1 .TH "LTTNG" "1" "May 13th, 2014" "" ""
4 lttng \(em LTTng 2.x tracer control command line tool
9 lttng [OPTIONS] <COMMAND>
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.
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
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.
29 We introduce the notion of \fBtracing domains\fP which is essentially a type of
30 tracer (kernel, user space, JUL, LOG4J or Python for now). In the future, we
31 could see more tracer like for instance an hypervisor. For some commands,
32 you'll need to specify on which domain the command operates (\-u, \-k, \-l, \-j
33 or \-p). For instance, the kernel domain must be specified when enabling a
36 In order to trace the kernel, the session daemon needs to be running as root.
37 LTTng provides the use of a \fBtracing group\fP (default: tracing). Whomever is
38 in that group can interact with the root session daemon and thus trace the
39 kernel. Session daemons can co-exist, meaning that you can have a session daemon
40 running as Alice that can be used to trace her applications along side with a
41 root daemon or even a Bob daemon. We highly recommend starting the session
42 daemon at boot time for stable and long term tracing.
44 Each user-space application instrumented with lttng-ust(3) will automatically
45 register with the root session daemon and its user session daemon. This allows
46 each daemon to list the available traceable applications and tracepoints at any
47 given moment (See the \fBlist\fP command).
51 This program follow the usual GNU command line syntax with long options starting with
52 two dashes. Below is a summary of the available options.
57 Show summary of possible options and commands.
59 .BR "\-v, \-\-verbose"
61 Three levels of verbosity are available which are triggered by putting additional v to
62 the option (\-vv or \-vvv)
65 Suppress all messages (even errors).
67 .BR "\-g, \-\-group NAME"
68 Set unix tracing group name. (default: tracing)
70 .BR "\-n, \-\-no-sessiond"
71 Don't automatically spawn a session daemon.
73 .BR "\-\-sessiond\-path PATH"
74 Set session daemon full binary path.
76 .BR "\-\-list\-options"
77 Simple listing of lttng options.
79 .BR "\-\-list\-commands"
80 Simple listing of lttng commands.
87 Machine interface (MI) mode converts the traditional pretty printing to a
88 machine output syntax. MI mode provides a format change-resistant way to access
89 information generated via the lttng command line.
91 When using MI mode, the data is printed on \fBstdout\fP. Error and warning are
92 printed on \fBstderr\fP with the pretty print default format.
94 If any errors occur during the execution of a command, the return value of the
95 command will be different than zero. In this case, lttng does NOT guarantee the
96 syntax and data validity of the generated MI output.
98 For XML output type, a schema definition (XSD) file used for validation can be
99 found under src/common/mi_lttng.xsd
104 \fBadd-context\fP [OPTIONS]
106 Add context to event(s) and/or channel(s).
108 A context is basically extra information appended to a channel. For instance,
109 you could ask the tracer to add the PID information for all events in a
110 channel. You can also add performance monitoring unit counters (perf PMU) using
113 For example, this command will add the context information 'prio' and two per-CPU
114 perf counters (hardware branch misses and cache misses), to all events in the trace
118 # lttng add-context \-k \-t prio \-t perf:cpu:branch-misses \\
119 \-t perf:cpu:cache-misses
122 Please take a look at the help (\-h/\-\-help) for a detailed list of available
125 Perf counters are available as per-CPU ("perf:cpu:...") and per-thread
126 ("perf:thread:...") counters. Currently, per-CPU counters can only be
127 used with the kernel tracing domain, and per-thread counters can only be
128 used with the UST tracing domain.
130 If no channel is given (\-c), the context is added to all channels that were
131 already enabled. If the session has no channel, a default channel is created.
132 Otherwise the context will be added only to the given channel (\-c).
134 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
141 Show summary of possible options and commands.
143 .BR "\-s, \-\-session NAME"
144 Apply on session name.
146 .BR "\-c, \-\-channel NAME"
147 Apply on channel name.
149 .BR "\-k, \-\-kernel"
150 Apply for the kernel tracer
152 .BR "\-u, \-\-userspace"
153 Apply for the user-space tracer
155 .BR "\-t, \-\-type TYPE"
156 Context type. You can repeat this option on the command line. Please
157 use "lttng add-context \-h" to list all available types.
