1 The LTTng Documentation
2 =======================
3 Philippe Proulx <pproulx@efficios.com>
7 include::../common/copyright.txt[]
10 include::../common/welcome.txt[]
13 include::../common/audience.txt[]
17 === What's in this documentation?
19 The LTTng Documentation is divided into the following sections:
21 * **<<nuts-and-bolts,Nuts and bolts>>** explains the
22 rudiments of software tracing and the rationale behind the
25 You can skip this section if you’re familiar with software tracing and
26 with the LTTng project.
28 * **<<installing-lttng,Installation>>** describes the steps to
29 install the LTTng packages on common Linux distributions and from
32 You can skip this section if you already properly installed LTTng on
35 * **<<getting-started,Quick start>>** is a concise guide to
36 getting started quickly with LTTng kernel and user space tracing.
38 We recommend this section if you're new to LTTng or to software tracing
41 You can skip this section if you're not new to LTTng.
43 * **<<core-concepts,Core concepts>>** explains the concepts at
46 It's a good idea to become familiar with the core concepts
47 before attempting to use the toolkit.
49 * **<<plumbing,Components of LTTng>>** describes the various components
50 of the LTTng machinery, like the daemons, the libraries, and the
51 command-line interface.
52 * **<<instrumenting,Instrumentation>>** shows different ways to
53 instrument user applications and the Linux kernel.
55 Instrumenting source code is essential to provide a meaningful
58 You can skip this section if you do not have a programming background.
60 * **<<controlling-tracing,Tracing control>>** is divided into topics
61 which demonstrate how to use the vast array of features that
62 LTTng{nbsp}{revision} offers.
63 * **<<reference,Reference>>** contains reference tables.
64 * **<<glossary,Glossary>>** is a specialized dictionary of terms related
65 to LTTng or to the field of software tracing.
68 include::../common/convention.txt[]
71 include::../common/acknowledgements.txt[]
75 == What's new in LTTng{nbsp}{revision}?
77 LTTng{nbsp}{revision} bears the name _Lafontaine_. This modern
78 https://en.wikipedia.org/wiki/saison[saison] from the
79 https://oshlag.com/[Oshlag] microbrewery is a refreshing--zesty--rice
80 beer with hints of fruit and spices. Some even say it makes for a great
81 https://en.wikipedia.org/wiki/Somaek[Somaek] when mixed with
82 Chamisul Soju, not that we've tried!
84 New features and changes in LTTng{nbsp}{revision}:
86 * Just like you can typically perform
87 https://en.wikipedia.org/wiki/Log_rotation[log rotation], you can
88 now <<session-rotation,_rotate_ a tracing session>>, that
89 is, according to man:lttng-rotate(1), archive the current trace
90 chunk (all the tracing session's trace data since the last rotation
91 or since its inception) so that LTTng does not manage it anymore.
93 Once LTTng archives a trace chunk, you are free to read it, modify it,
94 move it, or remove it.
96 You can rotate a tracing session immediately or set a rotation schedule
97 to automate rotations.
99 * When you <<enabling-disabling-events,create an event rule>>, the
100 filter expression syntax now supports the following new operators:
104 ** `<<` (bitwise left shift)
105 ** `>>` (bitwise right shift)
111 The syntax also supports array indexing with the usual square brackets:
114 regs[3][1] & 0xff7 == 0x240
117 There are peculiarities for both the new operators and the array
118 indexing brackets, like a custom precedence table and implicit casting.
119 See man:lttng-enable-event(1) to get all the details about the filter
122 * You can now dynamically instrument any application's or library's
123 function entry by symbol name thanks to the new
124 opt:lttng-enable-event(1):--userspace-probe option of
125 the `lttng enable-event` command:
129 $ lttng enable-event --kernel \
130 --userspace-probe=/usr/lib/libc.so.6:malloc libc_malloc
133 The option also supports tracing existing
134 https://www.sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps[SystemTap
135 Statically Defined Tracing] (USDT) probe (DTrace-style marker). For
136 example, given the following probe:
140 DTRACE_PROBE2("server", "accept-request", request_id, ip_addr);
143 You can trace this probe with:
146 $ lttng enable-event --kernel \
147 --userspace-probe=sdt:/path/to/server:server:accept-request \
148 server_accept_request
151 This feature makes use of Linux's
152 https://www.kernel.org/doc/Documentation/trace/uprobetracer.txt[uprobe]
153 mechanism, therefore you must use the `--userspace-probe`
154 instrumentation option with the opt:lttng-enable-event(1):--kernel
157 NOTE: As of LTTng{nbsp}{revision}, LTTng does not record function
158 parameters with the opt:lttng-enable-event(1):--userspace-probe option.
160 * Two new <<adding-context,context>> fields are available for Linux
161 kernel <<channel,channels>>:
164 ** `callstack-kernel`
168 Thanks to those, you can record the Linux kernel and user call stacks
169 when a kernel event occurs. For example:
173 $ lttng enable-event --kernel --syscall open
174 $ lttng add-context --kernel --type=callstack-kernel --type=callstack-user
177 When an man:open(2) system call occurs, LTTng attaches the kernel and
178 user call stacks to the recorded event.
180 NOTE: LTTng cannot always sample the user space call stack reliably.
181 For instance, LTTng cannot sample the call stack of user applications
182 and libraries compiled with the
183 https://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html[`-fomit-frame-pointer`]
184 option. In such a case, the tracing is not affected, but the sampled
185 user space call stack may only contain the user call stack's topmost
188 * User applications and libraries instrumented with
189 <<lttng-ust,LTTng-UST>> can now safely unload (man:dlclose(3)) a
191 <<building-tracepoint-providers-and-user-application,tracepoint
194 * The <<lttng-relayd,relay daemon>> is more efficient and presents fewer
195 connectivity issues, especially when a large number of targets send
196 trace data to a given relay daemon.
198 * LTTng-UST uses https://github.com/numactl/numactl[libnuma]
199 when available to allocate <<def-sub-buffer,sub-buffers>>, making them
201 https://en.wikipedia.org/wiki/Non-uniform_memory_access[NUMA] node.
203 This change makes the tracer more efficient on NUMA systems.
205 * The <<proc-lttng-logger-abi,LTTng logger>> kernel module now also
206 creates a ``misc'' device named `lttng-logger`, which udev will
207 make accessible as the path:{/dev/lttng-logger} special file.
209 The `lttng-logger` device shares the `/proc/lttng-logger` file's ABI,
210 but it works from within containers when the path is made accessible to
217 What is LTTng? As its name suggests, the _Linux Trace Toolkit: next
218 generation_ is a modern toolkit for tracing Linux systems and
219 applications. So your first question might be:
226 As the history of software engineering progressed and led to what
227 we now take for granted--complex, numerous and
228 interdependent software applications running in parallel on
229 sophisticated operating systems like Linux--the authors of such
230 components, software developers, began feeling a natural
231 urge to have tools that would ensure the robustness and good performance
232 of their masterpieces.
234 One major achievement in this field is, inarguably, the
235 https://www.gnu.org/software/gdb/[GNU debugger (GDB)],
236 an essential tool for developers to find and fix bugs. But even the best
237 debugger won't help make your software run faster, and nowadays, faster
238 software means either more work done by the same hardware, or cheaper
239 hardware for the same work.
241 A _profiler_ is often the tool of choice to identify performance
242 bottlenecks. Profiling is suitable to identify _where_ performance is
243 lost in a given software. The profiler outputs a profile, a statistical
244 summary of observed events, which you may use to discover which
245 functions took the most time to execute. However, a profiler won't
246 report _why_ some identified functions are the bottleneck. Bottlenecks
247 might only occur when specific conditions are met, conditions that are
248 sometimes impossible to capture by a statistical profiler, or impossible
249 to reproduce with an application altered by the overhead of an
250 event-based profiler. For a thorough investigation of software
251 performance issues, a history of execution is essential, with the
252 recorded values of variables and context fields you choose, and
253 with as little influence as possible on the instrumented software. This
254 is where tracing comes in handy.
256 _Tracing_ is a technique used to understand what goes on in a running
257 software system. The software used for tracing is called a _tracer_,
258 which is conceptually similar to a tape recorder. When recording,
259 specific instrumentation points placed in the software source code
260 generate events that are saved on a giant tape: a _trace_ file. You
261 can trace user applications and the operating system at the same time,
262 opening the possibility of resolving a wide range of problems that would
263 otherwise be extremely challenging.
265 Tracing is often compared to _logging_. However, tracers and loggers are
266 two different tools, serving two different purposes. Tracers are
267 designed to record much lower-level events that occur much more
268 frequently than log messages, often in the range of thousands per
269 second, with very little execution overhead. Logging is more appropriate
270 for a very high-level analysis of less frequent events: user accesses,
271 exceptional conditions (errors and warnings, for example), database
272 transactions, instant messaging communications, and such. Simply put,
273 logging is one of the many use cases that can be satisfied with tracing.
275 The list of recorded events inside a trace file can be read manually
276 like a log file for the maximum level of detail, but it is generally
277 much more interesting to perform application-specific analyses to
278 produce reduced statistics and graphs that are useful to resolve a
279 given problem. Trace viewers and analyzers are specialized tools
282 In the end, this is what LTTng is: a powerful, open source set of
283 tools to trace the Linux kernel and user applications at the same time.
284 LTTng is composed of several components actively maintained and
285 developed by its link:/community/#where[community].
288 [[lttng-alternatives]]
289 === Alternatives to noch:{LTTng}
291 Excluding proprietary solutions, a few competing software tracers
294 * https://github.com/dtrace4linux/linux[dtrace4linux] is a port of
295 Sun Microsystems's DTrace to Linux. The cmd:dtrace tool interprets
296 user scripts and is responsible for loading code into the
297 Linux kernel for further execution and collecting the outputted data.
298 * https://en.wikipedia.org/wiki/Berkeley_Packet_Filter[eBPF] is a
299 subsystem in the Linux kernel in which a virtual machine can execute
300 programs passed from the user space to the kernel. You can attach
301 such programs to tracepoints and kprobes thanks to a system call, and
302 they can output data to the user space when executed thanks to
303 different mechanisms (pipe, VM register values, and eBPF maps, to name
305 * https://www.kernel.org/doc/Documentation/trace/ftrace.txt[ftrace]
306 is the de facto function tracer of the Linux kernel. Its user
307 interface is a set of special files in sysfs.
308 * https://perf.wiki.kernel.org/[perf] is
309 a performance analysis tool for Linux which supports hardware
310 performance counters, tracepoints, as well as other counters and
311 types of probes. perf's controlling utility is the cmd:perf command
313 * http://linux.die.net/man/1/strace[strace]
314 is a command-line utility which records system calls made by a
315 user process, as well as signal deliveries and changes of process
316 state. strace makes use of https://en.wikipedia.org/wiki/Ptrace[ptrace]
317 to fulfill its function.
