displays for their sizes the content
775 amounts in their associated kernel buffers, if possible.
776 -S [t] This option specifies an optional time-out seconds value for
778 kernel functions - lstat(2), readlink(2), and stat(2) - that
779 might otherwise deadlock. The minimum for t is two; the
780 default, fifteen; when no value is specified, the default is
781 used.
782 See the BLOCKS AND TIMEOUTS section for more information.
784 -T [t] This option controls the reporting of some TCP/TPI informa‐
786 tion, also reported by netstat(1), following the network
787 addresses. In normal output the information appears in paren‐
788 theses, each item except TCP or TPI state name identified by a
789 keyword, followed by `=', separated from others by a single
790 space:
791 <TCP or TPI state name>
793 QR=<read queue length>
794 QS=<send queue length>
795 SO=<socket options and values>
796 SS=<socket states>
797 TF=<TCP flags and values>
798 WR=<window read length>
799 WW=<window write length>
800 Not all values are reported for all UNIX dialects. Items val‐
802 ues (when available) are reported after the item name and '='.
803 When the field output mode is in effect (See OUTPUT FOR OTHER
805 PROGRAMS.) each item appears as a field with a `T' leading
806 character.
807 -T with no following key characters disables TCP/TPI informa‐
809 tion reporting.
810 -T with following characters selects the reporting of specific
812 TCP/TPI information:
813 f selects reporting of socket options,
815 states and values, and TCP flags and
816 values.
817 q selects queue length reporting.
818 s selects connection state reporting.
819 w selects window size reporting.
820 Not all selections are enabled for some UNIX dialects. State
822 may be selected for all dialects and is reported by default.
823 The -h or -? help output for the -T option will show what
824 selections may be used with the UNIX dialect.
825 When -T is used to select information - i.e., it is followed
827 by one or more selection characters - the displaying of state
828 is disabled by default, and it must be explicitly selected
829 again in the characters following -T. (In effect, then, the
830 default is equivalent to -Ts.) For example, if queue lengths
831 and state are desired, use -Tqs.
832 Socket options, socket states, some socket values, TCP flags
834 and one TCP value may be reported (when available in the UNIX
835 dialect) in the form of the names that commonly appear after
836 SO_, so_, SS_, TCP_ and TF_ in the dialect's header files -
837 most often <sys/socket.h>, <sys/socketvar.h> and
838 <netinet/tcp_var.h>. Consult those header files for the mean‐
839 ing of the flags, options, states and values.
840 ``SO='' precedes socket options and values; ``SS='', socket
842 states; and ``TF='', TCP flags and values.
843 If a flag or option has a value, the value will follow an '='
845 and the name -- e.g., ``SO=LINGER=5'', ``SO=QLIM=5'',
846 ``TF=MSS=512''. The following seven values may be reported:
847 Name
849 Reported Description (Common Symbol)
850 KEEPALIVE keep alive time (SO_KEEPALIVE)
852 LINGER linger time (SO_LINGER)
853 MSS maximum segment size (TCP_MAXSEG)
854 PQLEN partial listen queue connections
855 QLEN established listen queue connections
856 QLIM established listen queue limit
857 RCVBUF receive buffer length (SO_RCVBUF)
858 SNDBUF send buffer length (SO_SNDBUF)
859 Details on what socket options and values, socket states, and
861 TCP flags and values may be displayed for particular UNIX
862 dialects may be found in the answer to the ``Why doesn't lsof
863 report socket options, socket states, and TCP flags and values
864 for my dialect?'' and ``Why doesn't lsof report the partial
865 listen queue connection count for my dialect?'' questions in
866 the lsof FAQ (The FAQ section gives its location.)
867 -t This option specifies that lsof should produce terse output
869 with process identifiers only and no header - e.g., so that
870 the output may be piped to kill(1). This option selects the
871 -w option.
872 -u s This option selects the listing of files for the user whose
874 login names or user ID numbers are in the comma-separated set
875 s - e.g., ``abe'', or ``548,root''. (There should be no spa‐
876 ces in the set.)
877 Multiple login names or user ID numbers are joined in a single
879 ORed set before participating in AND option selection.
880 If a login name or user ID is preceded by a `^', it becomes a
882 negation - i.e., files of processes owned by the login name or
883 user ID will never be listed. A negated login name or user ID
884 selection is neither ANDed nor ORed with other selections; it
885 is applied before all other selections and absolutely excludes
886 the listing of the files of the process. For example, to
887 direct lsof to exclude the listing of files belonging to root
888 processes, specify ``-u^root'' or ``-u^0''.
889 -U This option selects the listing of UNIX domain socket files.
891 -v This option selects the listing of lsof version information,
893 including: revision number; when the lsof binary was con‐
894 structed; who constructed the binary and where; the name of
895 the compiler used to construct the lsof binary; the version
896 number of the compiler when readily available; the compiler
897 and loader flags used to construct the lsof binary; and system
898 information, typically the output of uname's -a option.
899 -V This option directs lsof to indicate the items it was asked to
901 list and failed to find - command names, file names, Internet
902 addresses or files, login names, NFS files, PIDs, PGIDs, and
903 UIDs.
904 When other options are ANDed to search options, or com‐
906 pile-time options restrict the listing of some files, lsof may
907 not report that it failed to find a search item when an ANDed
908 option or compile-time option prevents the listing of the open
909 file containing the located search item.
910 For example, ``lsof -V -iTCP@foobar -a -d 999'' may not report
912 a failure to locate open files at ``TCP@foobar'' and may not
913 list any, if none have a file descriptor number of 999. A
914 similar situation arises when HASSECURITY and HASNOSOCKSECU‐
915 RITY are defined at compile time and they prevent the listing
916 of open files.
917 +|-w Enables (+) or disables (-) the suppression of warning mes‐
919 sages.
920 The lsof builder may choose to have warning messages disabled
922 or enabled by default. The default warning message state is
923 indicated in the output of the -h or -? option. Disabling
924 warning messages when they are already disabled or enabling
925 them when already enabled is acceptable.
926 The -t option selects the -w option.
928 -x [fl] This option may accompany the +d and +D options to direct
930 their processing to cross over symbolic links and|or file sys‐
931 tem mount points encountered when scanning the directory (+d)
932 or directory tree (+D).
933 If -x is specified by itself without a following parameter,
935 cross-over processing of both symbolic links and file system
936 mount points is enabled. Note that when -x is specified with‐
937 out a parameter, the next argument must begin with '-' or '+'.
938 The optional 'f' parameter enables file system mount point
940 cross-over processing; 'l', symbolic link cross-over process‐
941 ing.
942 The -x option may not be supplied without also supplying a +d
944 or +D option.
945 -X This is a dialect-specific option.
947 AIX:
949 This IBM AIX RISC/System 6000 option requests the reporting of
950 executed text file and shared library references.
951 WARNING: because this option uses the kernel readx() function,
953 its use on a busy AIX system might cause an application
954 process to hang so completely that it can neither be killed
955 nor stopped. I have never seen this happen or had a report of
956 its happening, but I think there is a remote possibility it
957 could happen.
958 By default use of readx() is disabled. On AIX 5L and above
960 lsof may need setuid-root permission to perform the actions
961 this option requests.
962 The lsof builder may specify that the -X option be restricted
964 to processes whose real UID is root. If that has been done,
965 the -X option will not appear in the -h or -? help output
966 unless the real UID of the lsof process is root. The default
967 lsof distribution allows any UID to specify -X, so by default
968 it will appear in the help output.
969 When AIX readx() use is disabled, lsof may not be able to
971 report information for all text and loader file references,
972 but it may also avoid exacerbating an AIX kernel directory
973 search kernel error, known as the Stale Segment ID bug.
974 The readx() function, used by lsof or any other program to
976 access some sections of kernel virtual memory, can trigger the
977 Stale Segment ID bug. It can cause the kernel's dir_search()
978 function to believe erroneously that part of an in-memory copy
979 of a file system directory has been zeroed. Another applica‐
980 tion process, distinct from lsof, asking the kernel to search
981 the directory - e.g., by using open(2) - can cause
982 dir_search() to loop forever, thus hanging the application
983 process.
984 Consult the lsof FAQ (The FAQ section gives its location.)
986 and the 00README file of the lsof distribution for a more com‐
987 plete description of the Stale Segment ID bug, its APAR, and
988 methods for defining readx() use when compiling lsof.
989 Linux:
991 This Linux option requests that lsof skip the reporting of
992 information on all open TCP, UDP and UDPLITE IPv4 and IPv6
993 files.
994 This Linux option is most useful when the system has an
996 extremely large number of open TCP, UDP and UDPLITE files, the
997 processing of whose information in the /proc/net/tcp* and
998 /proc/net/udp* files would take lsof a long time, and whose
999 reporting is not of interest.
1000 Use this option with care and only when you are sure that the
1002 information you want lsof to display isn't associated with
1003 open TCP, UDP or UDPLITE socket files.
1004 Solaris 10 and above:
1006 This Solaris 10 and above option requests the reporting of
1007 cached paths for files that have been deleted - i.e., removed
1008 with rm(1) or unlink(2).
1009 The cached path is followed by the string `` (deleted)'' to
1011 indicate that the path by which the file was opened has been
1012 deleted.
1013 Because intervening changes made to the path - i.e., renames
1015 with mv(1) or rename(2) - are not recorded in the cached path,
1016 what lsof reports is only the path by which the file was
1017 opened, not its possibly different final path.