162 \fBcalibrate\fP [OPTIONS]
164 Quantify LTTng overhead
166 The LTTng calibrate command can be used to find out the combined average
167 overhead of the LTTng tracer and the instrumentation mechanisms used. This
168 overhead can be calibrated in terms of time or using any of the PMU performance
169 counter available on the system.
171 For now, the only calibration implemented is that of the kernel function
172 instrumentation (kretprobes).
174 * Calibrate kernel function instrumentation
176 Let's use an example to show this calibration. We use an i7 processor with 4
177 general-purpose PMU registers. This information is available by issuing dmesg,
178 looking for "generic registers".
180 This sequence of commands will gather a trace executing a kretprobe hooked on
181 an empty function, gathering PMU counters LLC (Last Level Cache) misses
182 information (see lttng add-context \-\-help to see the list of available PMU
186 # lttng create calibrate-function
187 # lttng enable-event calibrate \-\-kernel \\
188 \-\-function lttng_calibrate_kretprobe
189 # lttng add-context \-\-kernel \-t perf:cpu:LLC-load-misses \\
190 \-t perf:cpu:LLC-store-misses \\
191 \-t perf:cpu:LLC-prefetch-misses
193 # for a in $(seq 1 10); do \\
194 lttng calibrate \-\-kernel \-\-function;
197 # babeltrace $(ls \-1drt ~/lttng-traces/calibrate-function-* \\
201 The output from babeltrace can be saved to a text file and opened in a
202 spreadsheet (e.g. oocalc) to focus on the per-PMU counter delta between
203 consecutive "calibrate_entry" and "calibrate_return" events. Note that these
204 counters are per-CPU, so scheduling events would need to be present to account
205 for migration between CPU. Therefore, for calibration purposes, only events
206 staying on the same CPU must be considered.
208 The average result, for the i7, on 10 samples:
212 perf_LLC_load_misses: 5.0 0.577
213 perf_LLC_store_misses: 1.6 0.516
214 perf_LLC_prefetch_misses: 9.0 14.742
217 As we can notice, the load and store misses are relatively stable across runs
218 (their standard deviation is relatively low) compared to the prefetch misses.
219 We can conclude from this information that LLC load and store misses can be
220 accounted for quite precisely, but prefetches within a function seems to behave
221 too erratically (not much causality link between the code executed and the CPU
222 prefetch activity) to be accounted for.
228 Show summary of possible options and commands.
230 .BR "\-k, \-\-kernel"
231 Apply for the kernel tracer
233 .BR "\-u, \-\-userspace"
234 Apply for the user-space tracer
237 Dynamic function entry/return probe (default)
242 \fBcreate\fP [NAME] [OPTIONS]
244 Create tracing session.
246 A tracing session contains channel(s) which contains event(s). It is domain
247 agnostic, meaning that channels and events can be enabled for the
248 user-space tracer and/or the kernel tracer. It acts like a container
249 aggregating multiple tracing sources.
251 On creation, a \fB.lttngrc\fP file is created in your $HOME directory
252 containing the current session name. If NAME is omitted, a session name is
253 automatically created having this form: 'auto-yyyymmdd-hhmmss'.
255 If no \fB\-o, \-\-output\fP is specified, the traces will be written in
258 The $HOME environment variable can be overridden by defining the environment
259 variable LTTNG_HOME. This is useful when the user running the commands has
260 a non-writeable home directory.
262 The session name MUST NOT contain the character '/'.
268 Show summary of possible options and commands.
270 .BR "\-\-list-options"
271 Simple listing of options
273 .BR "\-o, \-\-output PATH"
274 Specify output path for traces
277 Traces will not be output
280 Set the session in snapshot mode. Created in no-output mode and uses the
281 URL, if one is specified, as the default snapshot output. Every channel will be set
282 in overwrite mode and with mmap output (splice not supported).