318 * http://www.sysdig.org/[sysdig], like SystemTap, uses scripts to
319 analyze Linux kernel events. You write scripts, or _chisels_ in
320 sysdig's jargon, in Lua and sysdig executes them while it traces the
321 system or afterwards. sysdig's interface is the cmd:sysdig
322 command-line tool as well as the curses-based cmd:csysdig tool.
323 * https://sourceware.org/systemtap/[SystemTap] is a Linux kernel and
324 user space tracer which uses custom user scripts to produce plain text
325 traces. SystemTap converts the scripts to the C language, and then
326 compiles them as Linux kernel modules which are loaded to produce
327 trace data. SystemTap's primary user interface is the cmd:stap
330 The main distinctive features of LTTng is that it produces correlated
331 kernel and user space traces, as well as doing so with the lowest
332 overhead amongst other solutions. It produces trace files in the
333 http://diamon.org/ctf[CTF] format, a file format optimized
334 for the production and analyses of multi-gigabyte data.
336 LTTng is the result of more than 10{nbsp}years of active open source
337 development by a community of passionate developers.
338 LTTng{nbsp}{revision} is currently available on major desktop and server
341 The main interface for tracing control is a single command-line tool
342 named cmd:lttng. The latter can create several tracing sessions, enable
343 and disable events on the fly, filter events efficiently with custom
344 user expressions, start and stop tracing, and much more. LTTng can
345 record the traces on the file system or send them over the network, and
346 keep them totally or partially. You can view the traces once tracing
347 becomes inactive or in real-time.
349 <<installing-lttng,Install LTTng now>> and
350 <<getting-started,start tracing>>!
356 **LTTng** is a set of software <<plumbing,components>> which interact to
357 <<instrumenting,instrument>> the Linux kernel and user applications, and
358 to <<controlling-tracing,control tracing>> (start and stop
359 tracing, enable and disable event rules, and the rest). Those
360 components are bundled into the following packages:
363 Libraries and command-line interface to control tracing.
366 Linux kernel modules to instrument and trace the kernel.
369 Libraries and Java/Python packages to instrument and trace user
372 Most distributions mark the LTTng-modules and LTTng-UST packages as
373 optional when installing LTTng-tools (which is always required). In the
374 following sections, we always provide the steps to install all three,
377 * You only need to install LTTng-modules if you intend to trace the
379 * You only need to install LTTng-UST if you intend to trace user
383 .Availability of LTTng{nbsp}{revision} for major Linux distributions as of 5{nbsp}August{nbsp}2020.
385 |Distribution |Available in releases
387 |https://www.ubuntu.com/[Ubuntu]
388 |<<ubuntu,Ubuntu{nbsp}20.04 _Focal Fossa_>>.
390 Ubuntu{nbsp}16.04 _Xenial Xerus_ and Ubuntu{nbsp}18.04 _Bionic Beaver_:
391 <<ubuntu-ppa,use the LTTng Stable{nbsp}{revision} PPA>>.
393 |https://getfedora.org/[Fedora]
394 |<<fedora,Fedora{nbsp}32>>.
396 |https://www.redhat.com/[RHEL] and https://www.suse.com/[SLES]
397 |See http://packages.efficios.com/[EfficiOS Enterprise Packages].
399 |https://buildroot.org/[Buildroot]
400 |xref:buildroot[Buildroot{nbsp}2019.11, Buildroot{nbsp}2020.02, and
401 Buildroot{nbsp}2020.05].
403 |https://www.openembedded.org/wiki/Main_Page[OpenEmbedded] and
404 https://www.yoctoproject.org/[Yocto]
405 |<<oe-yocto,Yocto Project{nbsp}3.1 _Dunfell_>>
406 (`openembedded-core` layer).
411 === [[ubuntu-official-repositories]]Ubuntu
413 LTTng{nbsp}{revision} is available on Ubuntu{nbsp}20.04 _Focal Fossa_.
414 For previous supported releases of Ubuntu,
415 <<ubuntu-ppa,use the LTTng Stable{nbsp}{revision} PPA>>.
417 To install LTTng{nbsp}{revision} on Ubuntu{nbsp}20.04 _Focal Fossa_:
419 . Install the main LTTng{nbsp}{revision} packages:
424 # apt-get install lttng-tools
425 # apt-get install lttng-modules-dkms
426 # apt-get install liblttng-ust-dev
430 . **If you need to instrument and trace
431 <<java-application,Java applications>>**, install the LTTng-UST
437 # apt-get install liblttng-ust-agent-java
441 . **If you need to instrument and trace
442 <<python-application,Python{nbsp}3 applications>>**, install the
443 LTTng-UST Python agent:
448 # apt-get install python3-lttngust
453 === Ubuntu: noch:{LTTng} Stable {revision} PPA
455 The https://launchpad.net/~lttng/+archive/ubuntu/stable-{revision}[LTTng
456 Stable{nbsp}{revision} PPA] offers the latest stable
457 LTTng{nbsp}{revision} packages for Ubuntu{nbsp}16.04 _Xenial Xerus_ and
458 Ubuntu{nbsp}18.04 _Bionic Beaver_.
460 To install LTTng{nbsp}{revision} from the LTTng Stable{nbsp}{revision}
463 . Add the LTTng Stable{nbsp}{revision} PPA repository and update the
469 # apt-add-repository ppa:lttng/stable-2.11
474 . Install the main LTTng{nbsp}{revision} packages:
479 # apt-get install lttng-tools
480 # apt-get install lttng-modules-dkms
481 # apt-get install liblttng-ust-dev
485 . **If you need to instrument and trace
486 <<java-application,Java applications>>**, install the LTTng-UST
492 # apt-get install liblttng-ust-agent-java
496 . **If you need to instrument and trace
497 <<python-application,Python{nbsp}3 applications>>**, install the
498 LTTng-UST Python agent:
503 # apt-get install python3-lttngust
511 To install LTTng{nbsp}{revision} on Fedora{nbsp}32:
513 . Install the LTTng-tools{nbsp}{revision} and LTTng-UST{nbsp}{revision}
519 # yum install lttng-tools
520 # yum install lttng-ust
524 . Download, build, and install the latest LTTng-modules{nbsp}{revision}:
530 wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.11.tar.bz2 &&
531 tar -xf lttng-modules-latest-2.11.tar.bz2 &&
532 cd lttng-modules-2.11.* &&
534 sudo make modules_install &&
540 .Java and Python application instrumentation and tracing
542 If you need to instrument and trace <<java-application,Java
543 applications>> on Fedora, you need to build and install
544 LTTng-UST{nbsp}{revision} <<building-from-source,from source>> and pass
545 the `--enable-java-agent-jul`, `--enable-java-agent-log4j`, or
546 `--enable-java-agent-all` options to the `configure` script, depending
547 on which Java logging framework you use.
549 If you need to instrument and trace <<python-application,Python
550 applications>> on Fedora, you need to build and install
551 LTTng-UST{nbsp}{revision} from source and pass the
552 `--enable-python-agent` option to the `configure` script.
559 To install LTTng{nbsp}{revision} on Buildroot{nbsp}2019.11,
560 Buildroot{nbsp}2020.02, or Buildroot{nbsp}2020.05:
562 . Launch the Buildroot configuration tool:
571 . In **Kernel**, check **Linux kernel**.
572 . In **Toolchain**, check **Enable WCHAR support**.
573 . In **Target packages**{nbsp}→ **Debugging, profiling and benchmark**,
574 check **lttng-modules** and **lttng-tools**.
575 . In **Target packages**{nbsp}→ **Libraries**{nbsp}→
576 **Other**, check **lttng-libust**.
580 === OpenEmbedded and Yocto
582 LTTng{nbsp}{revision} recipes are available in the
583 https://layers.openembedded.org/layerindex/branch/master/layer/openembedded-core/[`openembedded-core`]
584 layer for Yocto Project{nbsp}3.1 _Dunfell_ under the following names:
590 With BitBake, the simplest way to include LTTng recipes in your target
591 image is to add them to `IMAGE_INSTALL_append` in path:{conf/local.conf}:
594 IMAGE_INSTALL_append = " lttng-tools lttng-modules lttng-ust"
599 . Select a machine and an image recipe.
600 . Click **Edit image recipe**.
601 . Under the **All recipes** tab, search for **lttng**.
602 . Check the desired LTTng recipes.
605 [[building-from-source]]
606 === Build from source
608 To build and install LTTng{nbsp}{revision} from source:
610 . Using your distribution's package manager, or from source, install
611 the following dependencies of LTTng-tools and LTTng-UST:
614 * https://sourceforge.net/projects/libuuid/[libuuid]
615 * http://directory.fsf.org/wiki/Popt[popt]
616 * http://liburcu.org/[Userspace RCU]
617 * http://www.xmlsoft.org/[libxml2]
618 * **Optional**: https://github.com/numactl/numactl[numactl]
621 . Download, build, and install the latest LTTng-modules{nbsp}{revision}:
627 wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.11.tar.bz2 &&
628 tar -xf lttng-modules-latest-2.11.tar.bz2 &&
629 cd lttng-modules-2.11.* &&
631 sudo make modules_install &&
636 . Download, build, and install the latest LTTng-UST{nbsp}{revision}:
642 wget http://lttng.org/files/lttng-ust/lttng-ust-latest-2.11.tar.bz2 &&
643 tar -xf lttng-ust-latest-2.11.tar.bz2 &&
644 cd lttng-ust-2.11.* &&
652 Add `--disable-numa` to `./configure` if you don't have
653 https://github.com/numactl/numactl[numactl].
657 .Java and Python application tracing
659 If you need to instrument and trace <<java-application,Java
660 applications>>, pass the `--enable-java-agent-jul`,
661 `--enable-java-agent-log4j`, or `--enable-java-agent-all` options to the
662 `configure` script, depending on which Java logging framework you use.
664 If you need to instrument and trace <<python-application,Python
665 applications>>, pass the `--enable-python-agent` option to the
666 `configure` script. You can set the `PYTHON` environment variable to the
667 path to the Python interpreter for which to install the LTTng-UST Python
675 By default, LTTng-UST libraries are installed to
676 dir:{/usr/local/lib}, which is the de facto directory in which to
677 keep self-compiled and third-party libraries.
679 When <<building-tracepoint-providers-and-user-application,linking an
680 instrumented user application with `liblttng-ust`>>:
682 * Append `/usr/local/lib` to the env:LD_LIBRARY_PATH environment
684 * Pass the `-L/usr/local/lib` and `-Wl,-rpath,/usr/local/lib` options to
685 man:gcc(1), man:g++(1), or man:clang(1).
