09c77d1d3a99feee543c4e96fc7a5555423d0d35
[lttng-tools.git] / doc / man / lttng.1
1 .TH "LTTNG" "1" "May 13th, 2014" "" ""
2
3 .SH "NAME"
4 lttng \(em LTTng 2.x tracer control command line tool
5
6 .SH "SYNOPSIS"
7
8 .PP
9 lttng [OPTIONS] <COMMAND>
10 .SH "DESCRIPTION"
11
12 .PP
13 The LTTng project aims at providing highly efficient tracing tools for Linux.
14 Its tracers help track down performance issues and debug problems
15 involving multiple concurrent processes and threads. Tracing across multiple
16 systems is also possible.
17
18 The \fBlttng\fP command line tool from the lttng-tools package is used to control
19 both kernel and user-space tracing. Every interaction with the tracer should
20 be done by this tool or by the liblttng-ctl library provided by the lttng-tools
21 package.
22
23 LTTng uses a session daemon (lttng-sessiond(8)), acting as a tracing registry,
24 which allows you to interact with multiple tracers (kernel and user-space)
25 inside the same container, a tracing session. Traces can be gathered from the
26 kernel and/or instrumented applications (lttng-ust(3)). Aggregating and reading
27 those traces is done using the babeltrace(1) text viewer.
28
29 We introduce the notion of \fBtracing domains\fP which is essentially a type of
30 tracer (kernel, user space or JUL for now). In the future, we could see more
31 tracer like for instance an hypervisor. For some commands, you'll need to
32 specify on which domain the command operates (\-u, \-k or \-j). For instance,
33 the kernel domain must be specified when enabling a kernel event.
34
35 In order to trace the kernel, the session daemon needs to be running as root.
36 LTTng provides the use of a \fBtracing group\fP (default: tracing). Whomever is
37 in that group can interact with the root session daemon and thus trace the
38 kernel. Session daemons can co-exist, meaning that you can have a session daemon
39 running as Alice that can be used to trace her applications along side with a
40 root daemon or even a Bob daemon. We highly recommend starting the session
41 daemon at boot time for stable and long term tracing.
42
43 Each user-space application instrumented with lttng-ust(3) will automatically
44 register with the root session daemon and its user session daemon. This allows
45 each daemon to list the available traceable applications and tracepoints at any
46 given moment (See the \fBlist\fP command).
47 .SH "OPTIONS"
48
49 .PP
50 This program follow the usual GNU command line syntax with long options starting with
51 two dashes. Below is a summary of the available options.
52 .PP
53
54 .TP
55 .BR "\-h, \-\-help"
56 Show summary of possible options and commands.
57 .TP
58 .BR "\-v, \-\-verbose"
59 Increase verbosity.
60 Three levels of verbosity are available which are triggered by putting additional v to
61 the option (\-vv or \-vvv)
62 .TP
63 .BR "\-q, \-\-quiet"
64 Suppress all messages (even errors).
65 .TP
66 .BR "\-g, \-\-group NAME"
67 Set unix tracing group name. (default: tracing)
68 .TP
69 .BR "\-n, \-\-no-sessiond"
70 Don't automatically spawn a session daemon.
71 .TP
72 .BR "\-\-sessiond\-path PATH"
73 Set session daemon full binary path.
74 .TP
75 .BR "\-\-list\-options"
76 Simple listing of lttng options.
77 .TP
78 .BR "\-\-list\-commands"
79 Simple listing of lttng commands.
80 .SH "COMMANDS"
81
82 .PP
83 \fBadd-context\fP [OPTIONS]
84 .RS
85 Add context to event(s) and/or channel(s).
86
87 A context is basically extra information appended to a channel. For instance,
88 you could ask the tracer to add the PID information for all events in a
89 channel. You can also add performance monitoring unit counters (perf PMU) using
90 the perf kernel API.
91
92 For example, this command will add the context information 'prio' and two per-CPU
93 perf counters (hardware branch misses and cache misses), to all events in the trace
94 data output:
95
96 .nf
97 # lttng add-context \-k \-t prio \-t perf:cpu:branch-misses \\
98 \-t perf:cpu:cache-misses
99 .fi
100
101 Please take a look at the help (\-h/\-\-help) for a detailed list of available
102 contexts.
103
104 Perf counters are available as per-CPU ("perf:cpu:...") and per-thread
105 ("perf:thread:...") counters. Currently, per-CPU counters can only be
106 used with the kernel tracing domain, and per-thread counters can only be
107 used with the UST tracing domain.
108
109 If no channel is given (\-c), the context is added to all channels that were
110 already enabled. If the session has no channel, a default channel is created.
111 Otherwise the context will be added only to the given channel (\-c).
112
113 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
114 file.