284 .BR "\-\-live [USEC]"
285 Set the session exclusively in live mode. The parameter is the delay in micro
286 seconds before the data is flushed and streamed. The live mode allows you to
287 stream the trace and view it while it's being recorded by any tracer. For that,
288 you need a lttng-relayd and this session requires a network URL (\-U or
289 \-C/\-D). If no USEC nor URL is provided, the default is to use a timer value
290 set to 1000000 and the network URL set to net://127.0.0.1.
292 To read a live session, you can use babeltrace(1) or the live streaming
293 protocol in doc/live-reading-protocol.txt. Here is an example:
296 $ lttng-relayd -o /tmp/lttng
297 $ lttng create --live 200000 -U net://localhost
298 $ lttng enable-event -a --userspace
302 After the start, you'll be able to read the events while they are being
303 recorded in /tmp/lttng.
306 .BR "\-\-shm-path PATH"
308 Path where shared memory holding buffers should be created. Useful
309 when used with PRAMFS or other persistent memory filesystems to extract
310 trace data in the event of a crash requiring a reboot.
313 .BR "\-U, \-\-set-url=URL"
314 Set URL for the consumer output destination. It is persistent for the
315 session lifetime. Redo the command to change it. This will set both data
316 and control URL for network.
318 .BR "\-C, \-\-ctrl-url=URL"
319 Set control path URL. (Must use -D also)
321 .BR "\-D, \-\-data-url=URL"
322 Set data path URL. (Must use -C also)
324 Using these options, each API call can be controlled individually. For
325 instance, \-C does not enable the consumer automatically. You'll need the \-e
330 proto://[HOST|IP][:PORT1[:PORT2]][/TRACE_PATH]
332 Supported protocols are (proto):
335 Local filesystem full path.
339 This will use the default network transport layer which is TCP for both
340 control (PORT1) and data port (PORT2). The default ports are
341 respectively 5342 and 5343. Note that net[6]:// is not yet supported.
345 Can only be used with -C and -D together
347 NOTE: IPv6 address MUST be enclosed in brackets '[]' (rfc2732)
352 # lttng create -U net://192.168.1.42
354 Uses TCP and default ports for the given destination.
357 # lttng create -U net6://[fe80::f66d:4ff:fe53:d220]
359 Uses TCP, default ports and IPv6.
362 # lttng create s1 -U net://myhost.com:3229
364 Create session s1 and set its consumer to myhost.com on port 3229 for control.
369 \fBdestroy\fP [NAME] [OPTIONS]
371 Teardown tracing session
373 Free memory on the session daemon and tracer side. It's gone!
375 If NAME is omitted, the session name is taken from the .lttngrc file.
381 Show summary of possible options and commands.
386 .BR "\-\-list-options"
387 Simple listing of options
392 \fBenable-channel\fP NAME[,NAME2,...] (\-k | \-u) [OPTIONS]
394 Enable tracing channel
396 To enable an event, you must enable both the event and the channel that
399 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
402 Exactly one of \-k or -u must be specified.
404 It is important to note that if a certain type of buffers is used, the session
405 will be set with that type and all other subsequent channel needs to have the
408 Note that once the session has been started and enabled on the tracer side,
409 it's not possible anymore to enable a new channel for that session.
417 .BR "\-\-list-options"
418 Simple listing of options
420 .BR "\-s, \-\-session NAME"
421 Apply on session name
423 .BR "\-k, \-\-kernel"
424 Apply to the kernel tracer
426 .BR "\-u, \-\-userspace"
427 Apply to the user-space tracer
430 Discard event when subbuffers are full (default)
433 Flight recorder mode: overwrites events when subbuffers are full. The
434 number of subbuffer must be 2 or more.