689 . Download, build, and install the latest LTTng-tools{nbsp}{revision}:
695 wget http://lttng.org/files/lttng-tools/lttng-tools-latest-2.11.tar.bz2 &&
696 tar -xf lttng-tools-latest-2.11.tar.bz2 &&
697 cd lttng-tools-2.11.* &&
705 TIP: The https://github.com/eepp/vlttng[vlttng tool] can do all the
706 previous steps automatically for a given version of LTTng and confine
707 the installed files in a specific directory. This can be useful to test
708 LTTng without installing it on your system.
714 This is a short guide to get started quickly with LTTng kernel and user
717 Before you follow this guide, make sure to <<installing-lttng,install>>
720 This tutorial walks you through the steps to:
722 . <<tracing-the-linux-kernel,Trace the Linux kernel>>.
723 . <<tracing-your-own-user-application,Trace a user application>> written
725 . <<viewing-and-analyzing-your-traces,View and analyze the
729 [[tracing-the-linux-kernel]]
730 === Trace the Linux kernel
732 The following command lines start with the `#` prompt because you need
733 root privileges to trace the Linux kernel. You can also trace the kernel
734 as a regular user if your Unix user is a member of the
735 <<tracing-group,tracing group>>.
737 . Create a <<tracing-session,tracing session>> which writes its traces
738 to dir:{/tmp/my-kernel-trace}:
743 # lttng create my-kernel-session --output=/tmp/my-kernel-trace
747 . List the available kernel tracepoints and system calls:
752 # lttng list --kernel
753 # lttng list --kernel --syscall
757 . Create <<event,event rules>> which match the desired instrumentation
758 point names, for example the `sched_switch` and `sched_process_fork`
759 tracepoints, and the man:open(2) and man:close(2) system calls:
764 # lttng enable-event --kernel sched_switch,sched_process_fork
765 # lttng enable-event --kernel --syscall open,close
769 You can also create an event rule which matches _all_ the Linux kernel
770 tracepoints (this will generate a lot of data when tracing):
775 # lttng enable-event --kernel --all
779 . <<basic-tracing-session-control,Start tracing>>:
788 . Do some operation on your system for a few seconds. For example,
789 load a website, or list the files of a directory.
790 . <<creating-destroying-tracing-sessions,Destroy>> the current
800 The man:lttng-destroy(1) command does not destroy the trace data; it
801 only destroys the state of the tracing session.
803 The man:lttng-destroy(1) command also runs the man:lttng-stop(1) command
804 implicitly (see <<basic-tracing-session-control,Start and stop a tracing
805 session>>). You need to stop tracing to make LTTng flush the remaining
806 trace data and make the trace readable.
808 . For the sake of this example, make the recorded trace accessible to
814 # chown -R $(whoami) /tmp/my-kernel-trace
818 See <<viewing-and-analyzing-your-traces,View and analyze the
819 recorded events>> to view the recorded events.
822 [[tracing-your-own-user-application]]
823 === Trace a user application
825 This section steps you through a simple example to trace a
826 _Hello world_ program written in C.
828 To create the traceable user application:
830 . Create the tracepoint provider header file, which defines the
831 tracepoints and the events they can generate:
837 #undef TRACEPOINT_PROVIDER
838 #define TRACEPOINT_PROVIDER hello_world
840 #undef TRACEPOINT_INCLUDE
841 #define TRACEPOINT_INCLUDE "./hello-tp.h"
843 #if !defined(_HELLO_TP_H) || defined(TRACEPOINT_HEADER_MULTI_READ)
846 #include <lttng/tracepoint.h>
856 ctf_string(my_string_field, my_string_arg)
857 ctf_integer(int, my_integer_field, my_integer_arg)
861 #endif /* _HELLO_TP_H */
863 #include <lttng/tracepoint-event.h>
867 . Create the tracepoint provider package source file:
873 #define TRACEPOINT_CREATE_PROBES
874 #define TRACEPOINT_DEFINE
876 #include "hello-tp.h"
880 . Build the tracepoint provider package:
885 $ gcc -c -I. hello-tp.c
889 . Create the _Hello World_ application source file:
896 #include "hello-tp.h"
898 int main(int argc, char *argv[])
902 puts("Hello, World!\nPress Enter to continue...");
905 * The following getchar() call is only placed here for the purpose
906 * of this demonstration, to pause the application in order for
907 * you to have time to list its tracepoints. It is not
913 * A tracepoint() call.
915 * Arguments, as defined in hello-tp.h:
917 * 1. Tracepoint provider name (required)
918 * 2. Tracepoint name (required)
919 * 3. my_integer_arg (first user-defined argument)
920 * 4. my_string_arg (second user-defined argument)
922 * Notice the tracepoint provider and tracepoint names are
923 * NOT strings: they are in fact parts of variables that the
924 * macros in hello-tp.h create.
926 tracepoint(hello_world, my_first_tracepoint, 23, "hi there!");
928 for (x = 0; x < argc; ++x) {
929 tracepoint(hello_world, my_first_tracepoint, x, argv[x]);
932 puts("Quitting now!");
933 tracepoint(hello_world, my_first_tracepoint, x * x, "x^2");
940 . Build the application:
949 . Link the application with the tracepoint provider package,
950 `liblttng-ust`, and `libdl`:
955 $ gcc -o hello hello.o hello-tp.o -llttng-ust -ldl
959 Here's the whole build process:
962 .User space tracing tutorial's build steps.
963 image::ust-flow.png[]
965 To trace the user application:
967 . Run the application with a few arguments:
972 $ ./hello world and beyond
981 Press Enter to continue...
985 . Start an LTTng <<lttng-sessiond,session daemon>>:
990 $ lttng-sessiond --daemonize
994 Note that a session daemon might already be running, for example as
995 a service that the distribution's service manager started.
997 . List the available user space tracepoints:
1002 $ lttng list --userspace
1006 You see the `hello_world:my_first_tracepoint` tracepoint listed
1007 under the `./hello` process.
1009 . Create a <<tracing-session,tracing session>>:
1014 $ lttng create my-user-space-session
1018 . Create an <<event,event rule>> which matches the
1019 `hello_world:my_first_tracepoint` event name:
1024 $ lttng enable-event --userspace hello_world:my_first_tracepoint
1028 . <<basic-tracing-session-control,Start tracing>>:
1037 . Go back to the running `hello` application and press Enter. The
1038 program executes all `tracepoint()` instrumentation points and exits.
1039 . <<creating-destroying-tracing-sessions,Destroy>> the current
1049 The man:lttng-destroy(1) command does not destroy the trace data; it
1050 only destroys the state of the tracing session.
1052 The man:lttng-destroy(1) command also runs the man:lttng-stop(1) command
1053 implicitly (see <<basic-tracing-session-control,Start and stop a tracing
1054 session>>). You need to stop tracing to make LTTng flush the remaining
1055 trace data and make the trace readable.
1057 By default, LTTng saves the traces in
1058 +$LTTNG_HOME/lttng-traces/__name__-__date__-__time__+,
1059 where +__name__+ is the tracing session name. The
1060 env:LTTNG_HOME environment variable defaults to `$HOME` if not set.
1062 See <<viewing-and-analyzing-your-traces,View and analyze the
1063 recorded events>> to view the recorded events.
1066 [[viewing-and-analyzing-your-traces]]
1067 === View and analyze the recorded events
1069 Once you have completed the <<tracing-the-linux-kernel,Trace the Linux
1070 kernel>> and <<tracing-your-own-user-application,Trace a user
1071 application>> tutorials, you can inspect the recorded events.
1073 There are many tools you can use to read LTTng traces:
1075 * **cmd:babeltrace** is a command-line utility which converts trace
1076 formats; it supports the format that LTTng produces, CTF, as well as a
1077 basic text output which can be ++grep++ed. The cmd:babeltrace command
1078 is part of the http://diamon.org/babeltrace[Babeltrace] project.
1079 * Babeltrace also includes
1080 **https://www.python.org/[Python{nbsp}3] bindings** so
1081 that you can easily open and read an LTTng trace with your own script,
1082 benefiting from the power of Python.
1083 * http://tracecompass.org/[**Trace Compass**]
1084 is a graphical user interface for viewing and analyzing any type of
1085 logs or traces, including LTTng's.
1086 * https://github.com/lttng/lttng-analyses[**LTTng analyses**] is a
1087 project which includes many high-level analyses of LTTng kernel
1088 traces, like scheduling statistics, interrupt frequency distribution,
1089 top CPU usage, and more.
1091 NOTE: This section assumes that LTTng saved the traces it recorded
1092 during the previous tutorials to their default location, in the
1093 dir:{$LTTNG_HOME/lttng-traces} directory. The env:LTTNG_HOME
1094 environment variable defaults to `$HOME` if not set.
1097 [[viewing-and-analyzing-your-traces-bt]]
1098 ==== Use the cmd:babeltrace command-line tool
1100 The simplest way to list all the recorded events of a trace is to pass
1101 its path to cmd:babeltrace with no options:
1105 $ babeltrace ~/lttng-traces/my-user-space-session*
1108 cmd:babeltrace finds all traces recursively within the given path and
1109 prints all their events, merging them in chronological order.
1111 You can pipe the output of cmd:babeltrace into a tool like man:grep(1) for
1116 $ babeltrace /tmp/my-kernel-trace | grep _switch
1119 You can pipe the output of cmd:babeltrace into a tool like man:wc(1) to
1120 count the recorded events:
1124 $ babeltrace /tmp/my-kernel-trace | grep _open | wc --lines
1128 [[viewing-and-analyzing-your-traces-bt-python]]
1129 ==== Use the Babeltrace{nbsp}1 Python bindings
1131 The <<viewing-and-analyzing-your-traces-bt,text output of cmd:babeltrace>>
1132 is useful to isolate events by simple matching using man:grep(1) and
1133 similar utilities. However, more elaborate filters, such as keeping only
1134 event records with a field value falling within a specific range, are
1135 not trivial to write using a shell. Moreover, reductions and even the
1136 most basic computations involving multiple event records are virtually
1137 impossible to implement.
1139 Fortunately, Babeltrace{nbsp}1 ships with Python{nbsp}3 bindings which
1140 makes it easy to read the event records of an LTTng trace sequentially
1141 and compute the desired information.