115
116 .B OPTIONS:
117
118 .TP
119 .BR "\-h, \-\-help"
120 Show summary of possible options and commands.
121 .TP
122 .BR "\-s, \-\-session NAME"
123 Apply on session name.
124 .TP
125 .BR "\-c, \-\-channel NAME"
126 Apply on channel name.
127 .TP
128 .BR "\-k, \-\-kernel"
129 Apply for the kernel tracer
130 .TP
131 .BR "\-u, \-\-userspace"
132 Apply for the user-space tracer
133 .TP
134 .BR "\-t, \-\-type TYPE"
135 Context type. You can repeat this option on the command line. Please
136 use "lttng add-context \-h" to list all available types.
137 .RE
138 .PP
139
140 .PP
141 \fBcalibrate\fP [OPTIONS]
142 .RS
143 Quantify LTTng overhead
144
145 The LTTng calibrate command can be used to find out the combined average
146 overhead of the LTTng tracer and the instrumentation mechanisms used. This
147 overhead can be calibrated in terms of time or using any of the PMU performance
148 counter available on the system.
149
150 For now, the only calibration implemented is that of the kernel function
151 instrumentation (kretprobes).
152
153 * Calibrate kernel function instrumentation
154
155 Let's use an example to show this calibration. We use an i7 processor with 4
156 general-purpose PMU registers. This information is available by issuing dmesg,
157 looking for "generic registers".
158
159 This sequence of commands will gather a trace executing a kretprobe hooked on
160 an empty function, gathering PMU counters LLC (Last Level Cache) misses
161 information (see lttng add-context \-\-help to see the list of available PMU
162 counters).
163
164 .nf
165 # lttng create calibrate-function
166 # lttng enable-event calibrate \-\-kernel \\
167 \-\-function lttng_calibrate_kretprobe
168 # lttng add-context \-\-kernel \-t perf:cpu:LLC-load-misses \\
169 \-t perf:cpu:LLC-store-misses \\
170 \-t perf:cpu:LLC-prefetch-misses
171 # lttng start
172 # for a in $(seq 1 10); do \\
173 lttng calibrate \-\-kernel \-\-function;
174 done
175 # lttng destroy
176 # babeltrace $(ls \-1drt ~/lttng-traces/calibrate-function-* \\
177 | tail \-n 1)
178 .fi
179
180 The output from babeltrace can be saved to a text file and opened in a
181 spreadsheet (e.g. oocalc) to focus on the per-PMU counter delta between
182 consecutive "calibrate_entry" and "calibrate_return" events. Note that these
183 counters are per-CPU, so scheduling events would need to be present to account
184 for migration between CPU. Therefore, for calibration purposes, only events
185 staying on the same CPU must be considered.
186
187 The average result, for the i7, on 10 samples:
188
189 .nf
190 Average Std.Dev.
191 perf_LLC_load_misses: 5.0 0.577
192 perf_LLC_store_misses: 1.6 0.516
193 perf_LLC_prefetch_misses: 9.0 14.742
194 .fi
195
196 As we can notice, the load and store misses are relatively stable across runs
197 (their standard deviation is relatively low) compared to the prefetch misses.
198 We can conclude from this information that LLC load and store misses can be
199 accounted for quite precisely, but prefetches within a function seems to behave
200 too erratically (not much causality link between the code executed and the CPU
201 prefetch activity) to be accounted for.
202
203 .B OPTIONS:
204
205 .TP
206 .BR "\-h, \-\-help"
207 Show summary of possible options and commands.
208 .TP
209 .BR "\-k, \-\-kernel"
210 Apply for the kernel tracer
211 .TP
212 .BR "\-u, \-\-userspace"
213 Apply for the user-space tracer
214 .TP
215 .BR "\-\-function"
216 Dynamic function entry/return probe (default)
217 .RE
218 .PP
219
220 .PP
221 \fBcreate\fP [NAME] [OPTIONS]
222 .RS
223 Create tracing session.
224
225 A tracing session contains channel(s) which contains event(s). It is domain
226 agnostic, meaning that channels and events can be enabled for the
227 user-space tracer and/or the kernel tracer. It acts like a container
228 aggregating multiple tracing sources.
229
230 On creation, a \fB.lttngrc\fP file is created in your $HOME directory
231 containing the current session name. If NAME is omitted, a session name is
232 automatically created having this form: 'auto-yyyymmdd-hhmmss'.
233
234 If no \fB\-o, \-\-output\fP is specified, the traces will be written in
235 $HOME/lttng-traces.
236
237 The $HOME environment variable can be overridden by defining the environment
238 variable LTTNG_HOME. This is useful when the user running the commands has
239 a non-writeable home directory.
240
241 The session name MUST NOT contain the character '/'.
242
243 .B OPTIONS:
244
245 .TP
246 .BR "\-h, \-\-help"
247 Show summary of possible options and commands.
248 .TP
249 .BR "\-\-list-options"
250 Simple listing of options
251 .TP
252 .BR "\-o, \-\-output PATH"
253 Specify output path for traces
254 .TP
255 .BR "\-\-no-output"
256 Traces will not be output
257 .TP
258 .BR "\-\-snapshot"
259 Set the session in snapshot mode. Created in no-output mode and uses the
260 URL, if one is specified, as the default snapshot output. Every channel will be set
261 in overwrite mode and with mmap output (splice not supported).