436 .BR "\-\-subbuf-size SIZE"
437 Subbuffer size in bytes {+k,+M,+G}.
438 (default UST uid: 131072, UST pid: 4096, kernel: 262144, metadata: 4096)
439 Rounded up to the next power of 2.
441 The minimum subbuffer size, for each tracer, is the max value between
442 the default above and the system page size. You can issue this command
443 to get the current page size on your system: \fB$ getconf PAGE_SIZE\fP
445 .BR "\-\-num-subbuf NUM"
446 Number of subbuffers. (default UST uid: 4, UST pid: 4, kernel: 4,
447 metadata: 2) Rounded up to the next power of 2.
449 .BR "\-\-switch-timer USEC"
450 Switch subbuffer timer interval in µsec.
451 (default UST uid: 0, UST pid: 0, kernel: 0, metadata: 0)
453 .BR "\-\-read-timer USEC"
454 Read timer interval in µsec.
455 (default UST uid: 0, UST pid: 0, kernel: 200000, metadata: 0)
457 .BR "\-\-output TYPE"
458 Channel output type. Possible values: mmap, splice
459 (default UST uid: mmap, UST pid: mmap, kernel: splice, metadata: mmap)
461 .BR "\-\-buffers-uid"
462 Use per UID buffer (\-u only). Buffers are shared between applications
463 that have the same UID.
465 .BR "\-\-buffers-pid"
466 Use per PID buffer (\-u only). Each application has its own buffers.
468 .BR "\-\-buffers-global"
469 Use shared buffer for the whole system (\-k only)
471 .BR "\-C, \-\-tracefile-size SIZE"
472 Maximum size of each tracefile within a stream (in bytes).
473 0 means unlimited. (default: 0)
474 Note: traces generated with this option may inaccurately report
475 discarded events as of CTF 1.8.
477 .BR "\-W, \-\-tracefile-count COUNT"
478 Used in conjunction with \-C option, this will limit the number of files
479 created to the specified count. 0 means unlimited. (default: 0)
484 $ lttng enable-channel -k -C 4096 -W 32 chan1
486 For each stream, the maximum size of each trace file will be 4096 bytes and
487 there will be a maximum of 32 different files. The file count is appended after
488 the stream number as seen in the following example. The last trace file is
489 smaller than 4096 since it was not completely filled.
492 ~/lttng-traces/[...]/chan1_0_0 (4096)
493 ~/lttng-traces/[...]/chan1_0_1 (4096)
494 ~/lttng-traces/[...]/chan1_0_2 (3245)
495 ~/lttng-traces/[...]/chan1_1_0 (4096)
500 $ lttng enable-channel -k -C 4096
502 This will create trace files of 4096 bytes and will create new ones as long as
503 there is data available.
508 \fBenable-event\fP NAME[,NAME2,...] [-k|-u] [OPTIONS]
512 A tracing event is always assigned to a channel. If \fB\-c, \-\-channel\fP is
513 omitted, a default channel named '\fBchannel0\fP' is created and the event is
514 added to it. If \fB\-c, \-\-channel\fP is omitted, but a non-default
515 channel already exists within the session, an error is returned. For the
516 user-space tracer, using \fB\-a, \-\-all\fP is the same as using the
519 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
526 Show summary of possible options and commands.
528 .BR "\-\-list-options"
529 Simple listing of options
531 .BR "\-s, \-\-session NAME"
532 Apply on session name
534 .BR "\-c, \-\-channel NAME"
535 Apply on channel name
538 Enable all tracepoints and syscalls. This actually enables a single
541 .BR "\-k, \-\-kernel"
542 Apply for the kernel tracer
544 .BR "\-u, \-\-userspace"
545 Apply for the user-space tracer
548 Apply for Java application using Java Util Logging interface (JUL)
551 Apply for Java application using LOG4J
553 .BR "\-p, \-\-python"
554 Apply for Python application using the logging module.
557 Tracepoint event (default). Userspace tracer supports wildcards at the end
558 of string. Don't forget to quote to deal with bash expansion.