1143 The following script accepts an LTTng Linux kernel trace path as its
1144 first argument and prints the short names of the top five running
1145 processes on CPU{nbsp}0 during the whole trace:
1150 from collections import Counter
1156 if len(sys.argv) != 2:
1157 msg = 'Usage: python3 {} TRACEPATH'.format(sys.argv[0])
1158 print(msg, file=sys.stderr)
1161 # A trace collection contains one or more traces
1162 col = babeltrace.TraceCollection()
1164 # Add the trace provided by the user (LTTng traces always have
1166 if col.add_trace(sys.argv[1], 'ctf') is None:
1167 raise RuntimeError('Cannot add trace')
1169 # This counter dict contains execution times:
1171 # task command name -> total execution time (ns)
1172 exec_times = Counter()
1174 # This contains the last `sched_switch` timestamp
1178 for event in col.events:
1179 # Keep only `sched_switch` events
1180 if event.name != 'sched_switch':
1183 # Keep only events which happened on CPU 0
1184 if event['cpu_id'] != 0:
1188 cur_ts = event.timestamp
1194 # Previous task command (short) name
1195 prev_comm = event['prev_comm']
1197 # Initialize entry in our dict if not yet done
1198 if prev_comm not in exec_times:
1199 exec_times[prev_comm] = 0
1201 # Compute previous command execution time
1202 diff = cur_ts - last_ts
1204 # Update execution time of this command
1205 exec_times[prev_comm] += diff
1207 # Update last timestamp
1211 for name, ns in exec_times.most_common(5):
1213 print('{:20}{} s'.format(name, s))
1218 if __name__ == '__main__':
1219 sys.exit(0 if top5proc() else 1)
1226 $ python3 top5proc.py /tmp/my-kernel-trace/kernel
1232 swapper/0 48.607245889 s
1233 chromium 7.192738188 s
1234 pavucontrol 0.709894415 s
1235 Compositor 0.660867933 s
1236 Xorg.bin 0.616753786 s
1239 Note that `swapper/0` is the "idle" process of CPU{nbsp}0 on Linux;
1240 since we weren't using the CPU that much when tracing, its first
1241 position in the list makes sense.
1245 == [[understanding-lttng]]Core concepts
1247 From a user's perspective, the LTTng system is built on a few concepts,
1248 or objects, on which the <<lttng-cli,cmd:lttng command-line tool>>
1249 operates by sending commands to the <<lttng-sessiond,session daemon>>.
1250 Understanding how those objects relate to eachother is key in mastering
1253 The core concepts are:
1255 * <<tracing-session,Tracing session>>
1256 * <<domain,Tracing domain>>
1257 * <<channel,Channel and ring buffer>>
1258 * <<"event","Instrumentation point, event rule, event, and event record">>
1264 A _tracing session_ is a stateful dialogue between you and
1265 a <<lttng-sessiond,session daemon>>. You can
1266 <<creating-destroying-tracing-sessions,create a new tracing
1267 session>> with the `lttng create` command.
1269 Anything that you do when you control LTTng tracers happens within a
1270 tracing session. In particular, a tracing session:
1273 * Has its own set of trace files.
1274 * Has its own state of activity (started or stopped).
1275 * Has its own <<tracing-session-mode,mode>> (local, network streaming,
1277 * Has its own <<channel,channels>> to which are associated their own
1278 <<event,event rules>>.
1281 .A _tracing session_ contains <<channel,channels>> that are members of <<domain,tracing domains>> and contain <<event,event rules>>.
1282 image::concepts.png[]
1284 Those attributes and objects are completely isolated between different
1287 A tracing session is analogous to a cash machine session:
1288 the operations you do on the banking system through the cash machine do
1289 not alter the data of other users of the same system. In the case of
1290 the cash machine, a session lasts as long as your bank card is inside.
1291 In the case of LTTng, a tracing session lasts from the `lttng create`
1292 command to the `lttng destroy` command.
1295 .Each Unix user has its own set of tracing sessions.
1296 image::many-sessions.png[]
1299 [[tracing-session-mode]]
1300 ==== Tracing session mode
1302 LTTng can send the generated trace data to different locations. The
1303 _tracing session mode_ dictates where to send it. The following modes
1304 are available in LTTng{nbsp}{revision}:
1307 LTTng writes the traces to the file system of the machine it traces
1310 Network streaming mode::
1311 LTTng sends the traces over the network to a
1312 <<lttng-relayd,relay daemon>> running on a remote system.
1315 LTTng does not write the traces by default. Instead, you can request
1316 LTTng to <<taking-a-snapshot,take a snapshot>>, that is, a copy of the
1317 tracing session's current sub-buffers, and to write it to the
1318 target's file system or to send it over the network to a
1319 <<lttng-relayd,relay daemon>> running on a remote system.
1322 This mode is similar to the network streaming mode, but a live
1323 trace viewer can connect to the distant relay daemon to
1324 <<lttng-live,view event records as LTTng generates them>>.
1330 A _tracing domain_ is a namespace for event sources. A tracing domain
1331 has its own properties and features.
1333 There are currently five available tracing domains:
1337 * `java.util.logging` (JUL)
1341 You must specify a tracing domain when using some commands to avoid
1342 ambiguity. For example, since all the domains support named tracepoints
1343 as event sources (instrumentation points that you manually insert in the
1344 source code), you need to specify a tracing domain when
1345 <<enabling-disabling-events,creating an event rule>> because all the
1346 tracing domains could have tracepoints with the same names.
1348 You can create <<channel,channels>> in the Linux kernel and user space
1349 tracing domains. The other tracing domains have a single default
1354 === Channel and ring buffer
1356 A _channel_ is an object which is responsible for a set of ring buffers.
1357 Each ring buffer is divided into multiple sub-buffers. When an LTTng
1358 tracer emits an event, it can record it to one or more
1359 sub-buffers. The attributes of a channel determine what to do when
1360 there's no space left for a new event record because all sub-buffers
1361 are full, where to send a full sub-buffer, and other behaviours.
1363 A channel is always associated to a <<domain,tracing domain>>. The
1364 `java.util.logging` (JUL), log4j, and Python tracing domains each have
1365 a default channel which you cannot configure.
1367 A channel also owns <<event,event rules>>. When an LTTng tracer emits
1368 an event, it records it to the sub-buffers of all
1369 the enabled channels with a satisfied event rule, as long as those
1370 channels are part of active <<tracing-session,tracing sessions>>.
1373 [[channel-buffering-schemes]]
1374 ==== Per-user vs. per-process buffering schemes
1376 A channel has at least one ring buffer _per CPU_. LTTng always
1377 records an event to the ring buffer associated to the CPU on which it
1380 Two _buffering schemes_ are available when you
1381 <<enabling-disabling-channels,create a channel>> in the
1382 user space <<domain,tracing domain>>:
1384 Per-user buffering::
1385 Allocate one set of ring buffers--one per CPU--shared by all the
1386 instrumented processes of each Unix user.
1390 .Per-user buffering scheme.
1391 image::per-user-buffering.png[]
1394 Per-process buffering::
1395 Allocate one set of ring buffers--one per CPU--for each
1396 instrumented process.
1400 .Per-process buffering scheme.
1401 image::per-process-buffering.png[]
1404 The per-process buffering scheme tends to consume more memory than the
1405 per-user option because systems generally have more instrumented
1406 processes than Unix users running instrumented processes. However, the
1407 per-process buffering scheme ensures that one process having a high
1408 event throughput won't fill all the shared sub-buffers of the same
1411 The Linux kernel tracing domain has only one available buffering scheme
1412 which is to allocate a single set of ring buffers for the whole system.
1413 This scheme is similar to the per-user option, but with a single, global
1414 user "running" the kernel.
1417 [[channel-overwrite-mode-vs-discard-mode]]
1418 ==== Overwrite vs. discard event record loss modes
1420 When an event occurs, LTTng records it to a specific sub-buffer (yellow
1421 arc in the following animations) of a specific channel's ring buffer.
1422 When there's no space left in a sub-buffer, the tracer marks it as
1423 consumable (red) and another, empty sub-buffer starts receiving the
1424 following event records. A <<lttng-consumerd,consumer daemon>>
1425 eventually consumes the marked sub-buffer (returns to white).
1428 [role="docsvg-channel-subbuf-anim"]
1433 In an ideal world, sub-buffers are consumed faster than they are filled,
1434 as it is the case in the previous animation. In the real world,
1435 however, all sub-buffers can be full at some point, leaving no space to
1436 record the following events.
1438 By default, LTTng-modules and LTTng-UST are _non-blocking_ tracers: when
1439 no empty sub-buffer is available, it is acceptable to lose event records
1440 when the alternative would be to cause substantial delays in the
1441 instrumented application's execution. LTTng privileges performance over
1442 integrity; it aims at perturbing the target system as little as possible
1443 in order to make tracing of subtle race conditions and rare interrupt
1446 Since LTTng{nbsp}2.10, the LTTng user space tracer, LTTng-UST, supports
1447 a _blocking mode_. See the <<blocking-timeout-example,blocking timeout
1448 example>> to learn how to use the blocking mode.
1450 When it comes to losing event records because no empty sub-buffer is
1451 available, or because the <<opt-blocking-timeout,blocking timeout>> is
1452 reached, the channel's _event record loss mode_ determines what to do.
1453 The available event record loss modes are:
1456 Drop the newest event records until the tracer releases a sub-buffer.
1458 This is the only available mode when you specify a
1459 <<opt-blocking-timeout,blocking timeout>>.
1462 Clear the sub-buffer containing the oldest event records and start
1463 writing the newest event records there.
1465 This mode is sometimes called _flight recorder mode_ because it's
1467 https://en.wikipedia.org/wiki/Flight_recorder[flight recorder]:
1468 always keep a fixed amount of the latest data.
1470 Which mechanism you should choose depends on your context: prioritize
1471 the newest or the oldest event records in the ring buffer?
1473 Beware that, in overwrite mode, the tracer abandons a _whole sub-buffer_
1474 as soon as a there's no space left for a new event record, whereas in
1475 discard mode, the tracer only discards the event record that doesn't
1478 In discard mode, LTTng increments a count of lost event records when an
1479 event record is lost and saves this count to the trace. In overwrite
1480 mode, since LTTng{nbsp}2.8, LTTng increments a count of lost sub-buffers
1481 when a sub-buffer is lost and saves this count to the trace. In this
1482 mode, LTTng does not write to the trace the exact number of lost event
1483 records in those lost sub-buffers. Trace analyses can use the trace's
1484 saved discarded event record and sub-buffer counts to decide whether or
1485 not to perform the analyses even if trace data is known to be missing.
1487 There are a few ways to decrease your probability of losing event
1489 <<channel-subbuf-size-vs-subbuf-count,Sub-buffer count and size>> shows
1490 how you can fine-tune the sub-buffer count and size of a channel to
1491 virtually stop losing event records, though at the cost of greater
1495 [[channel-subbuf-size-vs-subbuf-count]]
1496 ==== Sub-buffer count and size
1498 When you <<enabling-disabling-channels,create a channel>>, you can
1499 set its number of sub-buffers and their size.
1501 Note that there is noticeable CPU overhead introduced when
1502 switching sub-buffers (marking a full one as consumable and switching
1503 to an empty one for the following events to be recorded). Knowing this,
1504 the following list presents a few practical situations along with how
1505 to configure the sub-buffer count and size for them:
1507 * **High event throughput**: In general, prefer bigger sub-buffers to
1508 lower the risk of losing event records.
1510 Having bigger sub-buffers also ensures a lower
1511 <<channel-switch-timer,sub-buffer switching frequency>>.
1513 The number of sub-buffers is only meaningful if you create the channel
1514 in overwrite mode: in this case, if a sub-buffer overwrite happens, the
1515 other sub-buffers are left unaltered.