1018 -z [z] specifies how Solaris 10 and higher zone information is to be
1020 handled.
1021 Without a following argument - e.g., NO z - the option speci‐
1023 fies that zone names are to be listed in the ZONE output col‐
1024 umn.
1025 The -z option may be followed by a zone name, z. That causes
1027 lsof to list only open files for processes in that zone. Mul‐
1028 tiple -z z option and argument pairs may be specified to form
1029 a list of named zones. Any open file of any process in any of
1030 the zones will be listed, subject to other conditions speci‐
1031 fied by other options and arguments.
1032 -Z [Z] specifies how SELinux security contexts are to be handled.
1034 This option and 'Z' field output character support are inhib‐
1035 ited when SELinux is disabled in the running Linux kernel.
1036 See OUTPUT FOR OTHER PROGRAMS for more information on the 'Z'
1037 field output character.
1038 Without a following argument - e.g., NO Z - the option speci‐
1040 fies that security contexts are to be listed in the SECU‐
1041 RITY-CONTEXT output column.
1042 The -Z option may be followed by a wildcard security context
1044 name, Z. That causes lsof to list only open files for pro‐
1045 cesses in that security context. Multiple -Z Z option and
1046 argument pairs may be specified to form a list of security
1047 contexts. Any open file of any process in any of the security
1048 contexts will be listed, subject to other conditions specified
1049 by other options and arguments. Note that Z can be A:B:C or
1050 *:B:C or A:B:* or *:*:C to match against the A:B:C context.
1051 -- The double minus sign option is a marker that signals the end
1053 of the keyed options. It may be used, for example, when the
1054 first file name begins with a minus sign. It may also be used
1055 when the absence of a value for the last keyed option must be
1056 signified by the presence of a minus sign in the following
1057 option and before the start of the file names.
1058 names These are path names of specific files to list. Symbolic
1060 links are resolved before use. The first name may be sepa‐
1061 rated from the preceding options with the ``--'' option.
1062 If a name is the mounted-on directory of a file system or the
1064 device of the file system, lsof will list all the files open
1065 on the file system. To be considered a file system, the name
1066 must match a mounted-on directory name in mount(8) output, or
1067 match the name of a block device associated with a mounted-on
1068 directory name. The +|-f option may be used to force lsof to
1069 consider a name a file system identifier (+f) or a simple file
1070 (-f).
1071 If name is a path to a directory that is not the mounted-on
1073 directory name of a file system, it is treated just as a regu‐
1074 lar file is treated - i.e., its listing is restricted to pro‐
1075 cesses that have it open as a file or as a process-specific
1076 directory, such as the root or current working directory. To
1077 request that lsof look for open files inside a directory name,
1078 use the +d s and +D D options.
1079 If a name is the base name of a family of multiplexed files -
1081 e. g, AIX's /dev/pt[cs] - lsof will list all the associated
1082 multiplexed files on the device that are open - e.g.,
1083 /dev/pt[cs]/1, /dev/pt[cs]/2, etc.
1084 If a name is a UNIX domain socket name, lsof will usually
1086 search for it by the characters of the name alone - exactly as
1087 it is specified and is recorded in the kernel socket struc‐
1088 ture. (See the next paragraph for an exception to that rule
1089 for Linux.) Specifying a relative path - e.g., ./file - in
1090 place of the file's absolute path - e.g., /tmp/file - won't
1091 work because lsof must match the characters you specify with
1092 what it finds in the kernel UNIX domain socket structures.
1093 If a name is a Linux UNIX domain socket name, in one case lsof
1095 is able to search for it by its device and inode number,
1096 allowing name to be a relative path. The case requires that
1097 the absolute path -- i.e., one beginning with a slash ('/') be
1098 used by the process that created the socket, and hence be
1099 stored in the /proc/net/unix file; and it requires that lsof
1100 be able to obtain the device and node numbers of both the
1101 absolute path in /proc/net/unix and name via successful
1102 stat(2) system calls. When those conditions are met, lsof
1103 will be able to search for the UNIX domain socket when some
1104 path to it is is specified in name. Thus, for example, if the
1105 path is /dev/log, and an lsof search is initiated when the
1106 working directory is /dev, then name could be ./log.
1107 If a name is none of the above, lsof will list any open files
1109 whose device and inode match that of the specified path name.
1110 If you have also specified the -b option, the only names you
1112 may safely specify are file systems for which your mount table
1113 supplies alternate device numbers. See the AVOIDING KERNEL
1114 BLOCKS and ALTERNATE DEVICE NUMBERS sections for more informa‐
1115 tion.
1116 Multiple file names are joined in a single ORed set before
1118 participating in AND option selection.
1119
1121 Lsof supports the recognition of AFS files for these dialects (and AFS
1122 versions):
1123 AIX 4.1.4 (AFS 3.4a)
1125 HP-UX 9.0.5 (AFS 3.4a)
1126 Linux 1.2.13 (AFS 3.3)
1127 Solaris 2.[56] (AFS 3.4a)
1128 It may recognize AFS files on other versions of these dialects, but has
1130 not been tested there. Depending on how AFS is implemented, lsof may
1131 recognize AFS files in other dialects, or may have difficulties recog‐
1132 nizing AFS files in the supported dialects.
1133 Lsof may have trouble identifying all aspects of AFS files in supported
1135 dialects when AFS kernel support is implemented via dynamic modules
1136 whose addresses do not appear in the kernel's variable name list. In
1137 that case, lsof may have to guess at the identity of AFS files, and
1138 might not be able to obtain volume information from the kernel that is
1139 needed for calculating AFS volume node numbers. When lsof can't com‐
1140 pute volume node numbers, it reports blank in the NODE column.
1141 The -A A option is available in some dialect implementations of lsof
1143 for specifying the name list file where dynamic module kernel addresses
1144 may be found. When this option is available, it will be listed in the
1145 lsof help output, presented in response to the -h or -?
1146 See the lsof FAQ (The FAQ section gives its location.) for more infor‐
1148 mation about dynamic modules, their symbols, and how they affect lsof
1149 options.
1150 Because AFS path lookups don't seem to participate in the kernel's name
1152 cache operations, lsof can't identify path name components for AFS
1153 files.
1154
1156 Lsof has three features that may cause security concerns. First, its
1157 default compilation mode allows anyone to list all open files with it.
1158 Second, by default it creates a user-readable and user-writable device
1159 cache file in the home directory of the real user ID that executes
1160 lsof. (The list-all-open-files and device cache features may be dis‐
1161 abled when lsof is compiled.) Third, its -k and -m options name alter‐
1162 nate kernel name list or memory files.
1163 Restricting the listing of all open files is controlled by the com‐
1165 pile-time HASSECURITY and HASNOSOCKSECURITY options. When HASSECURITY
1166 is defined, lsof will allow only the root user to list all open files.
1167 The non-root user may list only open files of processes with the same
1168 user IDentification number as the real user ID number of the lsof
1169 process (the one that its user logged on with).
1170 However, if HASSECURITY and HASNOSOCKSECURITY are both defined, anyone
1172 may list open socket files, provided they are selected with the -i
1173 option.
1174 When HASSECURITY is not defined, anyone may list all open files.
1176 Help output, presented in response to the -h or -? option, gives the
1178 status of the HASSECURITY and HASNOSOCKSECURITY definitions.
1179 See the Security section of the 00README file of the lsof distribution
1181 for information on building lsof with the HASSECURITY and HASNOSOCKSE‐
1182 CURITY options enabled.
1183 Creation and use of a user-readable and user-writable device cache file
1185 is controlled by the compile-time HASDCACHE option. See the DEVICE
1186 CACHE FILE section and the sections that follow it for details on how
1187 its path is formed. For security considerations it is important to
1188 note that in the default lsof distribution, if the real user ID under
1189 which lsof is executed is root, the device cache file will be written
1190 in root's home directory - e.g., / or /root. When HASDCACHE is not
1191 defined, lsof does not write or attempt to read a device cache file.
1192 When HASDCACHE is defined, the lsof help output, presented in response
1194 to the -h, -D?, or -? options, will provide device cache file handling
1195 information. When HASDCACHE is not defined, the -h or -? output will
1196 have no -D option description.
1197 Before you decide to disable the device cache file feature - enabling
1199 it improves the performance of lsof by reducing the startup overhead of
1200 examining all the nodes in /dev (or /devices) - read the discussion of
1201 it in the 00DCACHE file of the lsof distribution and the lsof FAQ (The
1202 FAQ section gives its location.)
1203 WHEN IN DOUBT, YOU CAN TEMPORARILY DISABLE THE USE OF THE DEVICE CACHE
1205 FILE WITH THE -Di OPTION.
1206 When lsof user declares alternate kernel name list or memory files with
1208 the -k and -m options, lsof checks the user's authority to read them
1209 with access(2). This is intended to prevent whatever special power
1210 lsof's modes might confer on it from letting it read files not normally
1211 accessible via the authority of the real user ID.
1212
1214 This section describes the information lsof lists for each open file.
1215 See the OUTPUT FOR OTHER PROGRAMS section for additional information on
1216 output that can be processed by another program.
1217 Lsof only outputs printable (declared so by isprint(3)) 8 bit charac‐
1219 ters. Non-printable characters are printed in one of three forms: the
1220 C ``\[bfrnt]'' form; the control character `^' form (e.g., ``^@''); or
1221 hexadecimal leading ``\x'' form (e.g., ``\xab''). Space is non-print‐
1222 able in the COMMAND column (``\x20'') and printable elsewhere.