565 .BR "\-\-loglevel NAME"
566 Tracepoint loglevel range from 0 to loglevel. Listed in the help (\-h).
567 For the JUL domain, the loglevel ranges are detailed with the \-\-help
568 option thus starting from SEVERE to FINEST.
569 For the LOG4J domain, loglevels range from FATAL to TRACE which are also
570 detailed in the help.
571 For the Python domain, loglevels range from CRITICAL to DEBUG which are
572 detailed in the help as well.
574 .BR "\-\-loglevel-only NAME"
575 Tracepoint loglevel (only this loglevel).
576 The loglevel or loglevel-only options should be combined with a
577 tracepoint name or tracepoint wildcard.
579 .BR "\-\-probe (addr | symbol | symbol+offset)"
580 Dynamic probe. Addr and offset can be octal (0NNN...), decimal (NNN...)
581 or hexadecimal (0xNNN...)
583 .BR "\-\-function (addr | symbol | symbol+offset)"
584 Dynamic function entry/return probe. Addr and offset can be octal
585 (0NNN...), decimal (NNN...) or hexadecimal (0xNNN...)
588 System call event. Enabling syscalls tracing (kernel tracer), you will
589 not be able to disable them with disable-event. This is a known
590 limitation. You can disable the entire channel to do the trick. Also note
591 that per-syscall selection is not supported yet. Use with "-a" to enable
594 .BR "\-\-filter 'expression'"
595 Set a filter on a newly enabled event. Filter expression on event
596 fields and context. The event will be recorded if the filter's
597 expression evaluates to TRUE. Only specify on first activation of a
598 given event within a session.
599 Specifying a filter is only allowed when enabling events within a session before
600 tracing is started. If the filter fails to link with the event
601 within the traced domain, the event will be discarded.
602 Filtering is currently only implemented for the user-space tracer.
607 'intfield > 500 && intfield < 503'
608 '(strfield == "test" || intfield != 10) && intfield > 33'
609 'doublefield > 1.1 && intfield < 5.3'
612 Wildcards are allowed at the end of strings:
614 In string literals, the escape character is a '\\'. Use '\\*' for
615 the '*' character, and '\\\\' for the '\\' character sequence. Wildcard
616 matches any sequence of characters, including an empty sub-string
617 (matches 0 or more characters).
619 Context information can be used for filtering. The examples below shows
620 usage of context filtering on the process name (using a wildcard), process ID
621 range, and unique thread ID. The process and thread IDs of
622 running applications can be found under columns "PID" and "LWP" of the
626 '$ctx.procname == "demo*"'
627 '$ctx.vpid >= 4433 && $ctx.vpid < 4455'
631 Context information is available to all filters whether or not the add-context
632 command has been used to add it to the event's channel, as long as the context
633 field exists for that domain. For example, the filter examples given above will
634 never fail to link: no add-context is required for the event's channel.
637 .BR "\-x, \-\-exclude LIST"
638 Add exclusions to UST tracepoints:
639 Events that match any of the items in the comma-separated LIST are not
640 enabled, even if they match a wildcard definition of the event.
642 This option is also applicable with the \fB\-a, \-\-all\fP option,
643 in which case all UST tracepoints are enabled except the ones whose
644 names match any of the items in LIST.
649 \fBdisable-channel\fP NAME[,NAME2,...] (\-k | \-u) [OPTIONS]
651 Disable tracing channel
653 Disabling a channel disables the tracing of all of the channel's events. A channel
654 can be re-enabled by calling \fBlttng enable-channel NAME\fP again.
656 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
663 Show summary of possible options and commands.
665 .BR "\-\-list-options"
666 Simple listing of options
668 .BR "\-s, \-\-session NAME"
669 Apply on session name
671 .BR "\-k, \-\-kernel"
672 Apply for the kernel tracer
674 .BR "\-u, \-\-userspace"
675 Apply for the user-space tracer
680 \fBdisable-event\fP NAME[,NAME2,...] (\-k | \-u) [OPTIONS]
682 Disable tracing event
684 The event, once disabled, can be re-enabled by calling \fBlttng enable-event
687 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
690 If \fB\-c, \-\-channel\fP is omitted, the default channel name is used.
691 If \fB\-c, \-\-channel\fP is omitted, but a non-default channel already
692 exists within the session, an error is returned.