262 .TP
263 .BR "\-\-live [USEC]"
264 Set the session exclusively in live mode. The paremeter is the delay in micro
265 seconds before the data is flushed and streamed. The live mode allows you to
266 stream the trace and view it while it's being recorded by any tracer. For that,
267 you need a lttng-relayd and this session requires a network URL (\-U or
268 \-C/\-D). If no USEC nor URL is provided, the default is to use a timer value
269 set to 1000000 and the network URL set to net://127.0.0.1.
270
271 To read a live session, you can use babeltrace(1) or the live streaming
272 protocol in doc/live-reading-protocol.txt. Here is an example:
273
274 .nf
275 $ lttng-relayd -o /tmp/lttng
276 $ lttng create --live 200000 -U net://localhost
277 $ lttng enable-event -a --userspace
278 $ lttng start
279 .fi
280
281 After the start, you'll be able to read the events while they are being
282 recorded in /tmp/lttng.
283
284 .TP
285 .BR "\-U, \-\-set-url=URL"
286 Set URL for the consumer output destination. It is persistent for the
287 session lifetime. Redo the command to change it. This will set both data
288 and control URL for network.
289 .TP
290 .BR "\-C, \-\-ctrl-url=URL"
291 Set control path URL. (Must use -D also)
292 .TP
293 .BR "\-D, \-\-data-url=URL"
294 Set data path URL. (Must use -C also)
295 .PP
296 Using these options, each API call can be controlled individually. For
297 instance, \-C does not enable the consumer automatically. You'll need the \-e
298 option for that.
299
300 .B URL FORMAT:
301
302 proto://[HOST|IP][:PORT1[:PORT2]][/TRACE_PATH]
303
304 Supported protocols are (proto):
305 .TP
306 .BR "file://..."
307 Local filesystem full path.
308
309 .TP
310 .BR "net://..."
311 This will use the default network transport layer which is TCP for both
312 control (PORT1) and data port (PORT2). The default ports are
313 respectively 5342 and 5343. Note that net[6]:// is not yet supported.
314
315 .TP
316 .BR "tcp[6]://..."
317 Can only be used with -C and -D together
318
319 NOTE: IPv6 address MUST be enclosed in brackets '[]' (rfc2732)
320
321 .B EXAMPLES:
322
323 .nf
324 # lttng create -U net://192.168.1.42
325 .fi
326 Uses TCP and default ports for the given destination.
327
328 .nf
329 # lttng create -U net6://[fe80::f66d:4ff:fe53:d220]
330 .fi
331 Uses TCP, default ports and IPv6.
332
333 .nf
334 # lttng create s1 -U net://myhost.com:3229
335 .fi
336 Create session s1 and set its consumer to myhost.com on port 3229 for control.
337 .RE
338 .PP
339
340 .PP
341 \fBdestroy\fP [NAME] [OPTIONS]
342 .RS
343 Teardown tracing session
344
345 Free memory on the session daemon and tracer side. It's gone!
346
347 If NAME is omitted, the session name is taken from the .lttngrc file.
348
349 .B OPTIONS:
350
351 .TP
352 .BR "\-h, \-\-help"
353 Show summary of possible options and commands.
354 .TP
355 .BR "\-a, \-\-all"
356 Destroy all sessions
357 .TP
358 .BR "\-\-list-options"
359 Simple listing of options
360 .RE
361 .PP
362
363 .PP
364 \fBenable-channel\fP NAME[,NAME2,...] (\-k | \-u) [OPTIONS]
365 .RS
366 Enable tracing channel
367
368 To enable an event, you must enable both the event and the channel that
369 contains it.
370
371 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
372 file.
373
374 Exactly one of \-k or -u must be specified.
375
376 It is important to note that if a certain type of buffers is used, the session
377 will be set with that type and all other subsequent channel needs to have the
378 same type.
379
380 Note that once the session has been started and enabled on the tracer side,
381 it's not possible anymore to enable a new channel for that session.
382
383 .B OPTIONS:
384
385 .TP
386 .BR "\-h, \-\-help"
387 Show this help
388 .TP
389 .BR "\-\-list-options"
390 Simple listing of options
391 .TP
392 .BR "\-s, \-\-session NAME"
393 Apply on session name
394 .TP
395 .BR "\-k, \-\-kernel"
396 Apply to the kernel tracer
397 .TP
398 .BR "\-u, \-\-userspace"
399 Apply to the user-space tracer
400 .TP
401 .BR "\-\-discard"
402 Discard event when subbuffers are full (default)
403 .TP
404 .BR "\-\-overwrite"
405 Flight recorder mode: overwrites events when subbuffers are full. The
406 number of subbuffer must be 2 or more.