1223 For some dialects - if HASSETLOCALE is defined in the dialect's
1225 machine.h header file - lsof will print the extended 8 bit characters
1226 of a language locale. The lsof process must be supplied a language
1227 locale environment variable (e.g., LANG) whose value represents a known
1228 language locale in which the extended characters are considered print‐
1229 able by isprint(3). Otherwise lsof considers the extended characters
1230 non-printable and prints them according to its rules for non-printable
1231 characters, stated above. Consult your dialect's setlocale(3) man page
1232 for the names of other environment variables that may be used in place
1233 of LANG - e.g., LC_ALL, LC_CTYPE, etc.
1234 Lsof's language locale support for a dialect also covers wide charac‐
1236 ters - e.g., UTF-8 - when HASSETLOCALE and HASWIDECHAR are defined in
1237 the dialect's machine.h header file, and when a suitable language
1238 locale has been defined in the appropriate environment variable for the
1239 lsof process. Wide characters are printable under those conditions if
1240 iswprint(3) reports them to be. If HASSETLOCALE, HASWIDECHAR and a
1241 suitable language locale aren't defined, or if iswprint(3) reports wide
1242 characters that aren't printable, lsof considers the wide characters
1243 non-printable and prints each of their 8 bits according to its rules
1244 for non-printable characters, stated above.
1245 Consult the answers to the "Language locale support" questions in the
1247 lsof FAQ (The FAQ section gives its location.) for more information.
1248 Lsof dynamically sizes the output columns each time it runs, guarantee‐
1250 ing that each column is a minimum size. It also guarantees that each
1251 column is separated from its predecessor by at least one space.
1252 COMMAND contains the first nine characters of the name of the UNIX
1254 command associated with the process. If a non-zero w value
1255 is specified to the +c w option, the column contains the
1256 first w characters of the name of the UNIX command associ‐
1257 ated with the process up to the limit of characters supplied
1258 to lsof by the UNIX dialect. (See the description of the +c
1259 w command or the lsof FAQ for more information. The FAQ
1260 section gives its location.)
1261 If w is less than the length of the column title, ``COM‐
1263 MAND'', it will be raised to that length.
1264 If a zero w value is specified to the +c w option, the col‐
1266 umn contains all the characters of the name of the UNIX com‐
1267 mand associated with the process.
1268 All command name characters maintained by the kernel in its
1270 structures are displayed in field output when the command
1271 name descriptor (`c') is specified. See the OUTPUT FOR
1272 OTHER COMMANDS section for information on selecting field
1273 output and the associated command name descriptor.
1274 PID is the Process IDentification number of the process.
1276 ZONE is the Solaris 10 and higher zone name. This column must be
1278 selected with the -z option.
1279 SECURITY-CONTEXT
1281 is the SELinux security context. This column must be
1282 selected with the -Z option. Note that the -Z option is
1283 inhibited when SELinux is disabled in the running Linux ker‐
1284 nel.
1285 PPID is the Parent Process IDentification number of the process.
1287 It is only displayed when the -R option has been specified.
1288 PGID is the process group IDentification number associated with
1290 the process. It is only displayed when the -g option has
1291 been specified.
1292 USER is the user ID number or login name of the user to whom the
1294 process belongs, usually the same as reported by ps(1).
1295 However, on Linux USER is the user ID number or login that
1296 owns the directory in /proc where lsof finds information
1297 about the process. Usually that is the same value reported
1298 by ps(1), but may differ when the process has changed its
1299 effective user ID. (See the -l option description for
1300 information on when a user ID number or login name is dis‐
1301 played.)
1302 FD is the File Descriptor number of the file or:
1304 cwd current working directory;
1306 Lnn library references (AIX);
1307 err FD information error (see NAME column);
1308 jld jail directory (FreeBSD);
1309 ltx shared library text (code and data);
1310 Mxx hex memory-mapped type number xx.
1311 m86 DOS Merge mapped file;
1312 mem memory-mapped file;
1313 mmap memory-mapped device;
1314 pd parent directory;
1315 rtd root directory;
1316 tr kernel trace file (OpenBSD);
1317 txt program text (code and data);
1318 v86 VP/ix mapped file;
1319 FD is followed by one of these characters, describing the
1321 mode under which the file is open:
1322 r for read access;
1324 w for write access;
1325 u for read and write access;
1326 space if mode unknown and no lock
1327 character follows;
1328 `-' if mode unknown and lock
1329 character follows.
1330 The mode character is followed by one of these lock charac‐
1332 ters, describing the type of lock applied to the file:
1333 N for a Solaris NFS lock of unknown type;
1335 r for read lock on part of the file;
1336 R for a read lock on the entire file;
1337 w for a write lock on part of the file;
1338 W for a write lock on the entire file;
1339 u for a read and write lock of any length;
1340 U for a lock of unknown type;
1341 x for an SCO OpenServer Xenix lock on part of the
1342 file;
1343 X for an SCO OpenServer Xenix lock on the entire
1344 file;
1345 space if there is no lock.
1346 See the LOCKS section for more information on the lock
1348 information character.
1349 The FD column contents constitutes a single field for pars‐
1351 ing in post-processing scripts.
1352 TYPE is the type of the node associated with the file - e.g.,
1354 GDIR, GREG, VDIR, VREG, etc.
1355 or ``IPv4'' for an IPv4 socket;
1357 or ``IPv6'' for an open IPv6 network file - even if its
1359 address is IPv4, mapped in an IPv6 address;
1360 or ``ax25'' for a Linux AX.25 socket;
1362 or ``inet'' for an Internet domain socket;
1364 or ``lla'' for a HP-UX link level access file;
1366 or ``rte'' for an AF_ROUTE socket;
1368 or ``sock'' for a socket of unknown domain;
1370 or ``unix'' for a UNIX domain socket;
1372 or ``x.25'' for an HP-UX x.25 socket;
1374 or ``BLK'' for a block special file;
1376 or ``CHR'' for a character special file;
1378 or ``DEL'' for a Linux map file that has been deleted;
1380 or ``DIR'' for a directory;
1382 or ``DOOR'' for a VDOOR file;
1384 or ``FIFO'' for a FIFO special file;
1386 or ``KQUEUE'' for a BSD style kernel event queue file;
1388 or ``LINK'' for a symbolic link file;
1390 or ``MPB'' for a multiplexed block file;
1392 or ``MPC'' for a multiplexed character file;
1394 or ``NOFD'' for a Linux /proc/<PID>/fd directory that can't
1396 be opened -- the directory path appears in the NAME column,
1397 followed by an error message;
1398 or ``PAS'' for a /proc/as file;
1400 or ``PAXV'' for a /proc/auxv file;
1402 or ``PCRE'' for a /proc/cred file;
1404 or ``PCTL'' for a /proc control file;
1406 or ``PCUR'' for the current /proc process;
1408 or ``PCWD'' for a /proc current working directory;
1410 or ``PDIR'' for a /proc directory;
1412 or ``PETY'' for a /proc executable type (etype);
1414 or ``PFD'' for a /proc file descriptor;
1416 or ``PFDR'' for a /proc file descriptor directory;
1418 or ``PFIL'' for an executable /proc file;
1420 or ``PFPR'' for a /proc FP register set;
1422 or ``PGD'' for a /proc/pagedata file;
1424 or ``PGID'' for a /proc group notifier file;
1426 or ``PIPE'' for pipes;
1428 or ``PLC'' for a /proc/lwpctl file;
1430 or ``PLDR'' for a /proc/lpw directory;
1432 or ``PLDT'' for a /proc/ldt file;
1434 or ``PLPI'' for a /proc/lpsinfo file;
1436 or ``PLST'' for a /proc/lstatus file;
1438 or ``PLU'' for a /proc/lusage file;
1440 or ``PLWG'' for a /proc/gwindows file;
1442 or ``PLWI'' for a /proc/lwpsinfo file;
1444 or ``PLWS'' for a /proc/lwpstatus file;
1446 or ``PLWU'' for a /proc/lwpusage file;
1448 or ``PLWX'' for a /proc/xregs file'
1450 or ``PMAP'' for a /proc map file (map);
1452 or ``PMEM'' for a /proc memory image file;
1454 or ``PNTF'' for a /proc process notifier file;
1456 or ``POBJ'' for a /proc/object file;
1458 or ``PODR'' for a /proc/object directory;
1460 or ``POLP'' for an old format /proc light weight process
1462 file;
1463 or ``POPF'' for an old format /proc PID file;
1465 or ``POPG'' for an old format /proc page data file;
1467 or ``PORT'' for a SYSV named pipe;
1469 or ``PREG'' for a /proc register file;
1471 or ``PRMP'' for a /proc/rmap file;
1473 or ``PRTD'' for a /proc root directory;
1475 or ``PSGA'' for a /proc/sigact file;
1477 or ``PSIN'' for a /proc/psinfo file;
1479 or ``PSTA'' for a /proc status file;
1481 or ``PSXSEM'' for a POSIX semaphore file;
1483 or ``PSXSHM'' for a POSIX shared memory file;
1485 or ``PUSG'' for a /proc/usage file;
1487 or ``PW'' for a /proc/watch file;
1489 or ``PXMP'' for a /proc/xmap file;
1491 or ``REG'' for a regular file;
1493 or ``SMT'' for a shared memory transport file;
1495 or ``STSO'' for a stream socket;
1497 or ``UNNM'' for an unnamed type file;
1499 or ``XNAM'' for an OpenServer Xenix special file of unknown
1501 type;
1502 or ``XSEM'' for an OpenServer Xenix semaphore file;
1504 or ``XSD'' for an OpenServer Xenix shared data file;
1506 or the four type number octets if the corresponding name
1508 isn't known.