698 Show summary of possible options and commands.
700 .BR "\-\-list-options"
701 Simple listing of options
703 .BR "\-s, \-\-session NAME"
704 Apply on session name
706 .BR "\-c, \-\-channel NAME"
707 Apply on channel name
709 .BR "\-a, \-\-all-events"
710 Disable all events. This does NOT disable "*" but rather every known
711 events of the session.
713 .BR "\-k, \-\-kernel"
714 Apply for the kernel tracer
716 .BR "\-u, \-\-userspace"
717 Apply for the user-space tracer
720 Apply for Java application using Java Util Logging interface (JUL)
723 Apply for Java application using LOG4J
725 .BR "\-p, \-\-python"
726 Apply for Python application using the logging module.
731 \fBlist\fP [OPTIONS] [SESSION [SESSION OPTIONS]]
733 List tracing session information.
735 With no arguments, it will list available tracing session(s).
737 With the session name, it will display the details of the session including
738 the trace file path, the associated channels and their state (activated
739 and deactivated), the activated events and more.
741 With \-k alone, it will list all available kernel events (except the system
743 With \-j alone, the available JUL event from registered application will be
744 list. The event corresponds to the Logger name in the Java JUL application.
745 With \-l alone, the available LOG4J event from registered application will be
746 list. The event corresponds to the Logger name in the Java LOG4J application.
747 With \-p alone, the available Python event from registered application will be
748 list. The event corresponds to the Logger name in the Python application.
749 With \-u alone, it will list all available user-space events from registered
750 applications. Here is an example of 'lttng list \-u':
753 PID: 7448 - Name: /tmp/lttng-ust/tests/hello/.libs/lt-hello
754 ust_tests_hello:tptest_sighandler (type: tracepoint)
755 ust_tests_hello:tptest (type: tracepoint)
758 You can now enable any event listed by using the name :
759 \fBust_tests_hello:tptest\fP.
765 Show summary of possible options and commands.
767 .BR "\-\-list-options"
768 Simple listing of options
770 .BR "\-k, \-\-kernel"
773 .BR "\-u, \-\-userspace"
774 Select user-space domain.
777 Apply for Java application using JUL
780 Apply for Java application using LOG4J
782 .BR "\-p, \-\-python"
783 Apply for Python application using the logging module.
785 .BR "\-f, \-\-fields"
792 .BR "\-c, \-\-channel NAME"
793 List details of a channel
795 .BR "\-d, \-\-domain"
796 List available domain(s)
801 \fBload\fP [OPTIONS] [NAME]
803 Load tracing session configuration
805 If NAME is omitted, all session configurations found in both the user's session
806 configuration directory (default: ~/.lttng/sessions/) and the system session
807 configuration directory (default: /etc/lttng/sessions/) will be loaded. Note
808 that the sessions in the user directory are loaded first and then the system
809 wide directory are loaded.
815 Show summary of possible options and commands.
818 Load all session configurations (default).
820 .BR "\-i, \-\-input-path PATH"
821 Specify the input path for session configurations. This overrides the default
822 session configuration directory.
825 Overwrite current session configuration(s) if a session of the same name
831 \fBsave\fP [OPTIONS] [SESSION]
833 Save tracing session configuration
835 If SESSION is omitted, all session configurations will be saved to individual
836 \fB.lttng\fP files under the user's session configuration directory (default:
837 ~/.lttng/sessions/). The default session configuration file naming scheme is
840 For instance, a user in the tracing group saving a session from a root session
841 daemon will save it in her/his user directory.