407 .TP
408 .BR "\-\-subbuf-size SIZE"
409 Subbuffer size in bytes {+k,+M,+G}.
410 (default UST uid: 131072, UST pid: 4096, kernel: 262144, metadata: 4096)
411 Rounded up to the next power of 2.
412
413 The minimum subbuffer size, for each tracer, is the max value between
414 the default above and the system page size. You can issue this command
415 to get the current page size on your system: \fB$ getconf PAGE_SIZE\fP
416 .TP
417 .BR "\-\-num-subbuf NUM"
418 Number of subbuffers. (default UST uid: 4, UST pid: 4, kernel: 4,
419 metadata: 2) Rounded up to the next power of 2.
420 .TP
421 .BR "\-\-switch-timer USEC"
422 Switch subbuffer timer interval in µsec.
423 (default UST uid: 0, UST pid: 0, kernel: 0, metadata: 0)
424 .TP
425 .BR "\-\-read-timer USEC"
426 Read timer interval in µsec.
427 (default UST uid: 0, UST pid: 0, kernel: 200000, metadata: 0)
428 .TP
429 .BR "\-\-output TYPE"
430 Channel output type. Possible values: mmap, splice
431 (default UST uid: mmap, UST pid: mmap, kernel: splice, metadata: mmap)
432 .TP
433 .BR "\-\-buffers-uid"
434 Use per UID buffer (\-u only). Buffers are shared between applications
435 that have the same UID.
436 .TP
437 .BR "\-\-buffers-pid"
438 Use per PID buffer (\-u only). Each application has its own buffers.
439 .TP
440 .BR "\-\-buffers-global"
441 Use shared buffer for the whole system (\-k only)
442 .TP
443 .BR "\-C, \-\-tracefile-size SIZE"
444 Maximum size of each tracefile within a stream (in bytes).
445 0 means unlimited. (default: 0)
446 Note: traces generated with this option may inaccurately report
447 discarded events as per CTF 1.8.
448 .TP
449 .BR "\-W, \-\-tracefile-count COUNT"
450 Used in conjunction with \-C option, this will limit the number of files
451 created to the specified count. 0 means unlimited. (default: 0)
452
453 .B EXAMPLES:
454
455 .nf
456 $ lttng enable-channel -k -C 4096 -W 32 chan1
457 .fi
458 For each stream, the maximum size of each trace file will be 4096 bytes and
459 there will be a maximum of 32 different files. The file count is appended after
460 the stream number as seen in the following example. The last trace file is
461 smaller than 4096 since it was not completely filled.
462
463 .nf
464 ~/lttng-traces/[...]/chan1_0_0 (4096)
465 ~/lttng-traces/[...]/chan1_0_1 (4096)
466 ~/lttng-traces/[...]/chan1_0_2 (3245)
467 ~/lttng-traces/[...]/chan1_1_0 (4096)
468 ...
469 .fi
470
471 .nf
472 $ lttng enable-channel -k -C 4096
473 .fi
474 This will create trace files of 4096 bytes and will create new ones as long as
475 there is data available.
476 .RE
477 .PP
478
479 .PP
480 \fBenable-event\fP NAME[,NAME2,...] [-k|-u] [OPTIONS]
481 .RS
482 Enable tracing event
483
484 A tracing event is always assigned to a channel. If \fB\-c, \-\-channel\fP is
485 omitted, a default channel named '\fBchannel0\fP' is created and the event is
486 added to it. If \fB\-c, \-\-channel\fP is omitted, but a non-default
487 channel already exists within the session, an error is returned. For the
488 user-space tracer, using \fB\-a, \-\-all\fP is the same as using the
489 wildcard "*".
490
491 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
492 file.
493
494 .B OPTIONS:
495
496 .TP
497 .BR "\-h, \-\-help"
498 Show summary of possible options and commands.
499 .TP
500 .BR "\-\-list-options"
501 Simple listing of options
502 .TP
503 .BR "\-s, \-\-session NAME"
504 Apply on session name
505 .TP
506 .BR "\-c, \-\-channel NAME"
507 Apply on channel name
508 .TP
509 .BR "\-a, \-\-all"
510 Enable all tracepoints and syscalls. This actually enables a single
511 wildcard event "*".
512 .TP
513 .BR "\-k, \-\-kernel"
514 Apply for the kernel tracer
515 .TP
516 .BR "\-u, \-\-userspace"
517 Apply for the user-space tracer
518 .TP
519 .BR "\-j, \-\-jul"
520 Apply for Java application using Java Util Logging interface (JUL)
521 .TP
522 .BR "\-\-tracepoint"
523 Tracepoint event (default). Userspace tracer supports wildcards at the end
524 of string. Don't forget to quote to deal with bash expansion.