1509 FILE-ADDR contains the kernel file structure address when f has been
1511 specified to +f;
1512 FCT contains the file reference count from the kernel file
1514 structure when c has been specified to +f;
1515 FILE-FLAG when g or G has been specified to +f, this field contains
1517 the contents of the f_flag[s] member of the kernel file
1518 structure and the kernel's per-process open file flags (if
1519 available); `G' causes them to be displayed in hexadecimal;
1520 `g', as short-hand names; two lists may be displayed with
1521 entries separated by commas, the lists separated by a semi‐
1522 colon (`;'); the first list may contain short-hand names for
1523 f_flag[s] values from the following table:
1524 AIO asynchronous I/O (e.g., FAIO)
1526 AP append
1527 ASYN asynchronous I/O (e.g., FASYNC)
1528 BAS block, test, and set in use
1529 BKIU block if in use
1530 BL use block offsets
1531 BSK block seek
1532 CA copy avoid
1533 CIO concurrent I/O
1534 CLON clone
1535 CLRD CL read
1536 CR create
1537 DF defer
1538 DFI defer IND
1539 DFLU data flush
1540 DIR direct
1541 DLY delay
1542 DOCL do clone
1543 DSYN data-only integrity
1544 DTY must be a directory
1545 EVO event only
1546 EX open for exec
1547 EXCL exclusive open
1548 FSYN synchronous writes
1549 GCDF defer during unp_gc() (AIX)
1550 GCMK mark during unp_gc() (AIX)
1551 GTTY accessed via /dev/tty
1552 HUP HUP in progress
1553 KERN kernel
1554 KIOC kernel-issued ioctl
1555 LCK has lock
1556 LG large file
1557 MBLK stream message block
1558 MK mark
1559 MNT mount
1560 MSYN multiplex synchronization
1561 NATM don't update atime
1562 NB non-blocking I/O
1563 NBDR no BDRM check
1564 NBIO SYSV non-blocking I/O
1565 NBF n-buffering in effect
1566 NC no cache
1567 ND no delay
1568 NDSY no data synchronization
1569 NET network
1570 NFLK don't follow links
1571 NMFS NM file system
1572 NOTO disable background stop
1573 NSH no share
1574 NTTY no controlling TTY
1575 OLRM OLR mirror
1576 PAIO POSIX asynchronous I/O
1577 PP POSIX pipe
1578 R read
1579 RC file and record locking cache
1580 REV revoked
1581 RSH shared read
1582 RSYN read synchronization
1583 RW read and write access
1584 SL shared lock
1585 SNAP cooked snapshot
1586 SOCK socket
1587 SQSH Sequent shared set on open
1588 SQSV Sequent SVM set on open
1589 SQR Sequent set repair on open
1590 SQS1 Sequent full shared open
1591 SQS2 Sequent partial shared open
1592 STPI stop I/O
1593 SWR synchronous read
1594 SYN file integrity while writing
1595 TCPM avoid TCP collision
1596 TR truncate
1597 W write
1598 WKUP parallel I/O synchronization
1599 WTG parallel I/O synchronization
1600 VH vhangup pending
1601 VTXT virtual text
1602 XL exclusive lock
1603 this list of names was derived from F* #define's in dialect
1605 header files <fcntl.h>, <linux</fs.h>, <sys/fcntl.c>,
1606 <sys/fcntlcom.h>, and <sys/file.h>; see the lsof.h header
1607 file for a list showing the correspondence between the above
1608 short-hand names and the header file definitions;
1609 the second list (after the semicolon) may contain short-hand
1611 names for kernel per-process open file flags from this ta‐
1612 ble:
1613 ALLC allocated
1615 BR the file has been read
1616 BHUP activity stopped by SIGHUP
1617 BW the file has been written
1618 CLSG closing
1619 CX close-on-exec (see fcntl(F_SETFD))
1620 LCK lock was applied
1621 MP memory-mapped
1622 OPIP open pending - in progress
1623 RSVW reserved wait
1624 SHMT UF_FSHMAT set (AIX)
1625 USE in use (multi-threaded)
1626 NODE-ID (or INODE-ADDR for some dialects) contains a unique identi‐
1628 fier for the file node (usually the kernel vnode or inode
1629 address, but also occasionally a concatenation of device and
1630 node number) when n has been specified to +f;
1631 DEVICE contains the device numbers, separated by commas, for a
1633 character special, block special, regular, directory or NFS
1634 file;
1635 or ``memory'' for a memory file system node under Tru64
1637 UNIX;
1638 or the address of the private data area of a Solaris socket
1640 stream;
1641 or a kernel reference address that identifies the file (The
1643 kernel reference address may be used for FIFO's, for exam‐
1644 ple.);
1645 or the base address or device name of a Linux AX.25 socket
1647 device.
1648 Usually only the lower thirty two bits of Tru64 UNIX kernel
1650 addresses are displayed.
1651 SIZE, SIZE/OFF, or OFFSET
1653 is the size of the file or the file offset in bytes. A
1654 value is displayed in this column only if it is available.
1655 Lsof displays whatever value - size or offset - is appropri‐
1656 ate for the type of the file and the version of lsof.
1657 On some UNIX dialects lsof can't obtain accurate or consis‐
1659 tent file offset information from its kernel data sources,
1660 sometimes just for particular kinds of files (e.g., socket
1661 files.) In other cases, files don't have true sizes - e.g.,
1662 sockets, FIFOs, pipes - so lsof displays for their sizes the
1663 content amounts it finds in their kernel buffer descriptors
1664 (e.g., socket buffer size counts or TCP/IP window sizes.)
1665 Consult the lsof FAQ (The FAQ section gives its location.)
1666 for more information.
1667 The file size is displayed in decimal; the offset is nor‐
1669 mally displayed in decimal with a leading ``0t'' if it con‐
1670 tains 8 digits or less; in hexadecimal with a leading ``0x''
1671 if it is longer than 8 digits. (Consult the -o o option
1672 description for information on when 8 might default to some
1673 other value.)
1674 Thus the leading ``0t'' and ``0x'' identify an offset when
1676 the column may contain both a size and an offset (i.e., its
1677 title is SIZE/OFF).
1678 If the -o option is specified, lsof always displays the file
1680 offset (or nothing if no offset is available) and labels the
1681 column OFFSET. The offset always begins with ``0t'' or
1682 ``0x'' as described above.
1683 The lsof user can control the switch from ``0t'' to ``0x''
1685 with the -o o option. Consult its description for more
1686 information.
1687 If the -s option is specified, lsof always displays the file
1689 size (or nothing if no size is available) and labels the
1690 column SIZE. The -o and -s options are mutually exclusive;
1691 they can't both be specified.
1692 For files that don't have a fixed size - e.g., don't reside
1694 on a disk device - lsof will display appropriate information
1695 about the current size or position of the file if it is
1696 available in the kernel structures that define the file.
1697 NLINK contains the file link count when +L has been specified;
1699 NODE is the node number of a local file;
1701 or the inode number of an NFS file in the server host;
1703 or the Internet protocol type - e. g, ``TCP'';
1705 or ``STR'' for a stream;
1707 or ``CCITT'' for an HP-UX x.25 socket;
1709 or the IRQ or inode number of a Linux AX.25 socket device.
1711 NAME is the name of the mount point and file system on which the
1713 file resides;
1714 or the name of a file specified in the names option (after
1716 any symbolic links have been resolved);
1717 or the name of a character special or block special device;
1719 or the local and remote Internet addresses of a network
1721 file; the local host name or IP number is followed by a
1722 colon (':'), the port, ``->'', and the two-part remote
1723 address; IP addresses may be reported as numbers or names,
1724 depending on the +|-M, -n, and -P options; colon-separated
1725 IPv6 numbers are enclosed in square brackets; IPv4
1726 INADDR_ANY and IPv6 IN6_IS_ADDR_UNSPECIFIED addresses, and
1727 zero port numbers are represented by an asterisk ('*'); a
1728 UDP destination address may be followed by the amount of
1729 time elapsed since the last packet was sent to the destina‐
1730 tion; TCP, UDP and UDPLITE remote addresses may be followed
1731 by TCP/TPI information in parentheses - state (e.g.,
1732 ``(ESTABLISHED)'', ``(Unbound)''), queue sizes, and window
1733 sizes (not all dialects) - in a fashion similar to what net‐
1734 stat(1) reports; see the -T option description or the
1735 description of the TCP/TPI field in OUTPUT FOR OTHER PRO‐
1736 GRAMS for more information on state, queue size, and window
1737 size;
1738 or the address or name of a UNIX domain socket, possibly
1740 including a stream clone device name, a file system object's
1741 path name, local and foreign kernel addresses, socket pair
1742 information, and a bound vnode address;
1743 or the local and remote mount point names of an NFS file;
1745 or ``STR'', followed by the stream name;
1747 or a stream character device name, followed by ``->'' and
1749 the stream name or a list of stream module names, separated
1750 by ``->'';
1751 or ``STR:'' followed by the SCO OpenServer stream device and
1753 module names, separated by ``->'';
1754 or system directory name, `` -- '', and as many components
1756 of the path name as lsof can find in the kernel's name cache
1757 for selected dialects (See the KERNEL NAME CACHE section for
1758 more information.);
1759 or ``PIPE->'', followed by a Solaris kernel pipe destination
1761 address;
1762 or ``COMMON:'', followed by the vnode device information
1764 structure's device name, for a Solaris common vnode;
1765 or the address family, followed by a slash (`/'), followed
1767 by fourteen comma-separated bytes of a non-Internet raw
1768 socket address;
1769 or the HP-UX x.25 local address, followed by the virtual
1771 connection number (if any), followed by the remote address
1772 (if any);
1773 or ``(dead)'' for disassociated Tru64 UNIX files - typically
1775 terminal files that have been flagged with the TIOCNOTTY
1776 ioctl and closed by daemons;
1777 or ``rd=<offset>'' and ``wr=<offset>'' for the values of the
1779 read and write offsets of a FIFO;
1780 or ``clone n:/dev/event'' for SCO OpenServer file clones of
1782 the /dev/event device, where n is the minor device number of
1783 the file;
1784 or ``(socketpair: n)'' for a Solaris 2.6, 8, 9 or 10 UNIX
1786 domain socket, created by the socketpair(3N) network func‐
1787 tion;
1788 or ``no PCB'' for socket files that do not have a protocol
1790 block associated with them, optionally followed by ``,
1791 CANTSENDMORE'' if sending on the socket has been disabled,
1792 or ``, CANTRCVMORE'' if receiving on the socket has been
1793 disabled (e.g., by the shutdown(2) function);
1794 or the local and remote addresses of a Linux IPX socket file
1796 in the form <net>:[<node>:]<port>, followed in parentheses
1797 by the transmit and receive queue sizes, and the connection
1798 state;
1799 or ``dgram'' or ``stream'' for the type UnixWare 7.1.1 and
1801 above in-kernel UNIX domain sockets, followed by a colon
1802 (':') and the local path name when available, followed by
1803 ``->'' and the remote path name or kernel socket address in
1804 hexadecimal when available.