847 Show summary of possible options and commands.
850 Save all session configurations (default).
852 .BR "\-o, \-\-output-path PATH"
853 Specify the output path for saved sessions. This overrides the default session
854 configuration directory.
857 Overwrite session configuration file if session name clashes.
862 \fBset-session\fP NAME [OPTIONS]
864 Set current session name
866 Will change the session name in the .lttngrc file.
872 Show summary of possible options and commands.
874 .BR "\-\-list-options"
875 Simple listing of options
880 \fBsnapshot\fP [OPTIONS] ACTION
882 Snapshot command for LTTng session.
888 Show summary of possible options and commands.
890 .BR "\-\-list-options"
891 Simple listing of options
897 \fBadd-output\fP [-m <SIZE>] [-s <NAME>] [-n <NAME>] <URL> | -C <URL> -D <URL>
899 Setup and add a snapshot output for a session. Output is the destination
900 where the snapshot will be sent. Only one output is permitted. To change it,
901 you'll need to delete it and add back the new one.
904 \fBdel-output\fP ID | NAME [-s <NAME>]
906 Delete an output for a session using the output's ID. You can either specify the
907 output by name or use its ID as returned by the list-output command.
910 \fBlist-output\fP [-s <NAME>]
912 List the output of a session. Attributes of the output are printed.
915 \fBrecord\fP [-m <SIZE>] [-s <NAME>] [-n <NAME>] [<URL> | -C <URL> -D <URL>]
917 Snapshot a session's buffer(s) for all domains. If an URL is specified, it is
918 used instead of a previously added output. Specifying only a name or/and a max
919 size will override the current output values. For instance, you can record a
920 snapshot with a custom maximum size or with a different name.
923 $ lttng snapshot add-output -n mysnapshot file:///data/snapshot
925 $ lttng snapshot record -n new_name_snapshot
928 The above will create a snapshot in /data/snapshot/new_name_snapshot* directory
929 rather then in mysnapshot*/
932 .B DETAILED ACTION OPTIONS
935 .BR "\-s, \-\-session NAME"
936 Apply to session name.
938 .BR "\-n, \-\-name NAME"
939 Name of the snapshot's output.
941 .BR "\-m, \-\-max-size SIZE"
942 Maximum size in bytes of the snapshot. The maximum size does not include the
943 metadata file. Human readable format is accepted: {+k,+M,+G}. For instance,
946 .BR "\-C, \-\-ctrl-url URL"
947 Set control path URL. (Must use -D also)
949 .BR "\-D, \-\-data-url URL"
950 Set data path URL. (Must use -C also)
955 \fBstart\fP [NAME] [OPTIONS]
959 It will start tracing for all tracers for a specific tracing session.
960 If NAME is omitted, the session name is taken from the .lttngrc file.
966 Show summary of possible options and commands.
968 .BR "\-\-list-options"
969 Simple listing of options
974 \fBstop\fP [NAME] [OPTIONS]
978 It will stop tracing for all tracers for a specific tracing session. Before
979 returning, the command checks for data availability meaning that it will wait
980 until the trace is readable for the session. Use \-\-no-wait to avoid this
983 If NAME is omitted, the session name is taken from the .lttngrc file.
989 Show summary of possible options and commands.
991 .BR "\-\-list-options"
992 Simple listing of options
995 Don't wait for data availability.
1002 Show version information
1008 Show summary of possible options and commands.
1010 .BR "\-\-list-options"
1011 Simple listing of options
1016 \fBview\fP [SESSION_NAME] [OPTIONS]
1018 View traces of a tracing session. By default, the babeltrace viewer
1019 will be used for text viewing. If SESSION_NAME is omitted, the session
1020 name is taken from the .lttngrc file.