525 e.g.:
526 .nf
527 "*"
528 "app_component:na*"
529 .fi
530 .TP
531 .BR "\-\-loglevel NAME"
532 Tracepoint loglevel range from 0 to loglevel. Listed in the help (\-h).
533 For the JUL domain, the loglevel ranges are detailed with the \-\-help
534 option thus starting from SEVERE to FINEST.
535 .TP
536 .BR "\-\-loglevel-only NAME"
537 Tracepoint loglevel (only this loglevel).
538 The loglevel or loglevel-only options should be combined with a
539 tracepoint name or tracepoint wildcard.
540 .TP
541 .BR "\-\-probe (addr | symbol | symbol+offset)"
542 Dynamic probe. Addr and offset can be octal (0NNN...), decimal (NNN...)
543 or hexadecimal (0xNNN...)
544 .TP
545 .BR "\-\-function (addr | symbol | symbol+offset)"
546 Dynamic function entry/return probe. Addr and offset can be octal
547 (0NNN...), decimal (NNN...) or hexadecimal (0xNNN...)
548 .TP
549 .BR "\-\-syscall"
550 System call event. Enabling syscalls tracing (kernel tracer), you will
551 not be able to disable them with disable-event. This is a known
552 limitation. You can disable the entire channel to do the trick. Also note
553 that per-syscall selection is not supported yet. Use with "-a" to enable
554 all syscalls.
555 .TP
556 .BR "\-\-filter 'expression'"
557 Set a filter on a newly enabled event. Filter expression on event
558 fields and context. The event will be recorded if the filter's
559 expression evaluates to TRUE. Only specify on first activation of a
560 given event within a session.
561 Specifying a filter is only allowed when enabling events within a session before
562 tracing is started. If the filter fails to link with the event
563 within the traced domain, the event will be discarded.
564 Filtering is currently only implemented for the user-space tracer.
565
566 Expression examples:
567
568 .nf
569 'intfield > 500 && intfield < 503'
570 '(strfield == "test" || intfield != 10) && intfield > 33'
571 'doublefield > 1.1 && intfield < 5.3'
572 .fi
573
574 Wildcards are allowed at the end of strings:
575 'seqfield1 == "te*"'
576 In string literals, the escape character is a '\\'. Use '\\*' for
577 the '*' character, and '\\\\' for the '\\' character sequence. Wildcard
578 matches any sequence of characters, including an empty sub-string
579 (matches 0 or more characters).
580
581 Context information can be used for filtering. The examples below shows
582 usage of context filtering on the process name (using a wildcard), process ID
583 range, and unique thread ID. The process and thread IDs of
584 running applications can be found under columns "PID" and "LWP" of the
585 "ps -eLf" command.
586
587 .nf
588 '$ctx.procname == "demo*"'
589 '$ctx.vpid >= 4433 && $ctx.vpid < 4455'
590 '$ctx.vtid == 1234'
591 .fi
592
593 Context information is available to all filters whether or not the add-context
594 command has been used to add it to the event's channel, as long as the context
595 field exists for that domain. For example, the filter examples given above will
596 never fail to link: no add-context is required for the event's channel.
597
598 .TP
599 .BR "\-x, \-\-exclude LIST"
600 Add exclusions to UST tracepoints:
601 Events that match any of the items in the comma-separated LIST are not
602 enabled, even if they match a wildcard definition of the event.
603
604 This option is also applicable with the \fB\-a, \-\-all\fP option,
605 in which case all UST tracepoints are enabled except the ones whose
606 names match any of the items in LIST.
607 .RE
608 .PP
609
610 .PP
611 \fBdisable-channel\fP NAME[,NAME2,...] (\-k | \-u) [OPTIONS]
612 .RS
613 Disable tracing channel
614
615 Disabling a channel disables the tracing of all of the channel's events. A channel
616 can be re-enabled by calling \fBlttng enable-channel NAME\fP again.
617
618 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
619 file.
620
621 .B OPTIONS:
622
623 .TP
624 .BR "\-h, \-\-help"
625 Show summary of possible options and commands.
626 .TP
627 .BR "\-\-list-options"
628 Simple listing of options
629 .TP
630 .BR "\-s, \-\-session NAME"
631 Apply on session name
632 .TP
633 .BR "\-k, \-\-kernel"
634 Apply for the kernel tracer
635 .TP
636 .BR "\-u, \-\-userspace"
637 Apply for the user-space tracer
638 .RE
639 .PP
640
641 .PP
642 \fBdisable-event\fP NAME[,NAME2,...] (\-k | \-u) [OPTIONS]
643 .RS
644 Disable tracing event
645
646 The event, once disabled, can be re-enabled by calling \fBlttng enable-event
647 NAME\fP again.
648
649 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
650 file.
651
652 If \fB\-c, \-\-channel\fP is omitted, the default channel name is used.
653 If \fB\-c, \-\-channel\fP is omitted, but a non-default channel already
654 exists within the session, an error is returned.