1517 * **Low event throughput**: In general, prefer smaller sub-buffers
1518 since the risk of losing event records is low.
1520 Because events occur less frequently, the sub-buffer switching frequency
1521 should remain low and thus the tracer's overhead should not be a
1524 * **Low memory system**: If your target system has a low memory
1525 limit, prefer fewer first, then smaller sub-buffers.
1527 Even if the system is limited in memory, you want to keep the
1528 sub-buffers as big as possible to avoid a high sub-buffer switching
1531 Note that LTTng uses http://diamon.org/ctf/[CTF] as its trace format,
1532 which means event data is very compact. For example, the average
1533 LTTng kernel event record weights about 32{nbsp}bytes. Thus, a
1534 sub-buffer size of 1{nbsp}MiB is considered big.
1536 The previous situations highlight the major trade-off between a few big
1537 sub-buffers and more, smaller sub-buffers: sub-buffer switching
1538 frequency vs. how much data is lost in overwrite mode. Assuming a
1539 constant event throughput and using the overwrite mode, the two
1540 following configurations have the same ring buffer total size:
1543 [role="docsvg-channel-subbuf-size-vs-count-anim"]
1548 * **Two sub-buffers of 4{nbsp}MiB each**: Expect a very low sub-buffer
1549 switching frequency, but if a sub-buffer overwrite happens, half of
1550 the event records so far (4{nbsp}MiB) are definitely lost.
1551 * **Eight sub-buffers of 1{nbsp}MiB each**: Expect four times the tracer's
1552 overhead as the previous configuration, but if a sub-buffer
1553 overwrite happens, only the eighth of event records so far are
1556 In discard mode, the sub-buffers count parameter is pointless: use two
1557 sub-buffers and set their size according to the requirements of your
1561 [[channel-switch-timer]]
1562 ==== Switch timer period
1564 The _switch timer period_ is an important configurable attribute of
1565 a channel to ensure periodic sub-buffer flushing.
1567 When the _switch timer_ expires, a sub-buffer switch happens. You can
1568 set the switch timer period attribute when you
1569 <<enabling-disabling-channels,create a channel>> to ensure that LTTng
1570 consumes and commits trace data to trace files or to a distant relay
1571 daemon periodically in case of a low event throughput.
1574 [role="docsvg-channel-switch-timer"]
1579 This attribute is also convenient when you use big sub-buffers to cope
1580 with a sporadic high event throughput, even if the throughput is
1584 [[channel-read-timer]]
1585 ==== Read timer period
1587 By default, the LTTng tracers use a notification mechanism to signal a
1588 full sub-buffer so that a consumer daemon can consume it. When such
1589 notifications must be avoided, for example in real-time applications,
1590 you can use the channel's _read timer_ instead. When the read timer
1591 fires, the <<lttng-consumerd,consumer daemon>> checks for full,
1592 consumable sub-buffers.
1595 [[tracefile-rotation]]
1596 ==== Trace file count and size
1598 By default, trace files can grow as large as needed. You can set the
1599 maximum size of each trace file that a channel writes when you
1600 <<enabling-disabling-channels,create a channel>>. When the size of
1601 a trace file reaches the channel's fixed maximum size, LTTng creates
1602 another file to contain the next event records. LTTng appends a file
1603 count to each trace file name in this case.
1605 If you set the trace file size attribute when you create a channel, the
1606 maximum number of trace files that LTTng creates is _unlimited_ by
1607 default. To limit them, you can also set a maximum number of trace
1608 files. When the number of trace files reaches the channel's fixed
1609 maximum count, the oldest trace file is overwritten. This mechanism is
1610 called _trace file rotation_.
1614 Even if you don't limit the trace file count, you cannot assume that
1615 LTTng doesn't manage any trace file.
1617 In other words, there is no safe way to know if LTTng still holds a
1618 given trace file open with the trace file rotation feature.
1620 The only way to obtain an unmanaged, self-contained LTTng trace before
1621 you <<creating-destroying-tracing-sessions,destroy>> the tracing session
1622 is with the <<session-rotation,tracing session rotation>> feature
1623 (available since LTTng{nbsp}2.11).
1628 === Instrumentation point, event rule, event, and event record
1630 An _event rule_ is a set of conditions which must be **all** satisfied
1631 for LTTng to record an occuring event.
1633 You set the conditions when you <<enabling-disabling-events,create
1636 You always attach an event rule to a <<channel,channel>> when you create
1639 When an event passes the conditions of an event rule, LTTng records it
1640 in one of the attached channel's sub-buffers.
1642 The available conditions, as of LTTng{nbsp}{revision}, are:
1644 * The event rule _is enabled_.
1645 * The instrumentation point's type _is{nbsp}T_.
1646 * The instrumentation point's name (sometimes called _event name_)
1647 _matches{nbsp}N_, but _is not{nbsp}E_.
1648 * The instrumentation point's log level _is as severe as{nbsp}L_, or
1649 _is exactly{nbsp}L_.
1650 * The fields of the event's payload _satisfy_ a filter
1651 expression{nbsp}__F__.
1653 As you can see, all the conditions but the dynamic filter are related to
1654 the event rule's status or to the instrumentation point, not to the
1655 occurring events. This is why, without a filter, checking if an event
1656 passes an event rule is not a dynamic task: when you create or modify an
1657 event rule, all the tracers of its tracing domain enable or disable the
1658 instrumentation points themselves once. This is possible because the
1659 attributes of an instrumentation point (type, name, and log level) are
1660 defined statically. In other words, without a dynamic filter, the tracer
1661 _does not evaluate_ the arguments of an instrumentation point unless it
1662 matches an enabled event rule.
1664 Note that, for LTTng to record an event, the <<channel,channel>> to
1665 which a matching event rule is attached must also be enabled, and the
1666 <<tracing-session,tracing session>> owning this channel must be active
1670 .Logical path from an instrumentation point to an event record.
1671 image::event-rule.png[]
1673 .Event, event record, or event rule?
1675 With so many similar terms, it's easy to get confused.
1677 An **event** is the consequence of the execution of an _instrumentation
1678 point_, like a tracepoint that you manually place in some source code,
1679 or a Linux kernel kprobe. An event is said to _occur_ at a specific
1680 time. Different actions can be taken upon the occurrence of an event,
1681 like record the event's payload to a buffer.
1683 An **event record** is the representation of an event in a sub-buffer. A
1684 tracer is responsible for capturing the payload of an event, current
1685 context variables, the event's ID, and the event's timestamp. LTTng
1686 can append this sub-buffer to a trace file.
1688 An **event rule** is a set of conditions which must _all_ be satisfied
1689 for LTTng to record an occuring event. Events still occur without
1690 satisfying event rules, but LTTng does not record them.
1695 == Components of noch:{LTTng}
1697 The second _T_ in _LTTng_ stands for _toolkit_: it would be wrong
1698 to call LTTng a simple _tool_ since it is composed of multiple
1699 interacting components. This section describes those components,
1700 explains their respective roles, and shows how they connect together to
1701 form the LTTng ecosystem.
1703 The following diagram shows how the most important components of LTTng
1704 interact with user applications, the Linux kernel, and you:
1707 .Control and trace data paths between LTTng components.
1708 image::plumbing.png[]
1710 The LTTng project incorporates:
1712 * **LTTng-tools**: Libraries and command-line interface to
1713 control tracing sessions.
1714 ** <<lttng-sessiond,Session daemon>> (man:lttng-sessiond(8)).
1715 ** <<lttng-consumerd,Consumer daemon>> (cmd:lttng-consumerd).
1716 ** <<lttng-relayd,Relay daemon>> (man:lttng-relayd(8)).
1717 ** <<liblttng-ctl-lttng,Tracing control library>> (`liblttng-ctl`).
1718 ** <<lttng-cli,Tracing control command-line tool>> (man:lttng(1)).
1719 * **LTTng-UST**: Libraries and Java/Python packages to trace user
1721 ** <<lttng-ust,User space tracing library>> (`liblttng-ust`) and its
1722 headers to instrument and trace any native user application.
1723 ** <<prebuilt-ust-helpers,Preloadable user space tracing helpers>>:
1724 *** `liblttng-ust-libc-wrapper`
1725 *** `liblttng-ust-pthread-wrapper`
1726 *** `liblttng-ust-cyg-profile`
1727 *** `liblttng-ust-cyg-profile-fast`
1728 *** `liblttng-ust-dl`
1729 ** User space tracepoint provider source files generator command-line
1730 tool (man:lttng-gen-tp(1)).
1731 ** <<lttng-ust-agents,LTTng-UST Java agent>> to instrument and trace
1732 Java applications using `java.util.logging` or
1733 Apache log4j{nbsp}1.2 logging.
1734 ** <<lttng-ust-agents,LTTng-UST Python agent>> to instrument
1735 Python applications using the standard `logging` package.
1736 * **LTTng-modules**: <<lttng-modules,Linux kernel modules>> to trace
1738 ** LTTng kernel tracer module.
1739 ** Tracing ring buffer kernel modules.
1740 ** Probe kernel modules.
1741 ** LTTng logger kernel module.
1745 === Tracing control command-line interface
1748 .The tracing control command-line interface.
1749 image::plumbing-lttng-cli.png[]
1751 The _man:lttng(1) command-line tool_ is the standard user interface to
1752 control LTTng <<tracing-session,tracing sessions>>. The cmd:lttng tool
1753 is part of LTTng-tools.
1755 The cmd:lttng tool is linked with
1756 <<liblttng-ctl-lttng,`liblttng-ctl`>> to communicate with
1757 one or more <<lttng-sessiond,session daemons>> behind the scenes.
1759 The cmd:lttng tool has a Git-like interface:
1763 $ lttng <GENERAL OPTIONS> <COMMAND> <COMMAND OPTIONS>
1766 The <<controlling-tracing,Tracing control>> section explores the
1767 available features of LTTng using the cmd:lttng tool.
1770 [[liblttng-ctl-lttng]]
1771 === Tracing control library
1774 .The tracing control library.
1775 image::plumbing-liblttng-ctl.png[]
1777 The _LTTng control library_, `liblttng-ctl`, is used to communicate
1778 with a <<lttng-sessiond,session daemon>> using a C API that hides the
1779 underlying protocol's details. `liblttng-ctl` is part of LTTng-tools.
1781 The <<lttng-cli,cmd:lttng command-line tool>>
1782 is linked with `liblttng-ctl`.
1784 You can use `liblttng-ctl` in C or $$C++$$ source code by including its
1789 #include <lttng/lttng.h>
1792 Some objects are referenced by name (C string), such as tracing
1793 sessions, but most of them require to create a handle first using
1794 `lttng_create_handle()`.
1796 As of LTTng{nbsp}{revision}, the best available developer documentation for
1797 `liblttng-ctl` is its installed header files. Every function and structure is
1798 thoroughly documented.