1028 .BR "\-\-list-options"
1029 Simple listing of options
1031 .BR "\-t, \-\-trace-path PATH"
1032 Trace directory path for the viewer
1034 .BR "\-e, \-\-viewer CMD"
1035 Specify viewer and/or options to use This will completely override the
1036 default viewers so please make sure to specify the full command. The
1037 trace directory path of the session will be appended at the end to the
1042 .SH "JUL/LOG4J DOMAIN"
1044 This section explains the JUL and LOG4J domain where JUL stands for Java Util
1045 Logging. You can use these by using the \fBliblttng-ust-<domain>-jni.so\fP from
1046 the lttng-ust(3) project.
1048 The LTTng Java Agent uses JNI to link the UST tracer to the Java application
1049 that uses the agent. Thus, it behaves similarly to the UST domain (\-u). When
1050 enabling events, you enable a Logger name that will then be mapped to a default
1051 UST tracepoint called \fBlttng_jul:<domain>_event\fP in the
1052 \fBlttng_<domain>_channel\fP. Using the lttng-ctl API, any JUL/LOG4J events
1053 must use the tracepoint event type (same as \-\-tracepoint).
1055 Because of the default immutable channel, the \fBenable-channel\fP command CAN
1056 NOT be used with the JUL and LOG4J domain thus not having any options.
1058 Also, loglevels are supported. Use \fBlttng enable-event \-h\fP to list them.
1059 Wildcards are NOT supported except the "*" meaning all events (same as \-a).
1061 Exactly like the UST domain, if the Java application has the same UID as you,
1062 you can trace it. Same goes for the tracing group accessing root applications.
1064 Finally, you can list every Logger name that are available from registered
1065 applications to the session daemon by using \fBlttng list \-j\fP or \fB\-l\fP.
1067 Here is an example on how to use the JUL domain.
1072 $ lttng create aSession
1073 $ lttng enable-event -s aSession -j MyCustomLoggerName
1077 More information can be found in the lttng-ust documentation, see
1078 java-util-logging.txt
1083 On success 0 is returned and a positive value on error. Value of 1 means a command
1084 error, 2 an undefined command, 3 a fatal error and 4 a command warning meaning that
1085 something went wrong during the command.
1087 Any other value above 10, please refer to
1088 .BR "<lttng/lttng-error.h>"
1089 for a detailed list or use lttng_strerror() to get a human readable string of
1093 .SH "ENVIRONMENT VARIABLES"
1096 Note that all command line options override environment variables.
1100 .IP "LTTNG_SESSIOND_PATH"
1101 Allows one to specify the full session daemon binary path to lttng command line
1102 tool. You can also use \-\-sessiond-path option having the same effect.
1106 .IP "LTTNG_SESSION_CONFIG_XSD_PATH"
1107 Set the path in which the \fBsession.xsd\fP session configuration schema may be
1114 .BR lttng-sessiond(8),
1115 .BR lttng-relayd(8),
1120 If you encounter any issues or usability problem, please report it on our
1121 mailing list <lttng-dev@lists.lttng.org> to help improve this project or
1122 at https://bugs.lttng.org which is a bug tracker.
1128 lttng is distributed under the GNU General Public License version 2. See the file
1129 COPYING for details.
1131 A Web site is available at http://lttng.org for more information on the LTTng
1134 You can also find our git tree at http://git.lttng.org.
1136 Mailing lists for support and development: <lttng-dev@lists.lttng.org>.
1138 You can find us on IRC server irc.oftc.net (OFTC) in #lttng.
1143 Thanks to Yannick Brosseau without whom this project would never have been so
1144 lean and mean! Also thanks to the Ericsson teams working on tracing which
1145 helped us greatly with detailed bug reports and unusual test cases.
1147 Thanks to our beloved packager Alexandre Montplaisir-Goncalves (Ubuntu and PPA
1148 maintainer) and Jon Bernard for our Debian packages.
1150 Special thanks to Michel Dagenais and the DORSAL laboratory at Polytechnique de
1151 Montreal for the LTTng journey.
1156 lttng-tools was originally written by Mathieu Desnoyers, Julien Desfossez and
1157 David Goulet. More people have since contributed to it. It is currently
1158 maintained by Jérémie Galarneau <jeremie.galarneau@efficios.com>.