655
656 .B OPTIONS:
657
658 .TP
659 .BR "\-h, \-\-help"
660 Show summary of possible options and commands.
661 .TP
662 .BR "\-\-list-options"
663 Simple listing of options
664 .TP
665 .BR "\-s, \-\-session NAME"
666 Apply on session name
667 .TP
668 .BR "\-c, \-\-channel NAME"
669 Apply on channel name
670 .TP
671 .BR "\-a, \-\-all-events"
672 Disable all events. This does NOT disable "*" but rather every known
673 events of the session.
674 .TP
675 .BR "\-k, \-\-kernel"
676 Apply for the kernel tracer
677 .TP
678 .BR "\-u, \-\-userspace"
679 Apply for the user-space tracer
680 .TP
681 .BR "\-j, \-\-jul"
682 Apply for Java application using Java Util Logging interface (JUL)
683 .RE
684 .PP
685
686 .PP
687 \fBlist\fP [OPTIONS] [SESSION [SESSION OPTIONS]]
688 .RS
689 List tracing session information.
690
691 With no arguments, it will list available tracing session(s).
692
693 With the session name, it will display the details of the session including
694 the trace file path, the associated channels and their state (activated
695 and deactivated), the activated events and more.
696
697 With \-k alone, it will list all available kernel events (except the system
698 calls events).
699 With \-j alone, the available JUL event from registered application will be
700 list. The event corresponds to the Logger name in the Java JUL application.
701 With \-u alone, it will list all available user-space events from registered
702 applications. Here is an example of 'lttng list \-u':
703
704 .nf
705 PID: 7448 - Name: /tmp/lttng-ust/tests/hello/.libs/lt-hello
706 ust_tests_hello:tptest_sighandler (type: tracepoint)
707 ust_tests_hello:tptest (type: tracepoint)
708 .fi
709
710 You can now enable any event listed by using the name :
711 \fBust_tests_hello:tptest\fP.
712
713 .B OPTIONS:
714
715 .TP
716 .BR "\-h, \-\-help"
717 Show summary of possible options and commands.
718 .TP
719 .BR "\-\-list-options"
720 Simple listing of options
721 .TP
722 .BR "\-k, \-\-kernel"
723 Select kernel domain
724 .TP
725 .BR "\-u, \-\-userspace"
726 Select user-space domain.
727 .TP
728 .BR "\-j, \-\-jul"
729 Apply for Java application using JUL
730 .TP
731 .BR "\-f, \-\-fields"
732 List event fields
733
734 .PP
735 .B SESSION OPTIONS:
736
737 .TP
738 .BR "\-c, \-\-channel NAME"
739 List details of a channel
740 .TP
741 .BR "\-d, \-\-domain"
742 List available domain(s)
743 .RE
744 .PP
745
746 .PP
747 \fBload\fP [OPTIONS] [NAME]
748 .RS
749 Load tracing session configuration
750
751 If NAME is omitted, all session configurations found in both the user's session
752 configuration directory and the system session configuration directory will be
753 loaded.
754
755 .B OPTIONS:
756
757 .TP
758 .BR "\-h, \-\-help"
759 Show summary of possible options and commands.
760 .TP
761 .BR "\-a, \-\-all"
762 Load all session configurations (default).
763 .TP
764 .BR "\-i, \-\-input-path PATH"
765 Specify the input path for session configurations.
766 .TP
767 .BR "\-f, -\-force"
768 Overwrite current session configuration(s) if a session of the same name
769 already exists.
770 .RE
771 .PP
772
773 .PP
774 \fBsave\fP [OPTIONS] [SESSION]
775 .RS
776 Save tracing session configuration
777
778 If SESSION is omitted, all session configurations will be saved to individual
779 \fB.lttng\fP files under the user's session configuration directory. The
780 default session configuration file naming scheme is \fBSESSION.lttng\fP.
781
782 .B OPTIONS:
783
784 .TP
785 .BR "\-h, \-\-help"
786 Show summary of possible options and commands.
787 .TP
788 .BR "\-a, \-\-all"
789 Save all session configurations (default).
790 .TP
791 .BR "\-o, \-\-output-path PATH"
792 Specify the output path for saved sessions (default: ~/.lttng/sessions/).
793 .TP
794 .BR "\-f, -\-force"
795 Overwrite session configuration file if session name clashes.
796 .RE
797 .PP
798
799 .PP
800 \fBset-session\fP NAME [OPTIONS]
801 .RS
802 Set current session name
803
804 Will change the session name in the .lttngrc file.
805
806 .B OPTIONS:
807
808 .TP
809 .BR "\-h, \-\-help"
810 Show summary of possible options and commands.
811 .TP
812 .BR "\-\-list-options"
813 Simple listing of options
814 .RE
815 .PP
816
817 .PP
818 \fBsnapshot\fP [OPTIONS] ACTION
819 .RS
820 Snapshot command for LTTng session.
821
822 .B OPTIONS:
823
824 .TP
825 .BR "\-h, \-\-help"
826 Show summary of possible options and commands.
827 .TP
828 .BR "\-\-list-options"
829 Simple listing of options
830
831 .PP
832 .B ACTION:
833
834 .TP
835 \fBadd-output\fP [-m <SIZE>] [-s <NAME>] [-n <NAME>] <URL> | -C <URL> -D <URL>
836
837 Setup and add an snapshot output for a session. Output are the destination
838 where the snapshot will be sent. Only one output is permitted. To change it,
839 you'll need to delete it and add back the new one.