1802 === User space tracing library
1805 .The user space tracing library.
1806 image::plumbing-liblttng-ust.png[]
1808 The _user space tracing library_, `liblttng-ust` (see man:lttng-ust(3)),
1809 is the LTTng user space tracer. It receives commands from a
1810 <<lttng-sessiond,session daemon>>, for example to
1811 enable and disable specific instrumentation points, and writes event
1812 records to ring buffers shared with a
1813 <<lttng-consumerd,consumer daemon>>.
1814 `liblttng-ust` is part of LTTng-UST.
1816 Public C header files are installed beside `liblttng-ust` to
1817 instrument any <<c-application,C or $$C++$$ application>>.
1819 <<lttng-ust-agents,LTTng-UST agents>>, which are regular Java and Python
1820 packages, use their own library providing tracepoints which is
1821 linked with `liblttng-ust`.
1823 An application or library does not have to initialize `liblttng-ust`
1824 manually: its constructor does the necessary tasks to properly register
1825 to a session daemon. The initialization phase also enables the
1826 instrumentation points matching the <<event,event rules>> that you
1830 [[lttng-ust-agents]]
1831 === User space tracing agents
1834 .The user space tracing agents.
1835 image::plumbing-lttng-ust-agents.png[]
1837 The _LTTng-UST Java and Python agents_ are regular Java and Python
1838 packages which add LTTng tracing capabilities to the
1839 native logging frameworks. The LTTng-UST agents are part of LTTng-UST.
1841 In the case of Java, the
1842 https://docs.oracle.com/javase/7/docs/api/java/util/logging/package-summary.html[`java.util.logging`
1843 core logging facilities] and
1844 https://logging.apache.org/log4j/1.2/[Apache log4j{nbsp}1.2] are supported.
1845 Note that Apache Log4{nbsp}2 is not supported.
1847 In the case of Python, the standard
1848 https://docs.python.org/3/library/logging.html[`logging`] package
1849 is supported. Both Python{nbsp}2 and Python{nbsp}3 modules can import the
1850 LTTng-UST Python agent package.
1852 The applications using the LTTng-UST agents are in the
1853 `java.util.logging` (JUL),
1854 log4j, and Python <<domain,tracing domains>>.
1856 Both agents use the same mechanism to trace the log statements. When an
1857 agent initializes, it creates a log handler that attaches to the root
1858 logger. The agent also registers to a <<lttng-sessiond,session daemon>>.
1859 When the application executes a log statement, the root logger passes it
1860 to the agent's log handler. The agent's log handler calls a native
1861 function in a tracepoint provider package shared library linked with
1862 <<lttng-ust,`liblttng-ust`>>, passing the formatted log message and
1863 other fields, like its logger name and its log level. This native
1864 function contains a user space instrumentation point, hence tracing the
1867 The log level condition of an
1868 <<event,event rule>> is considered when tracing
1869 a Java or a Python application, and it's compatible with the standard
1870 JUL, log4j, and Python log levels.
1874 === LTTng kernel modules
1877 .The LTTng kernel modules.
1878 image::plumbing-lttng-modules.png[]
1880 The _LTTng kernel modules_ are a set of Linux kernel modules
1881 which implement the kernel tracer of the LTTng project. The LTTng
1882 kernel modules are part of LTTng-modules.
1884 The LTTng kernel modules include:
1886 * A set of _probe_ modules.
1888 Each module attaches to a specific subsystem
1889 of the Linux kernel using its tracepoint instrument points. There are
1890 also modules to attach to the entry and return points of the Linux
1891 system call functions.
1893 * _Ring buffer_ modules.
1895 A ring buffer implementation is provided as kernel modules. The LTTng
1896 kernel tracer writes to the ring buffer; a
1897 <<lttng-consumerd,consumer daemon>> reads from the ring buffer.
1899 * The _LTTng kernel tracer_ module.
1900 * The _LTTng logger_ module.
1902 The LTTng logger module implements the special path:{/proc/lttng-logger}
1903 (and path:{/dev/lttng-logger} since LTTng{nbsp}2.11) files so that any
1904 executable can generate LTTng events by opening and writing to those
1907 See <<proc-lttng-logger-abi,LTTng logger>>.
1909 Generally, you do not have to load the LTTng kernel modules manually
1910 (using man:modprobe(8), for example): a root <<lttng-sessiond,session
1911 daemon>> loads the necessary modules when starting. If you have extra
1912 probe modules, you can specify to load them to the session daemon on
1915 The LTTng kernel modules are installed in
1916 +/usr/lib/modules/__release__/extra+ by default, where +__release__+ is
1917 the kernel release (see `uname --kernel-release`).
1924 .The session daemon.
1925 image::plumbing-sessiond.png[]
1927 The _session daemon_, man:lttng-sessiond(8), is a daemon responsible for
1928 managing tracing sessions and for controlling the various components of
1929 LTTng. The session daemon is part of LTTng-tools.
1931 The session daemon sends control requests to and receives control
1934 * The <<lttng-ust,user space tracing library>>.
1936 Any instance of the user space tracing library first registers to
1937 a session daemon. Then, the session daemon can send requests to
1938 this instance, such as:
1941 ** Get the list of tracepoints.
1942 ** Share an <<event,event rule>> so that the user space tracing library
1943 can enable or disable tracepoints. Amongst the possible conditions
1944 of an event rule is a filter expression which `liblttng-ust` evalutes
1945 when an event occurs.
1946 ** Share <<channel,channel>> attributes and ring buffer locations.
1949 The session daemon and the user space tracing library use a Unix
1950 domain socket for their communication.
1952 * The <<lttng-ust-agents,user space tracing agents>>.
1954 Any instance of a user space tracing agent first registers to
1955 a session daemon. Then, the session daemon can send requests to
1956 this instance, such as:
1959 ** Get the list of loggers.
1960 ** Enable or disable a specific logger.
1963 The session daemon and the user space tracing agent use a TCP connection
1964 for their communication.
1966 * The <<lttng-modules,LTTng kernel tracer>>.
1967 * The <<lttng-consumerd,consumer daemon>>.
1969 The session daemon sends requests to the consumer daemon to instruct
1970 it where to send the trace data streams, amongst other information.
1972 * The <<lttng-relayd,relay daemon>>.
1974 The session daemon receives commands from the
1975 <<liblttng-ctl-lttng,tracing control library>>.
1977 The root session daemon loads the appropriate
1978 <<lttng-modules,LTTng kernel modules>> on startup. It also spawns
1979 a <<lttng-consumerd,consumer daemon>> as soon as you create
1980 an <<event,event rule>>.
1982 The session daemon does not send and receive trace data: this is the
1983 role of the <<lttng-consumerd,consumer daemon>> and
1984 <<lttng-relayd,relay daemon>>. It does, however, generate the
1985 http://diamon.org/ctf/[CTF] metadata stream.
1987 Each Unix user can have its own session daemon instance. The
1988 tracing sessions which different session daemons manage are completely
1991 The root user's session daemon is the only one which is
1992 allowed to control the LTTng kernel tracer, and its spawned consumer
1993 daemon is the only one which is allowed to consume trace data from the
1994 LTTng kernel tracer. Note, however, that any Unix user which is a member
1995 of the <<tracing-group,tracing group>> is allowed
1996 to create <<channel,channels>> in the
1997 Linux kernel <<domain,tracing domain>>, and thus to trace the Linux
2000 The <<lttng-cli,cmd:lttng command-line tool>> automatically starts a
2001 session daemon when using its `create` command if none is currently
2002 running. You can also start the session daemon manually.
2009 .The consumer daemon.
2010 image::plumbing-consumerd.png[]
2012 The _consumer daemon_, cmd:lttng-consumerd, is a daemon which shares
2013 ring buffers with user applications or with the LTTng kernel modules to
2014 collect trace data and send it to some location (on disk or to a
2015 <<lttng-relayd,relay daemon>> over the network). The consumer daemon
2016 is part of LTTng-tools.
2018 You do not start a consumer daemon manually: a consumer daemon is always
2019 spawned by a <<lttng-sessiond,session daemon>> as soon as you create an
2020 <<event,event rule>>, that is, before you start tracing. When you kill
2021 its owner session daemon, the consumer daemon also exits because it is
2022 the session daemon's child process. Command-line options of
2023 man:lttng-sessiond(8) target the consumer daemon process.
2025 There are up to two running consumer daemons per Unix user, whereas only
2026 one session daemon can run per user. This is because each process can be
2027 either 32-bit or 64-bit: if the target system runs a mixture of 32-bit
2028 and 64-bit processes, it is more efficient to have separate
2029 corresponding 32-bit and 64-bit consumer daemons. The root user is an
2030 exception: it can have up to _three_ running consumer daemons: 32-bit
2031 and 64-bit instances for its user applications, and one more
2032 reserved for collecting kernel trace data.
2040 image::plumbing-relayd.png[]
2042 The _relay daemon_, man:lttng-relayd(8), is a daemon acting as a bridge
2043 between remote session and consumer daemons, local trace files, and a
2044 remote live trace viewer. The relay daemon is part of LTTng-tools.
2046 The main purpose of the relay daemon is to implement a receiver of
2047 <<sending-trace-data-over-the-network,trace data over the network>>.
2048 This is useful when the target system does not have much file system
2049 space to record trace files locally.
2051 The relay daemon is also a server to which a
2052 <<lttng-live,live trace viewer>> can
2053 connect. The live trace viewer sends requests to the relay daemon to
2054 receive trace data as the target system emits events. The
2055 communication protocol is named _LTTng live_; it is used over TCP
2058 Note that you can start the relay daemon on the target system directly.
2059 This is the setup of choice when the use case is to view events as
2060 the target system emits them without the need of a remote system.
2064 == [[using-lttng]]Instrumentation
2066 There are many examples of tracing and monitoring in our everyday life:
2068 * You have access to real-time and historical weather reports and
2069 forecasts thanks to weather stations installed around the country.
2070 * You know your heart is safe thanks to an electrocardiogram.
2071 * You make sure not to drive your car too fast and to have enough fuel
2072 to reach your destination thanks to gauges visible on your dashboard.
2074 All the previous examples have something in common: they rely on
2075 **instruments**. Without the electrodes attached to the surface of your
2076 body's skin, cardiac monitoring is futile.
2078 LTTng, as a tracer, is no different from those real life examples. If
2079 you're about to trace a software system or, in other words, record its
2080 history of execution, you better have **instrumentation points** in the
2081 subject you're tracing, that is, the actual software.
2083 Various ways were developed to instrument a piece of software for LTTng
2084 tracing. The most straightforward one is to manually place
2085 instrumentation points, called _tracepoints_, in the software's source
2086 code. It is also possible to add instrumentation points dynamically in
2087 the Linux kernel <<domain,tracing domain>>.