1805 For dialects that support a ``namefs'' file system, allowing one file
1807 to be attached to another with fattach(3C), lsof will add
1808 ``(FA:<address1><direction><address2>)'' to the NAME column.
1809 <address1> and <address2> are hexadecimal vnode addresses. <direction>
1810 will be ``<-'' if <address2> has been fattach'ed to this vnode whose
1811 address is <address1>; and ``->'' if <address1>, the vnode address of
1812 this vnode, has been fattach'ed to <address2>. <address1> may be omit‐
1813 ted if it already appears in the DEVICE column.
1814 Lsof may add two parenthetical notes to the NAME column for open
1816 Solaris 10 files: ``(?)'' if lsof considers the path name of question‐
1817 able accuracy; and ``(deleted)'' if the -X option has been specified
1818 and lsof detects the open file's path name has been deleted. Consult
1819 the lsof FAQ (The FAQ section gives its location.) for more informa‐
1820 tion on these NAME column additions.
1821
1823 Lsof can't adequately report the wide variety of UNIX dialect file
1824 locks in a single character. What it reports in a single character is
1825 a compromise between the information it finds in the kernel and the
1826 limitations of the reporting format.
1827 Moreover, when a process holds several byte level locks on a file, lsof
1829 only reports the status of the first lock it encounters. If it is a
1830 byte level lock, then the lock character will be reported in lower case
1831 - i.e., `r', `w', or `x' - rather than the upper case equivalent
1832 reported for a full file lock.
1833 Generally lsof can only report on locks held by local processes on
1835 local files. When a local process sets a lock on a remotely mounted
1836 (e.g., NFS) file, the remote server host usually records the lock
1837 state. One exception is Solaris - at some patch levels of 2.3, and in
1838 all versions above 2.4, the Solaris kernel records information on
1839 remote locks in local structures.
1840 Lsof has trouble reporting locks for some UNIX dialects. Consult the
1842 BUGS section of this manual page or the lsof FAQ (The FAQ section gives
1843 its location.) for more information.
1844
1846 When the -F option is specified, lsof produces output that is suitable
1847 for processing by another program - e.g, an awk or Perl script, or a C
1848 program.
1849 Each unit of information is output in a field that is identified with a
1851 leading character and terminated by a NL (012) (or a NUL (000) if the 0
1852 (zero) field identifier character is specified.) The data of the field
1853 follows immediately after the field identification character and
1854 extends to the field terminator.
1855 It is possible to think of field output as process and file sets. A
1857 process set begins with a field whose identifier is `p' (for process
1858 IDentifier (PID)). It extends to the beginning of the next PID field
1859 or the beginning of the first file set of the process, whichever comes
1860 first. Included in the process set are fields that identify the com‐
1861 mand, the process group IDentification (PGID) number, and the user ID
1862 (UID) number or login name.
1863 A file set begins with a field whose identifier is `f' (for file
1865 descriptor). It is followed by lines that describe the file's access
1866 mode, lock state, type, device, size, offset, inode, protocol, name and
1867 stream module names. It extends to the beginning of the next file or
1868 process set, whichever comes first.
1869 When the NUL (000) field terminator has been selected with the 0 (zero)
1871 field identifier character, lsof ends each process and file set with a
1872 NL (012) character.
1873 Lsof always produces one field, the PID (`p') field. All other fields
1875 may be declared optionally in the field identifier character list that
1876 follows the -F option. When a field selection character identifies an
1877 item lsof does not normally list - e.g., PPID, selected with -R - spec‐
1878 ification of the field character - e.g., ``-FR'' - also selects the
1879 listing of the item.
1880 It is entirely possible to select a set of fields that cannot easily be
1882 parsed - e.g., if the field descriptor field is not selected, it may be
1883 difficult to identify file sets. To help you avoid this difficulty,
1884 lsof supports the -F option; it selects the output of all fields with
1885 NL terminators (the -F0 option pair selects the output of all fields
1886 with NUL terminators). For compatibility reasons neither -F nor -F0
1887 select the raw device field.
1888 These are the fields that lsof will produce. The single character
1890 listed first is the field identifier.
1891 a file access mode
1893 c process command name (all characters from proc or
1894 user structure)
1895 C file structure share count
1896 d file's device character code
1897 D file's major/minor device number (0x<hexadecimal>)
1898 f file descriptor
1899 F file structure address (0x<hexadecimal>)
1900 G file flaGs (0x<hexadecimal>; names if +fg follows)
1901 i file's inode number
1902 k link count
1903 l file's lock status
1904 L process login name
1905 m marker between repeated output
1906 n file name, comment, Internet address
1907 N node identifier (ox<hexadecimal>
1908 o file's offset (decimal)
1909 p process ID (always selected)
1910 g process group ID
1911 P protocol name
1912 r raw device number (0x<hexadecimal>)
1913 R parent process ID
1914 s file's size (decimal)
1915 S file's stream identification
1916 t file's type
1917 T TCP/TPI information, identified by prefixes (the
1918 `=' is part of the prefix):
1919 QR=<read queue size>
1920 QS=<send queue size>
1921 SO=<socket options and values> (not all dialects)
1922 SS=<socket states> (not all dialects)
1923 ST=<connection state>
1924 TF=<TCP flags and values> (not all dialects)
1925 WR=<window read size> (not all dialects)
1926 WW=<window write size> (not all dialects)
1927 (TCP/TPI information isn't reported for all supported
1928 UNIX dialects. The -h or -? help output for the
1929 -T option will show what TCP/TPI reporting can be
1930 requested.)
1931 u process user ID
1932 z Solaris 10 and higher zone name
1933 Z SELinux security context (inhibited when SELinux is disabled)
1934 0 use NUL field terminator character in place of NL
1935 1-9 dialect-specific field identifiers (The output
1936 of -F? identifies the information to be found
1937 in dialect-specific fields.)
1938 You can get on-line help information on these characters and their
1940 descriptions by specifying the -F? option pair. (Escape the `?' char‐
1941 acter as your shell requires.) Additional information on field content
1942 can be found in the OUTPUT section.
1943 As an example, ``-F pcfn'' will select the process ID (`p'), command
1945 name (`c'), file descriptor (`f') and file name (`n') fields with an NL
1946 field terminator character; ``-F pcfn0'' selects the same output with a
1947 NUL (000) field terminator character.
1948 Lsof doesn't produce all fields for every process or file set, only
1950 those that are available. Some fields are mutually exclusive: file
1951 device characters and file major/minor device numbers; file inode num‐
1952 ber and protocol name; file name and stream identification; file size
1953 and offset. One or the other member of these mutually exclusive sets
1954 will appear in field output, but not both.
1955 Normally lsof ends each field with a NL (012) character. The 0 (zero)
1957 field identifier character may be specified to change the field termi‐
1958 nator character to a NUL (000). A NUL terminator may be easier to
1959 process with xargs [4m(1), for example, or with programs whose quoting
1960 mechanisms may not easily cope with the range of characters in the
1961 field output. When the NUL field terminator is in use, lsof ends each
1962 process and file set with a NL (012).
1963 Three aids to producing programs that can process lsof field output are
1965 included in the lsof distribution. The first is a C header file,
1966 lsof_fields.h, that contains symbols for the field identification char‐
1967 acters, indexes for storing them in a table, and explanation strings
1968 that may be compiled into programs. Lsof uses this header file.
1969 The second aid is a set of sample scripts that process field output,
1971 written in awk, Perl 4, and Perl 5. They're located in the scripts
1972 subdirectory of the lsof distribution.