840
841 .TP
842 \fBdel-output\fP ID | NAME [-s <NAME>]
843
844 Delete an output for a session using the ID. You can either specify the
845 output's ID that can be found with list-output or the name.
846
847 .TP
848 \fBlist-output\fP [-s <NAME>]
849
850 List the output of a session. Attributes of the output are printed.
851
852 .TP
853 \fBrecord\fP [-m <SIZE>] [-s <NAME>] [-n <NAME>] [<URL> | -C <URL> -D <URL>]
854
855 Snapshot a session's buffer(s) for all domains. If an URL is specified, it is
856 used instead of a previously added output. Specifying only a name or/and a max
857 size will override the current output values. For instance, you can record a
858 snapshot with a custom maximum size or with a different name.
859
860 .nf
861 $ lttng snapshot add-output -n mysnapshot file:///data/snapshot
862 [...]
863 $ lttng snapshot record -n new_name_snapshot
864 .fi
865
866 The above will create a snapshot in /data/snapshot/new_name_snapshot* directory
867 rather then in mysnapshot*/
868
869 .PP
870 .B DETAILED ACTION OPTIONS
871
872 .TP
873 .BR "\-s, \-\-session NAME"
874 Apply to session name.
875 .TP
876 .BR "\-n, \-\-name NAME"
877 Name of the snapshot's output.
878 .TP
879 .BR "\-m, \-\-max-size SIZE"
880 Maximum size in bytes of the snapshot. The maxium size does not include the
881 metadata file. Human readable format is accepted: {+k,+M,+G}. For instance,
882 \-\-max-size 5M
883 .TP
884 .BR "\-C, \-\-ctrl-url URL"
885 Set control path URL. (Must use -D also)
886 .TP
887 .BR "\-D, \-\-data-url URL"
888 Set data path URL. (Must use -C also)
889 .RE
890 .PP
891
892 .PP
893 \fBstart\fP [NAME] [OPTIONS]
894 .RS
895 Start tracing
896
897 It will start tracing for all tracers for a specific tracing session.
898 If NAME is omitted, the session name is taken from the .lttngrc file.
899
900 .B OPTIONS:
901
902 .TP
903 .BR "\-h, \-\-help"
904 Show summary of possible options and commands.
905 .TP
906 .BR "\-\-list-options"
907 Simple listing of options
908 .RE
909 .PP
910
911 .PP
912 \fBstop\fP [NAME] [OPTIONS]
913 .RS
914 Stop tracing
915
916 It will stop tracing for all tracers for a specific tracing session. Before
917 returning, the command checks for data availability meaning that it will wait
918 until the trace is readable for the session. Use \-\-no-wait to avoid this
919 behavior.
920
921 If NAME is omitted, the session name is taken from the .lttngrc file.
922
923 .B OPTIONS:
924
925 .TP
926 .BR "\-h, \-\-help"
927 Show summary of possible options and commands.
928 .TP
929 .BR "\-\-list-options"
930 Simple listing of options
931 .TP
932 .BR "\-\-no-wait"
933 Don't wait for data availability.
934 .RE
935 .PP
936
937 .PP
938 \fBversion\fP
939 .RS
940 Show version information
941
942 .B OPTIONS:
943
944 .TP
945 .BR "\-h, \-\-help"
946 Show summary of possible options and commands.
947 .TP
948 .BR "\-\-list-options"
949 Simple listing of options
950 .RE
951 .PP
952
953 .PP
954 \fBview\fP [SESSION_NAME] [OPTIONS]
955 .RS
956 View traces of a tracing session. By default, the babeltrace viewer
957 will be used for text viewing. If SESSION_NAME is omitted, the session
958 name is taken from the .lttngrc file.
959
960 .B OPTIONS:
961
962 .TP
963 .BR "\-h, \-\-help"
964 Show this help
965 .TP
966 .BR "\-\-list-options"
967 Simple listing of options
968 .TP
969 .BR "\-t, \-\-trace-path PATH"
970 Trace directory path for the viewer
971 .TP
972 .BR "\-e, \-\-viewer CMD"
973 Specify viewer and/or options to use This will completely override the
974 default viewers so please make sure to specify the full command. The
975 trace directory path of the session will be appended at the end to the
976 arguments
977 .RE
978 .PP
979
980 .SH "JUL DOMAIN"
981 This section explains the JUL domain (\-j, \-\-jul) where JUL stands for Java
982 Util Logging. You can use that feature by using the \fBliblttng-ust-jul.so\fP
983 from the lttng-ust(3) project.