2089 If you're only interested in tracing the Linux kernel, your
2090 instrumentation needs are probably already covered by LTTng's built-in
2091 <<lttng-modules,Linux kernel tracepoints>>. You may also wish to trace a
2092 user application which is already instrumented for LTTng tracing.
2093 In such cases, you can skip this whole section and read the topics of
2094 the <<controlling-tracing,Tracing control>> section.
2096 Many methods are available to instrument a piece of software for LTTng
2099 * <<c-application,User space instrumentation for C and $$C++$$
2101 * <<prebuilt-ust-helpers,Prebuilt user space tracing helpers>>.
2102 * <<java-application,User space Java agent>>.
2103 * <<python-application,User space Python agent>>.
2104 * <<proc-lttng-logger-abi,LTTng logger>>.
2105 * <<instrumenting-linux-kernel,LTTng kernel tracepoints>>.
2109 === [[cxx-application]]User space instrumentation for C and $$C++$$ applications
2111 The procedure to instrument a C or $$C++$$ user application with
2112 the <<lttng-ust,LTTng user space tracing library>>, `liblttng-ust`, is:
2114 . <<tracepoint-provider,Create the source files of a tracepoint provider
2116 . <<probing-the-application-source-code,Add tracepoints to
2117 the application's source code>>.
2118 . <<building-tracepoint-providers-and-user-application,Build and link
2119 a tracepoint provider package and the user application>>.
2121 If you need quick, man:printf(3)-like instrumentation, you can skip
2122 those steps and use <<tracef,`tracef()`>> or <<tracelog,`tracelog()`>>
2125 IMPORTANT: You need to <<installing-lttng,install>> LTTng-UST to
2126 instrument a user application with `liblttng-ust`.
2129 [[tracepoint-provider]]
2130 ==== Create the source files of a tracepoint provider package
2132 A _tracepoint provider_ is a set of compiled functions which provide
2133 **tracepoints** to an application, the type of instrumentation point
2134 supported by LTTng-UST. Those functions can emit events with
2135 user-defined fields and serialize those events as event records to one
2136 or more LTTng-UST <<channel,channel>> sub-buffers. The `tracepoint()`
2137 macro, which you <<probing-the-application-source-code,insert in a user
2138 application's source code>>, calls those functions.
2140 A _tracepoint provider package_ is an object file (`.o`) or a shared
2141 library (`.so`) which contains one or more tracepoint providers.
2142 Its source files are:
2144 * One or more <<tpp-header,tracepoint provider header>> (`.h`).
2145 * A <<tpp-source,tracepoint provider package source>> (`.c`).
2147 A tracepoint provider package is dynamically linked with `liblttng-ust`,
2148 the LTTng user space tracer, at run time.
2151 .User application linked with `liblttng-ust` and containing a tracepoint provider.
2152 image::ust-app.png[]
2154 NOTE: If you need quick, man:printf(3)-like instrumentation, you can
2155 skip creating and using a tracepoint provider and use
2156 <<tracef,`tracef()`>> or <<tracelog,`tracelog()`>> instead.
2160 ===== Create a tracepoint provider header file template
2162 A _tracepoint provider header file_ contains the tracepoint
2163 definitions of a tracepoint provider.
2165 To create a tracepoint provider header file:
2167 . Start from this template:
2171 .Tracepoint provider header file template (`.h` file extension).
2173 #undef TRACEPOINT_PROVIDER
2174 #define TRACEPOINT_PROVIDER provider_name
2176 #undef TRACEPOINT_INCLUDE
2177 #define TRACEPOINT_INCLUDE "./tp.h"
2179 #if !defined(_TP_H) || defined(TRACEPOINT_HEADER_MULTI_READ)
2182 #include <lttng/tracepoint.h>
2185 * Use TRACEPOINT_EVENT(), TRACEPOINT_EVENT_CLASS(),
2186 * TRACEPOINT_EVENT_INSTANCE(), and TRACEPOINT_LOGLEVEL() here.
2191 #include <lttng/tracepoint-event.h>
2197 * `provider_name` with the name of your tracepoint provider.
2198 * `"tp.h"` with the name of your tracepoint provider header file.
2200 . Below the `#include <lttng/tracepoint.h>` line, put your
2201 <<defining-tracepoints,tracepoint definitions>>.
2203 Your tracepoint provider name must be unique amongst all the possible
2204 tracepoint provider names used on the same target system. We
2205 suggest to include the name of your project or company in the name,
2206 for example, `org_lttng_my_project_tpp`.
2208 TIP: [[lttng-gen-tp]]You can use the man:lttng-gen-tp(1) tool to create
2209 this boilerplate for you. When using cmd:lttng-gen-tp, all you need to
2210 write are the <<defining-tracepoints,tracepoint definitions>>.
2213 [[defining-tracepoints]]
2214 ===== Create a tracepoint definition
2216 A _tracepoint definition_ defines, for a given tracepoint:
2218 * Its **input arguments**. They are the macro parameters that the
2219 `tracepoint()` macro accepts for this particular tracepoint
2220 in the user application's source code.
2221 * Its **output event fields**. They are the sources of event fields
2222 that form the payload of any event that the execution of the
2223 `tracepoint()` macro emits for this particular tracepoint.
2225 You can create a tracepoint definition by using the
2226 `TRACEPOINT_EVENT()` macro below the `#include <lttng/tracepoint.h>`
2228 <<tpp-header,tracepoint provider header file template>>.
2230 The syntax of the `TRACEPOINT_EVENT()` macro is:
2233 .`TRACEPOINT_EVENT()` macro syntax.
2236 /* Tracepoint provider name */
2239 /* Tracepoint name */
2242 /* Input arguments */
2247 /* Output event fields */
2256 * `provider_name` with your tracepoint provider name.
2257 * `tracepoint_name` with your tracepoint name.
2258 * `arguments` with the <<tpp-def-input-args,input arguments>>.
2259 * `fields` with the <<tpp-def-output-fields,output event field>>
2262 This tracepoint emits events named `provider_name:tracepoint_name`.
2265 .Event name's length limitation
2267 The concatenation of the tracepoint provider name and the
2268 tracepoint name must not exceed **254{nbsp}characters**. If it does, the
2269 instrumented application compiles and runs, but LTTng throws multiple
2270 warnings and you could experience serious issues.
2273 [[tpp-def-input-args]]The syntax of the `TP_ARGS()` macro is:
2276 .`TP_ARGS()` macro syntax.
2285 * `type` with the C type of the argument.
2286 * `arg_name` with the argument name.
2288 You can repeat `type` and `arg_name` up to 10{nbsp}times to have
2289 more than one argument.
2291 .`TP_ARGS()` usage with three arguments.
2303 The `TP_ARGS()` and `TP_ARGS(void)` forms are valid to create a
2304 tracepoint definition with no input arguments.
2306 [[tpp-def-output-fields]]The `TP_FIELDS()` macro contains a list of
2307 `ctf_*()` macros. Each `ctf_*()` macro defines one event field. See
2308 man:lttng-ust(3) for a complete description of the available `ctf_*()`
2309 macros. A `ctf_*()` macro specifies the type, size, and byte order of
2312 Each `ctf_*()` macro takes an _argument expression_ parameter. This is a
2313 C expression that the tracer evalutes at the `tracepoint()` macro site
2314 in the application's source code. This expression provides a field's
2315 source of data. The argument expression can include input argument names
2316 listed in the `TP_ARGS()` macro.
2318 Each `ctf_*()` macro also takes a _field name_ parameter. Field names
2319 must be unique within a given tracepoint definition.
2321 Here's a complete tracepoint definition example:
2323 .Tracepoint definition.
2325 The following tracepoint definition defines a tracepoint which takes
2326 three input arguments and has four output event fields.
2330 #include "my-custom-structure.h"
2336 const struct my_custom_structure*, my_custom_structure,
2341 ctf_string(query_field, query)
2342 ctf_float(double, ratio_field, ratio)
2343 ctf_integer(int, recv_size, my_custom_structure->recv_size)
2344 ctf_integer(int, send_size, my_custom_structure->send_size)
2349 You can refer to this tracepoint definition with the `tracepoint()`
2350 macro in your application's source code like this:
2354 tracepoint(my_provider, my_tracepoint,
2355 my_structure, some_ratio, the_query);
2359 NOTE: The LTTng tracer only evaluates tracepoint arguments at run time
2360 if they satisfy an enabled <<event,event rule>>.
2363 [[using-tracepoint-classes]]
2364 ===== Use a tracepoint class
2366 A _tracepoint class_ is a class of tracepoints which share the same
2367 output event field definitions. A _tracepoint instance_ is one
2368 instance of such a defined tracepoint class, with its own tracepoint
2371 The <<defining-tracepoints,`TRACEPOINT_EVENT()` macro>> is actually a
2372 shorthand which defines both a tracepoint class and a tracepoint
2373 instance at the same time.
2375 When you build a tracepoint provider package, the C or $$C++$$ compiler
2376 creates one serialization function for each **tracepoint class**. A
2377 serialization function is responsible for serializing the event fields
2378 of a tracepoint to a sub-buffer when tracing.
2380 For various performance reasons, when your situation requires multiple
2381 tracepoint definitions with different names, but with the same event
2382 fields, we recommend that you manually create a tracepoint class
2383 and instantiate as many tracepoint instances as needed. One positive
2384 effect of such a design, amongst other advantages, is that all
2385 tracepoint instances of the same tracepoint class reuse the same
2386 serialization function, thus reducing
2387 https://en.wikipedia.org/wiki/Cache_pollution[cache pollution].
2389 .Use a tracepoint class and tracepoint instances.
2391 Consider the following three tracepoint definitions:
2403 ctf_integer(int, userid, userid)
2404 ctf_integer(size_t, len, len)
2416 ctf_integer(int, userid, userid)
2417 ctf_integer(size_t, len, len)
2429 ctf_integer(int, userid, userid)
2430 ctf_integer(size_t, len, len)
2435 In this case, we create three tracepoint classes, with one implicit
2436 tracepoint instance for each of them: `get_account`, `get_settings`, and
2437 `get_transaction`. However, they all share the same event field names
2438 and types. Hence three identical, yet independent serialization
2439 functions are created when you build the tracepoint provider package.
2441 A better design choice is to define a single tracepoint class and three
2442 tracepoint instances:
2446 /* The tracepoint class */
2447 TRACEPOINT_EVENT_CLASS(
2448 /* Tracepoint provider name */
2451 /* Tracepoint class name */
2454 /* Input arguments */
2460 /* Output event fields */
2462 ctf_integer(int, userid, userid)
2463 ctf_integer(size_t, len, len)
2467 /* The tracepoint instances */
2468 TRACEPOINT_EVENT_INSTANCE(
2469 /* Tracepoint provider name */
2472 /* Tracepoint class name */
2475 /* Tracepoint name */
2478 /* Input arguments */
2484 TRACEPOINT_EVENT_INSTANCE(
2493 TRACEPOINT_EVENT_INSTANCE(
2506 [[assigning-log-levels]]
2507 ===== Assign a log level to a tracepoint definition
2509 You can assign an optional _log level_ to a
2510 <<defining-tracepoints,tracepoint definition>>.