1973 The third aid is the C library used for the lsof test suite. The test
1975 suite is written in C and uses field output to validate the correct
1976 operation of lsof. The library can be found in the tests/LTlib.c file
1977 of the lsof distribution. The library uses the first aid, the
1978 lsof_fields.h header file.
1979
1981 Lsof can be blocked by some kernel functions that it uses - lstat(2),
1982 readlink(2), and stat(2). These functions are stalled in the kernel,
1983 for example, when the hosts where mounted NFS file systems reside
1984 become inaccessible.
1985 Lsof attempts to break these blocks with timers and child processes,
1987 but the techniques are not wholly reliable. When lsof does manage to
1988 break a block, it will report the break with an error message. The
1989 messages may be suppressed with the -t and -w options.
1990 The default timeout value may be displayed with the -h or -? option,
1992 and it may be changed with the -S [t] option. The minimum for t is two
1993 seconds, but you should avoid small values, since slow system respon‐
1994 siveness can cause short timeouts to expire unexpectedly and perhaps
1995 stop lsof before it can produce any output.
1996 When lsof has to break a block during its access of mounted file system
1998 information, it normally continues, although with less information
1999 available to display about open files.
2000 Lsof can also be directed to avoid the protection of timers and child
2002 processes when using the kernel functions that might block by specify‐
2003 ing the -O option. While this will allow lsof to start up with less
2004 overhead, it exposes lsof completely to the kernel situations that
2005 might block it. Use this option cautiously.
2006
2008 You can use the -b option to tell lsof to avoid using kernel functions
2009 that would block. Some cautions apply.
2010 First, using this option usually requires that your system supply
2012 alternate device numbers in place of the device numbers that lsof would
2013 normally obtain with the lstat(2) and stat(2) kernel functions. See
2014 the ALTERNATE DEVICE NUMBERS section for more information on alternate
2015 device numbers.
2016 Second, you can't specify names for lsof to locate unless they're file
2018 system names. This is because lsof needs to know the device and inode
2019 numbers of files listed with names in the lsof options, and the -b
2020 option prevents lsof from obtaining them. Moreover, since lsof only
2021 has device numbers for the file systems that have alternates, its abil‐
2022 ity to locate files on file systems depends completely on the avail‐
2023 ability and accuracy of the alternates. If no alternates are avail‐
2024 able, or if they're incorrect, lsof won't be able to locate files on
2025 the named file systems.
2026 Third, if the names of your file system directories that lsof obtains
2028 from your system's mount table are symbolic links, lsof won't be able
2029 to resolve the links. This is because the -b option causes lsof to
2030 avoid the kernel readlink(2) function it uses to resolve symbolic
2031 links.
2032 Finally, using the -b option causes lsof to issue warning messages when
2034 it needs to use the kernel functions that the -b option directs it to
2035 avoid. You can suppress these messages by specifying the -w option,
2036 but if you do, you won't see the alternate device numbers reported in
2037 the warning messages.
2038
2040 On some dialects, when lsof has to break a block because it can't get
2041 information about a mounted file system via the lstat(2) and stat(2)
2042 kernel functions, or because you specified the -b option, lsof can
2043 obtain some of the information it needs - the device number and possi‐
2044 bly the file system type - from the system mount table. When that is
2045 possible, lsof will report the device number it obtained. (You can
2046 suppress the report by specifying the -w option.)
2047 You can assist this process if your mount table is supported with an
2049 /etc/mtab or /etc/mnttab file that contains an options field by adding
2050 a ``dev=xxxx'' field for mount points that do not have one in their
2051 options strings. Note: you must be able to edit the file - i.e., some
2052 mount tables like recent Solaris /etc/mnttab or Linux /proc/mounts are
2053 read-only and can't be modified.
2054 You may also be able to supply device numbers using the +m and +m m
2056 options, provided they are supported by your dialect. Check the output
2057 of lsof's -h or -? options to see if the +m and +m m options are
2058 available.
2059 The ``xxxx'' portion of the field is the hexadecimal value of the file
2061 system's device number. (Consult the st_dev field of the output of the
2062 lstat(2) and stat(2) functions for the appropriate values for your file
2063 systems.) Here's an example from a Sun Solaris 2.6 /etc/mnttab for a
2064 file system remotely mounted via NFS:
2065 nfs ignore,noquota,dev=2a40001
2067 There's an advantage to having ``dev=xxxx'' entries in your mount table
2069 file, especially for file systems that are mounted from remote NFS
2070 servers. When a remote server crashes and you want to identify its
2071 users by running lsof on one of its clients, lsof probably won't be
2072 able to get output from the lstat(2) and stat(2) functions for the file
2073 system. If it can obtain the file system's device number from the
2074 mount table, it will be able to display the files open on the crashed
2075 NFS server.
2076 Some dialects that do not use an ASCII /etc/mtab or /etc/mnttab file
2078 for the mount table may still provide an alternative device number in
2079 their internal mount tables. This includes AIX, Apple Darwin, FreeBSD,
2080 NetBSD, OpenBSD, and Tru64 UNIX. Lsof knows how to obtain the alterna‐
2081 tive device number for these dialects and uses it when its attempt to
2082 lstat(2) or stat(2) the file system is blocked.
2083 If you're not sure your dialect supplies alternate device numbers for
2085 file systems from its mount table, use this lsof incantation to see if
2086 it reports any alternate device numbers:
2087 lsof -b
2089 Look for standard error file warning messages that begin ``assuming
2091 "dev=xxxx" from ...''.
2092
2094 Lsof is able to examine the kernel's name cache or use other kernel
2095 facilities (e.g., the ADVFS 4.x tag_to_path() function under Tru64
2096 UNIX) on some dialects for most file system types, excluding AFS, and
2097 extract recently used path name components from it. (AFS file system
2098 path lookups don't use the kernel's name cache; some Solaris VxFS file
2099 system operations apparently don't use it, either.)
2100 Lsof reports the complete paths it finds in the NAME column. If lsof
2102 can't report all components in a path, it reports in the NAME column
2103 the file system name, followed by a space, two `-' characters, another
2104 space, and the name components it has located, separated by the `/'
2105 character.
2106 When lsof is run in repeat mode - i.e., with the -r option specified -
2108 the extent to which it can report path name components for the same
2109 file may vary from cycle to cycle. That's because other running pro‐
2110 cesses can cause the kernel to remove entries from its name cache and
2111 replace them with others.
2112 Lsof's use of the kernel name cache to identify the paths of files can
2114 lead it to report incorrect components under some circumstances. This
2115 can happen when the kernel name cache uses device and node number as a
2116 key (e.g., SCO OpenServer) and a key on a rapidly changing file system
2117 is reused. If the UNIX dialect's kernel doesn't purge the name cache
2118 entry for a file when it is unlinked, lsof may find a reference to the
2119 wrong entry in the cache. The lsof FAQ (The FAQ section gives its
2120 location.) has more information on this situation.
2121 Lsof can report path name components for these dialects:
2123 FreeBSD
2125 HP-UX
2126 Linux
2127 NetBSD
2128 NEXTSTEP
2129 OpenBSD
2130 OPENSTEP
2131 SCO OpenServer
2132 SCO|Caldera UnixWare
2133 Solaris
2134 Tru64 UNIX
2135 Lsof can't report path name components for these dialects:
2137 AIX
2139 If you want to know why lsof can't report path name components for some
2141 dialects, see the lsof FAQ (The FAQ section gives its location.)
2142
2144 Examining all members of the /dev (or /devices) node tree with stat(2)
2145 functions can be time consuming. What's more, the information that
2146 lsof needs - device number, inode number, and path - rarely changes.
2147 Consequently, lsof normally maintains an ASCII text file of cached /dev
2149 (or /devices) information (exception: the /proc-based Linux lsof where
2150 it's not needed.) The local system administrator who builds lsof can
2151 control the way the device cache file path is formed, selecting from
2152 these options:
2153 Path from the -D option;
2155 Path from an environment variable;
2156 System-wide path;
2157 Personal path (the default);
2158 Personal path, modified by an environment variable.
2159 Consult the output of the -h, -D? , or -? help options for the current
2161 state of device cache support. The help output lists the default
2162 read-mode device cache file path that is in effect for the current
2163 invocation of lsof. The -D? option output lists the read-only and
2164 write device cache file paths, the names of any applicable environment
2165 variables, and the personal device cache path format.
2166 Lsof can detect that the current device cache file has been acciden‐
2168 tally or maliciously modified by integrity checks, including the compu‐
2169 tation and verification of a sixteen bit Cyclic Redundancy Check (CRC)
2170 sum on the file's contents. When lsof senses something wrong with the
2171 file, it issues a warning and attempts to remove the current cache file
2172 and create a new copy, but only to a path that the process can legiti‐
2173 mately write.
2174 The path from which a lsof process may attempt to read a device cache
2176 file may not be the same as the path to which it can legitimately
2177 write. Thus when lsof senses that it needs to update the device cache
2178 file, it may choose a different path for writing it from the path from
2179 which it read an incorrect or outdated version.
2180 If available, the -Dr option will inhibit the writing of a new device
2182 cache file. (It's always available when specified without a path name
2183 argument.)
2184 When a new device is added to the system, the device cache file may
2186 need to be recreated. Since lsof compares the mtime of the device
2187 cache file with the mtime and ctime of the /dev (or /devices) direc‐
2188 tory, it usually detects that a new device has been added; in that case
2189 lsof issues a warning message and attempts to rebuild the device cache
2190 file.