984
985 The LTTng Java Agent uses JNI to link the UST tracer to the Java application
986 that uses the agent. Thus, it behaves similarly to the UST domain (\-u). When
987 enabling events with the JUL domain, you enable a Logger name that will then be
988 mapped to a default UST tracepoint called \fBlttng_jul:jul_event\fP in the
989 \fBlttng_jul_channel\fP. Using the lttng-ctl API, any JUL events must use the
990 tracepoint event type (same as \-\-tracepoint).
991
992 Because of the default immutable channel (\fBlttng_jul_channel\fP), the
993 \fBenable-channel\fP command CAN NOT be used with the JUL domain thus not
994 having any \-j option.
995
996 For JUL event, loglevels are supported with the JUL ABI values. Use \fBlttng
997 enable-event \-h\fP to list them. Wildcards are NOT supported except the "*"
998 meaning all events (same as \-a).
999
1000 Exactly like the UST domain, if the Java application has the same UID as you,
1001 you can trace it. Same goes for the tracing group accessing root applications.
1002
1003 Finally, you can list every Logger name that are available from JUL registered
1004 applications to the session daemon by using \fBlttng list \-j\fP.
1005
1006 Here is an example on how to use this domain.
1007
1008 .nf
1009 $ lttng list -j
1010 [...]
1011 $ lttng create aSession
1012 $ lttng enable-event -s aSession -j MyCustomLoggerName
1013 $ lttng start
1014 .fi
1015
1016 More information can be found in the lttng-ust documentation, see
1017 java-util-logging.txt
1018 .PP
1019
1020 .SH "EXIT VALUES"
1021 .PP
1022 On success 0 is returned and a positive value on error. Value of 1 means a command
1023 error, 2 an undefined command, 3 a fatal error and 4 a command warning meaning that
1024 something went wrong during the command.
1025
1026 Any other value above 10, please refer to
1027 .BR "<lttng/lttng-error.h>"
1028 for a detailed list or use lttng_strerror() to get a human readable string of
1029 the error code.
1030 .PP
1031
1032 .SH "ENVIRONMENT VARIABLES"
1033
1034 .PP
1035 Note that all command line options override environment variables.
1036 .PP
1037
1038 .PP
1039 .IP "LTTNG_SESSIOND_PATH"
1040 Allows one to specify the full session daemon binary path to lttng command line
1041 tool. You can also use \-\-sessiond-path option having the same effect.
1042 .PP
1043
1044 .PP
1045 .IP "LTTNG_SESSION_CONFIG_XSD_PATH"
1046 Set the path in which the \fBsession.xsd\fP session configuration schema may be
1047 found.
1048 .PP
1049
1050 .SH "SEE ALSO"
1051 .BR babeltrace(1),
1052 .BR lttng-ust(3),
1053 .BR lttng-sessiond(8),
1054 .BR lttng-relayd(8),
1055
1056 .SH "BUGS"
1057
1058 .PP
1059 If you encounter any issues or usability problem, please report it on our
1060 mailing list <lttng-dev@lists.lttng.org> to help improve this project or
1061 at https://bugs.lttng.org which is a bugtracker.
1062 .PP
1063
1064 .SH "CREDITS"
1065
1066 .PP
1067 lttng is distributed under the GNU General Public License version 2. See the file
1068 COPYING for details.
1069 .PP
1070 A Web site is available at http://lttng.org for more information on the LTTng
1071 project.
1072 .PP
1073 You can also find our git tree at http://git.lttng.org.
1074 .PP
1075 Mailing lists for support and development: <lttng-dev@lists.lttng.org>.
1076 .PP
1077 You can find us on IRC server irc.oftc.net (OFTC) in #lttng.
1078 .PP
1079 .SH "THANKS"
1080
1081 .PP
1082 Thanks to Yannick Brosseau without whom this project would never have been so
1083 lean and mean! Also thanks to the Ericsson teams working on tracing which
1084 helped us greatly with detailed bug reports and unusual test cases.
1085
1086 Thanks to our beloved packager Alexandre Montplaisir-Goncalves (Ubuntu and PPA
1087 maintainer) and Jon Bernard for our Debian packages.
1088
1089 Special thanks to Michel Dagenais and the DORSAL laboratory at Polytechnique de
1090 Montreal for the LTTng journey.
1091 .PP
1092 .SH "AUTHORS"
1093
1094 .PP
1095 lttng-tools was originally written by Mathieu Desnoyers, Julien Desfossez and
1096 David Goulet. More people have since contributed to it. It is currently
1097 maintained by David Goulet <dgoulet@efficios.com>.
1098 .PP
This page took 0.04999 seconds and 3 git commands to generate.