2191 Whenever lsof writes a device cache file, it sets its ownership to the
2193 real UID of the executing process, and its permission modes to 0600,
2194 this restricting its reading and writing to the file's owner.
2195
2197 Two permissions of the lsof executable affect its ability to access
2198 device cache files. The permissions are set by the local system admin‐
2199 istrator when lsof is installed.
2200 The first and rarer permission is setuid-root. It comes into effect
2202 when lsof is executed; its effective UID is then root, while its real
2203 (i.e., that of the logged-on user) UID is not. The lsof distribution
2204 recommends that versions for these dialects run setuid-root.
2205 HP-UX 11.11 and 11.23
2207 Linux
2208 The second and more common permission is setgid. It comes into effect
2210 when the effective group IDentification number (GID) of the lsof
2211 process is set to one that can access kernel memory devices - e.g.,
2212 ``kmem'', ``sys'', or ``system''.
2213 An lsof process that has setgid permission usually surrenders the per‐
2215 mission after it has accessed the kernel memory devices. When it does
2216 that, lsof can allow more liberal device cache path formations. The
2217 lsof distribution recommends that versions for these dialects run set‐
2218 gid and be allowed to surrender setgid permission.
2219 AIX 5.[12] and 5.3-ML1
2221 Apple Darwin 7.x Power Macintosh systems
2222 FreeBSD 4.x, 4.1x, 5.x and [6789].x for x86-based systems
2223 FreeBSD 5.x and [6789].x for Alpha, AMD64 and Sparc64-based
2224 systems
2225 HP-UX 11.00
2226 NetBSD 1.[456], 2.x and 3.x for Alpha, x86, and SPARC-based
2227 systems
2228 NEXTSTEP 3.[13] for NEXTSTEP architectures
2229 OpenBSD 2.[89] and 3.[0-9] for x86-based systems
2230 OPENSTEP 4.x
2231 SCO OpenServer Release 5.0.6 for x86-based systems
2232 SCO|Caldera UnixWare 7.1.4 for x86-based systems
2233 Solaris 2.6, 8, 9 and 10
2234 Tru64 UNIX 5.1
2235 (Note: lsof for AIX 5L and above needs setuid-root permission if its -X
2237 option is used.)
2238 Lsof for these dialects does not support a device cache, so the permis‐
2240 sions given to the executable don't apply to the device cache file.
2241 Linux
2243
2245 The -D option provides limited means for specifying the device cache
2246 file path. Its ? function will report the read-only and write device
2247 cache file paths that lsof will use.
2248 When the -D b, r, and u functions are available, you can use them to
2250 request that the cache file be built in a specific location (b[path]);
2251 read but not rebuilt (r[path]); or read and rebuilt (u[path]). The b,
2252 r, and u functions are restricted under some conditions. They are
2253 restricted when the lsof process is setuid-root. The path specified
2254 with the r function is always read-only, even when it is available.
2255 The b, r, and u functions are also restricted when the lsof process
2257 runs setgid and lsof doesn't surrender the setgid permission. (See the
2258 LSOF PERMISSIONS THAT AFFECT DEVICE CACHE FILE ACCESS section for a
2259 list of implementations that normally don't surrender their setgid per‐
2260 mission.)
2261 A further -D function, i (for ignore), is always available.
2263 When available, the b function tells lsof to read device information
2265 from the kernel with the stat(2) function and build a device cache file
2266 at the indicated path.
2267 When available, the r function tells lsof to read the device cache
2269 file, but not update it. When a path argument accompanies -Dr, it
2270 names the device cache file path. The r function is always available
2271 when it is specified without a path name argument. If lsof is not run‐
2272 ning setuid-root and surrenders its setgid permission, a path name
2273 argument may accompany the r function.
2274 When available, the u function tells lsof to attempt to read and use
2276 the device cache file. If it can't read the file, or if it finds the
2277 contents of the file incorrect or outdated, it will read information
2278 from the kernel, and attempt to write an updated version of the device
2279 cache file, but only to a path it considers legitimate for the lsof
2280 process effective and real UIDs.
2281
2283 Lsof's second choice for the device cache file is the contents of the
2284 LSOFDEVCACHE environment variable. It avoids this choice if the lsof
2285 process is setuid-root, or the real UID of the process is root.
2286 A further restriction applies to a device cache file path taken from
2288 the LSOFDEVCACHE environment variable: lsof will not write a device
2289 cache file to the path if the lsof process doesn't surrender its setgid
2290 permission. (See the LSOF PERMISSIONS THAT AFFECT DEVICE CACHE FILE
2291 ACCESS section for information on implementations that don't surrender
2292 their setgid permission.)
2293 The local system administrator can disable the use of the LSOFDEVCACHE
2295 environment variable or change its name when building lsof. Consult
2296 the output of -D? for the environment variable's name.
2297
2299 The local system administrator may choose to have a system-wide device
2300 cache file when building lsof. That file will generally be constructed
2301 by a special system administration procedure when the system is booted
2302 or when the contents of /dev or /devices) changes. If defined, it is
2303 lsof's third device cache file path choice.
2304 You can tell that a system-wide device cache file is in effect for your
2306 local installation by examining the lsof help option output - i.e., the
2307 output from the -h or -? option.
2308 Lsof will never write to the system-wide device cache file path by
2310 default. It must be explicitly named with a -D function in a
2311 root-owned procedure. Once the file has been written, the procedure
2312 must change its permission modes to 0644 (owner-read and owner-write,
2313 group-read, and other-read).
2314
2316 The default device cache file path of the lsof distribution is one
2317 recorded in the home directory of the real UID that executes lsof.
2318 Added to the home directory is a second path component of the form
2319 .lsof_hostname.
2320 This is lsof's fourth device cache file path choice, and is usually the
2322 default. If a system-wide device cache file path was defined when lsof
2323 was built, this fourth choice will be applied when lsof can't find the
2324 system-wide device cache file. This is the only time lsof uses two
2325 paths when reading the device cache file.
2326 The hostname part of the second component is the base name of the exe‐
2328 cuting host, as returned by gethostname(2). The base name is defined
2329 to be the characters preceding the first `.' in the gethostname(2)
2330 output, or all the gethostname(2) output if it contains no `.'.
2331 The device cache file belongs to the user ID and is readable and
2333 writable by the user ID alone - i.e., its modes are 0600. Each dis‐
2334 tinct real user ID on a given host that executes lsof has a distinct
2335 device cache file. The hostname part of the path distinguishes device
2336 cache files in an NFS-mounted home directory into which device cache
2337 files are written from several different hosts.
2338 The personal device cache file path formed by this method represents a
2340 device cache file that lsof will attempt to read, and will attempt to
2341 write should it not exist or should its contents be incorrect or out‐
2342 dated.
2343 The -Dr option without a path name argument will inhibit the writing of
2345 a new device cache file.
2346 The -D? option will list the format specification for constructing the
2348 personal device cache file. The conversions used in the format speci‐
2349 fication are described in the 00DCACHE file of the lsof distribution.
2350
2352 If this option is defined by the local system administrator when lsof
2353 is built, the LSOFPERSDCPATH environment variable contents may be used
2354 to add a component of the personal device cache file path.
2355 The LSOFPERSDCPATH variable contents are inserted in the path at the
2357 place marked by the local system administrator with the ``%p'' conver‐
2358 sion in the HASPERSDC format specification of the dialect's machine.h
2359 header file. (It's placed right after the home directory in the
2360 default lsof distribution.)
2361 Thus, for example, if LSOFPERSDCPATH contains ``LSOF'', the home direc‐
2363 tory is ``/Homes/abe'', the host name is ``lsof.itap.purdue.edu'', and
2364 the HASPERSDC format is the default (``%h/%p.lsof_%L''), the modified
2365 personal device cache file path is:
2366 /Homes/abe/LSOF/.lsof_vic
2368 The LSOFPERSDCPATH environment variable is ignored when the lsof
2370 process is setuid-root or when the real UID of the process is root.
2371 Lsof will not write to a modified personal device cache file path if
2373 the lsof process doesn't surrender setgid permission. (See the LSOF
2374 PERMISSIONS THAT AFFECT DEVICE CACHE FILE ACCESS section for a list of
2375 implementations that normally don't surrender their setgid permission.)
2376 If, for example, you want to create a sub-directory of personal device
2378 cache file paths by using the LSOFPERSDCPATH environment variable to
2379 name it, and lsof doesn't surrender its setgid permission, you will
2380 have to allow lsof to create device cache files at the standard per‐
2381 sonal path and move them to your subdirectory with shell commands.
2382 The local system administrator may: disable this option when lsof is
2384 built; change the name of the environment variable from LSOFPERSDCPATH
2385 to something else; change the HASPERSDC format to include the personal
2386 path component in another place; or exclude the personal path component
2387 entirely. Consult the output of the -D? option for the environment
2388 variable's name and the HASPERSDC format specification.
2389
2391 Errors are identified with messages on the standard error file.
2392 Lsof returns a one (1) if any error was detected, including the failure
2394 to locate command names, file names, Internet addresses or files, login
2395 names, NFS files, PIDs, PGIDs, or UIDs it was asked to list. If the -V
2396 option is specified, lsof will indicate the search items it failed to
2397 list.
2398 It returns a zero (0) if no errors were detected and if it was able to
2400 list some information about all the specified search arguments.
2401 When lsof cannot open access to /dev (or /devices) or one of its subdi‐
2403 rectories, or get information on a file in them with stat(2), it issues
2404 a warning message and continues. That lsof will issue warning messages
2405 about inaccessible files in /dev (or /devices) is indicated in its help
2406 output - requested with the -h or >B -? options - with the message:
2407 Inaccessible /dev warnings are enabled.
2409 The warning message may be suppressed with the -w option. It may also
2411 have been suppressed by the system administrator when lsof was compiled
2412 by the setting of the WARNDEVACCESS definition. In this case, the out‐
2413 put from the help options will include the message:
2414 Inaccessible /dev warnings are disabled.
2416 Inaccessible device warning messages usually disappear after lsof has
2418 created a working device cache file.
2419
2421 For a more extensive set of examples, documented more fully, see the
2422 00QUICKSTART file of the lsof distribution.
2423 To list all open files, use:
2425 lsof
2427 To list all open Internet, x.25 (HP-UX), and UNIX domain files, use:
2429 lsof -i -U
2431 To list all open IPv4 network files in use by the process whose PID is
2433 1234, use:
2434 lsof -i 4 -a -p 1234
2436 Presuming the UNIX dialect supports IPv6, to list only open IPv6 net‐
2438 work files, use:
2439 lsof -i 6
2441 To list all files using any protocol on ports 513, 514, or 515 of host
2443 wonderland.cc.purdue.edu, use:
2444 lsof -i @wonderland.cc.purdue.edu:513-515
2446 To list all files using any protocol on any port of mace.cc.purdue.edu
2448 (cc.purdue.edu is the default domain), use:
2449 lsof -i @mace
2451 To list all open files for login name ``abe'', or user ID 1234, or
2453 process 456, or process 123, or process 789, use:
2454 lsof -p 456,123,789 -u 1234,abe
2456 To list all open files on device /dev/hd4, use:
2458 lsof /dev/hd4
2460 To find the process that has /u/abe/foo open, use:
2462 lsof /u/abe/foo
2464 To send a SIGHUP to the processes that have /u/abe/bar open, use:
2466 kill -HUP `lsof -t /u/abe/bar`
2468 To find any open file, including an open UNIX domain socket file, with
2470 the name /dev/log, use:
2471 lsof /dev/log
2473 To find processes with open files on the NFS file system named
2475 /nfs/mount/point whose server is inaccessible, and presuming your mount
2476 table supplies the device number for /nfs/mount/point, use:
2477 lsof -b /nfs/mount/point
2479 To do the preceding search with warning messages suppressed, use:
2481 lsof -bw /nfs/mount/point
2483 To ignore the device cache file, use:
2485 lsof -Di
2487 To obtain PID and command name field output for each process, file
2489 descriptor, file device number, and file inode number for each file of
2490 each process, use:
2491 lsof -FpcfDi
2493 To list the files at descriptors 1 and 3 of every process running the
2495 lsof command for login ID ``abe'' every 10 seconds, use:
2496 lsof -c lsof -a -d 1 -d 3 -u abe -r10
2498 To list the current working directory of processes running a command
2500 that is exactly four characters long and has an 'o' or 'O' in character
2501 three, use this regular expression form of the -c c option:
2502 lsof -c /^..o.$/i -a -d cwd
2504 To find an IP version 4 socket file by its associated numeric dot-form
2506 address, use:
2507 lsof -i@128.210.15.17
2509 To find an IP version 6 socket file (when the UNIX dialect supports
2511 IPv6) by its associated numeric colon-form address, use:
2512 lsof -i@[0:1:2:3:4:5:6:7]
2514 To find an IP version 6 socket file (when the UNIX dialect supports
2516 IPv6) by an associated numeric colon-form address that has a run of
2517 zeroes in it - e.g., the loop-back address - use:
2518 lsof -i@[::1]
2520 To obtain a repeat mode marker line that contains the current time,
2522 use:
2523 lsof -rm====%T====
2525 To add spaces to the previous marker line, use:
2527 lsof -r "m==== %T ===="
2529
2531 Since lsof reads kernel memory in its search for open files, rapid
2532 changes in kernel memory may produce unpredictable results.
2533 When a file has multiple record locks, the lock status character (fol‐
2535 lowing the file descriptor) is derived from a test of the first lock
2536 structure, not from any combination of the individual record locks that
2537 might be described by multiple lock structures.
2538 Lsof can't search for files with restrictive access permissions by name
2540 unless it is installed with root set-UID permission. Otherwise it is
2541 limited to searching for files to which its user or its set-GID group
2542 (if any) has access permission.
2543 The display of the destination address of a raw socket (e.g., for ping)
2545 depends on the UNIX operating system. Some dialects store the destina‐
2546 tion address in the raw socket's protocol control block, some do not.
2547 Lsof can't always represent Solaris device numbers in the same way that
2549 ls(1) does. For example, the major and minor device numbers that the
2550 lstat(2) and stat(2) functions report for the directory on which CD-ROM
2551 files are mounted (typically /cdrom) are not the same as the ones that
2552 it reports for the device on which CD-ROM files are mounted (typically
2553 /dev/sr0). (Lsof reports the directory numbers.)
2554 The support for /proc file systems is available only for BSD and Tru64
2556 UNIX dialects, Linux, and dialects derived from SYSV R4 - e.g., Free‐
2557 BSD, NetBSD, OpenBSD, Solaris, UnixWare.
2558 Some /proc file items - device number, inode number, and file size -
2560 are unavailable in some dialects. Searching for files in a /proc file
2561 system may require that the full path name be specified.
2562 No text (txt) file descriptors are displayed for Linux processes. All
2564 entries for files other than the current working directory, the root
2565 directory, and numerical file descriptors are labeled mem descriptors.
2566 Lsof can't search for Tru64 UNIX named pipes by name, because their
2568 kernel implementation of lstat(2) returns an improper device number for
2569 a named pipe.
2570 Lsof can't report fully or correctly on HP-UX 9.01, 10.20, and 11.00
2572 locks because of insufficient access to kernel data or errors in the
2573 kernel data. See the lsof FAQ (The FAQ section gives its location.)
2574 for details.
2575 The AIX SMT file type is a fabrication. It's made up for file struc‐
2577 tures whose type (15) isn't defined in the AIX /usr/include/sys/file.h
2578 header file. One way to create such file structures is to run X
2579 clients with the DISPLAY variable set to ``:0.0''.
2580 The +|-f[cfgGn] option is not supported under /proc-based Linux lsof,
2582 because it doesn't read kernel structures from kernel memory.
2583
2585 Lsof may access these environment variables.
2586 LANG defines a language locale. See setlocale(3) for the
2588 names of other variables that can be used in place of
2589 LANG - e.g., LC_ALL, LC_TYPE, etc.
2590 LSOFDEVCACHE defines the path to a device cache file. See the
2592 DEVICE CACHE PATH FROM AN ENVIRONMENT VARIABLE sec‐
2593 tion for more information.
2594 LSOFPERSDCPATH defines the middle component of a modified personal
2596 device cache file path. See the MODIFIED PERSONAL
2597 DEVICE CACHE PATH section for more information.
2598
2600 Frequently-asked questions and their answers (an FAQ) are available in
2601 the 00FAQ file of the lsof distribution.
2602 That file is also available via anonymous ftp from lsof.itap.purdue.edu
2604 at pub/tools/unix/lsofFAQ. The URL is:
2605 ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof/FAQ
2607
2609 /dev/kmem kernel virtual memory device
2610 /dev/mem physical memory device
2612 /dev/swap system paging device
2614 .lsof_hostname lsof's device cache file (The suffix, hostname, is
2616 the first component of the host's name returned by
2617 gethostname(2).)
2618
2620 Lsof was written by Victor A. Abell <abe@purdue.edu> of Purdue Univer‐
2621 sity. Many others have contributed to lsof. They're listed in the
2622 00CREDITS file of the lsof distribution.
2623
2625 The latest distribution of lsof is available via anonymous ftp from the
2626 host lsof.itap.purdue.edu. You'll find the lsof distribution in the
2627 pub/tools/unix/lsof directory.
2628 You can also use this URL:
2630 ftp://lsof.itap.purdue.edu/pub/tools/unix/lsof
2632 Lsof is also mirrored elsewhere. When you access lsof.itap.purdue.edu
2634 and change to its pub/tools/unix/lsof directory, you'll be given a list
2635 of some mirror sites. The pub/tools/unix/lsof directory also contains
2636 a more complete list in its mirrors file. Use mirrors with caution -
2637 not all mirrors always have the latest lsof revision.
2638 Some pre-compiled Lsof executables are available on lsof.itap.pur‐
2640 due.edu, but their use is discouraged - it's better that you build your
2641 own from the sources. If you feel you must use a pre-compiled exe‐
2642 cutable, please read the cautions that appear in the README files of
2643 the pub/tools/unix/lsof/binaries subdirectories and in the 00* files of
2644 the distribution.
2645 More information on the lsof distribution can be found in its
2647 README.lsof_<version> file. If you intend to get the lsof distribution
2648 and build it, please read README.lsof_<version> and the other 00* files
2649 of the distribution before sending questions to the author.
2650
2652 Not all the following manual pages may exist in every UNIX dialect to
2653 which lsof has been ported.
2654 access(2), awk(1), crash(1), fattach(3C), ff(1), fstat(8), fuser(1),
2656 gethostname(2), isprint(3), kill(1), localtime(3), lstat(2), mod‐
2657 load(8), mount(8), netstat(1), ofiles(8L), perl(1), ps(1), readlink(2),
2658 setlocale(3), stat(2), strftime(3), time(2), uname(1).
2659 Revision-4.83 LSOF(8)