2512 Assigning different levels of severity to tracepoint definitions can
2513 be useful: when you <<enabling-disabling-events,create an event rule>>,
2514 you can target tracepoints having a log level as severe as a specific
2517 The concept of LTTng-UST log levels is similar to the levels found
2518 in typical logging frameworks:
2520 * In a logging framework, the log level is given by the function
2521 or method name you use at the log statement site: `debug()`,
2522 `info()`, `warn()`, `error()`, and so on.
2523 * In LTTng-UST, you statically assign the log level to a tracepoint
2524 definition; any `tracepoint()` macro invocation which refers to
2525 this definition has this log level.
2527 You can assign a log level to a tracepoint definition with the
2528 `TRACEPOINT_LOGLEVEL()` macro. You must use this macro _after_ the
2529 <<defining-tracepoints,`TRACEPOINT_EVENT()`>> or
2530 <<using-tracepoint-classes,`TRACEPOINT_INSTANCE()`>> macro for a given
2533 The syntax of the `TRACEPOINT_LOGLEVEL()` macro is:
2536 .`TRACEPOINT_LOGLEVEL()` macro syntax.
2538 TRACEPOINT_LOGLEVEL(provider_name, tracepoint_name, log_level)
2543 * `provider_name` with the tracepoint provider name.
2544 * `tracepoint_name` with the tracepoint name.
2545 * `log_level` with the log level to assign to the tracepoint
2546 definition named `tracepoint_name` in the `provider_name`
2547 tracepoint provider.
2549 See man:lttng-ust(3) for a list of available log level names.
2551 .Assign the `TRACE_DEBUG_UNIT` log level to a tracepoint definition.
2555 /* Tracepoint definition */
2564 ctf_integer(int, userid, userid)
2565 ctf_integer(size_t, len, len)
2569 /* Log level assignment */
2570 TRACEPOINT_LOGLEVEL(my_app, get_transaction, TRACE_DEBUG_UNIT)
2576 ===== Create a tracepoint provider package source file
2578 A _tracepoint provider package source file_ is a C source file which
2579 includes a <<tpp-header,tracepoint provider header file>> to expand its
2580 macros into event serialization and other functions.
2582 You can always use the following tracepoint provider package source
2586 .Tracepoint provider package source file template.
2588 #define TRACEPOINT_CREATE_PROBES
2593 Replace `tp.h` with the name of your <<tpp-header,tracepoint provider
2594 header file>> name. You may also include more than one tracepoint
2595 provider header file here to create a tracepoint provider package
2596 holding more than one tracepoint providers.
2599 [[probing-the-application-source-code]]
2600 ==== Add tracepoints to an application's source code
2602 Once you <<tpp-header,create a tracepoint provider header file>>, you
2603 can use the `tracepoint()` macro in your application's
2604 source code to insert the tracepoints that this header
2605 <<defining-tracepoints,defines>>.
2607 The `tracepoint()` macro takes at least two parameters: the tracepoint
2608 provider name and the tracepoint name. The corresponding tracepoint
2609 definition defines the other parameters.
2611 .`tracepoint()` usage.
2613 The following <<defining-tracepoints,tracepoint definition>> defines a
2614 tracepoint which takes two input arguments and has two output event
2618 .Tracepoint provider header file.
2620 #include "my-custom-structure.h"
2627 const char*, cmd_name
2630 ctf_string(cmd_name, cmd_name)
2631 ctf_integer(int, number_of_args, argc)
2636 You can refer to this tracepoint definition with the `tracepoint()`
2637 macro in your application's source code like this:
2640 .Application's source file.
2644 int main(int argc, char* argv[])
2646 tracepoint(my_provider, my_tracepoint, argc, argv[0]);
2652 Note how the application's source code includes
2653 the tracepoint provider header file containing the tracepoint
2654 definitions to use, path:{tp.h}.
2657 .`tracepoint()` usage with a complex tracepoint definition.
2659 Consider this complex tracepoint definition, where multiple event
2660 fields refer to the same input arguments in their argument expression
2664 .Tracepoint provider header file.
2666 /* For `struct stat` */
2667 #include <sys/types.h>
2668 #include <sys/stat.h>
2680 ctf_integer(int, my_constant_field, 23 + 17)
2681 ctf_integer(int, my_int_arg_field, my_int_arg)
2682 ctf_integer(int, my_int_arg_field2, my_int_arg * my_int_arg)
2683 ctf_integer(int, sum4_field, my_str_arg[0] + my_str_arg[1] +
2684 my_str_arg[2] + my_str_arg[3])
2685 ctf_string(my_str_arg_field, my_str_arg)
2686 ctf_integer_hex(off_t, size_field, st->st_size)
2687 ctf_float(double, size_dbl_field, (double) st->st_size)
2688 ctf_sequence_text(char, half_my_str_arg_field, my_str_arg,
2689 size_t, strlen(my_str_arg) / 2)
2694 You can refer to this tracepoint definition with the `tracepoint()`
2695 macro in your application's source code like this:
2698 .Application's source file.
2700 #define TRACEPOINT_DEFINE
2707 stat("/etc/fstab", &s);
2708 tracepoint(my_provider, my_tracepoint, 23, "Hello, World!", &s);
2714 If you look at the event record that LTTng writes when tracing this
2715 program, assuming the file size of path:{/etc/fstab} is 301{nbsp}bytes,
2716 it should look like this:
2718 .Event record fields
2720 |Field's name |Field's value
2721 |`my_constant_field` |40
2722 |`my_int_arg_field` |23
2723 |`my_int_arg_field2` |529
2725 |`my_str_arg_field` |`Hello, World!`
2726 |`size_field` |0x12d
2727 |`size_dbl_field` |301.0
2728 |`half_my_str_arg_field` |`Hello,`
2732 Sometimes, the arguments you pass to `tracepoint()` are expensive to
2733 compute--they use the call stack, for example. To avoid this
2734 computation when the tracepoint is disabled, you can use the
2735 `tracepoint_enabled()` and `do_tracepoint()` macros.
2737 The syntax of the `tracepoint_enabled()` and `do_tracepoint()` macros
2741 .`tracepoint_enabled()` and `do_tracepoint()` macros syntax.
2743 tracepoint_enabled(provider_name, tracepoint_name)
2744 do_tracepoint(provider_name, tracepoint_name, ...)
2749 * `provider_name` with the tracepoint provider name.
2750 * `tracepoint_name` with the tracepoint name.
2752 `tracepoint_enabled()` returns a non-zero value if the tracepoint named
2753 `tracepoint_name` from the provider named `provider_name` is enabled
2756 `do_tracepoint()` is like `tracepoint()`, except that it doesn't check
2757 if the tracepoint is enabled. Using `tracepoint()` with
2758 `tracepoint_enabled()` is dangerous since `tracepoint()` also contains
2759 the `tracepoint_enabled()` check, thus a race condition is
2760 possible in this situation:
2763 .Possible race condition when using `tracepoint_enabled()` with `tracepoint()`.
2765 if (tracepoint_enabled(my_provider, my_tracepoint)) {
2766 stuff = prepare_stuff();
2769 tracepoint(my_provider, my_tracepoint, stuff);
2772 If the tracepoint is enabled after the condition, then `stuff` is not
2773 prepared: the emitted event will either contain wrong data, or the whole
2774 application could crash (segmentation fault, for example).
2776 NOTE: Neither `tracepoint_enabled()` nor `do_tracepoint()` have an
2777 `STAP_PROBEV()` call. If you need it, you must emit
2781 [[building-tracepoint-providers-and-user-application]]
2782 ==== Build and link a tracepoint provider package and an application
2784 Once you have one or more <<tpp-header,tracepoint provider header
2785 files>> and a <<tpp-source,tracepoint provider package source file>>,
2786 you can create the tracepoint provider package by compiling its source
2787 file. From here, multiple build and run scenarios are possible. The
2788 following table shows common application and library configurations
2789 along with the required command lines to achieve them.
2791 In the following diagrams, we use the following file names:
2794 Executable application.
2797 Application's object file.
2800 Tracepoint provider package object file.
2803 Tracepoint provider package archive file.
2806 Tracepoint provider package shared object file.
2809 User library object file.
2812 User library shared object file.
2814 We use the following symbols in the diagrams of table below:
2817 .Symbols used in the build scenario diagrams.
2818 image::ust-sit-symbols.png[]
2820 We assume that path:{.} is part of the env:LD_LIBRARY_PATH environment
2821 variable in the following instructions.
2823 [role="growable ust-scenarios",cols="asciidoc,asciidoc"]
2824 .Common tracepoint provider package scenarios.
2826 |Scenario |Instructions
2829 The instrumented application is statically linked with
2830 the tracepoint provider package object.
2832 image::ust-sit+app-linked-with-tp-o+app-instrumented.png[]
2835 include::../common/ust-sit-step-tp-o.txt[]
2837 To build the instrumented application:
2839 . In path:{app.c}, before including path:{tpp.h}, add the following line:
2844 #define TRACEPOINT_DEFINE
2848 . Compile the application source file:
2857 . Build the application:
2862 $ gcc -o app app.o tpp.o -llttng-ust -ldl
2866 To run the instrumented application:
2868 * Start the application:
2878 The instrumented application is statically linked with the
2879 tracepoint provider package archive file.
2881 image::ust-sit+app-linked-with-tp-a+app-instrumented.png[]
2884 To create the tracepoint provider package archive file:
2886 . Compile the <<tpp-source,tracepoint provider package source file>>:
2895 . Create the tracepoint provider package archive file:
2900 $ ar rcs tpp.a tpp.o
2904 To build the instrumented application:
2906 . In path:{app.c}, before including path:{tpp.h}, add the following line:
2911 #define TRACEPOINT_DEFINE
2915 . Compile the application source file:
2924 . Build the application:
2929 $ gcc -o app app.o tpp.a -llttng-ust -ldl
2933 To run the instrumented application:
2935 * Start the application:
2945 The instrumented application is linked with the tracepoint provider
2946 package shared object.
2948 image::ust-sit+app-linked-with-tp-so+app-instrumented.png[]
2951 include::../common/ust-sit-step-tp-so.txt[]
2953 To build the instrumented application:
2955 . In path:{app.c}, before including path:{tpp.h}, add the following line:
2960 #define TRACEPOINT_DEFINE
2964 . Compile the application source file:
2973 . Build the application:
2978 $ gcc -o app app.o -ldl -L. -ltpp
2982 To run the instrumented application:
2984 * Start the application:
2994 The tracepoint provider package shared object is preloaded before the
2995 instrumented application starts.
2997 image::ust-sit+tp-so-preloaded+app-instrumented.png[]
3000 include::../common/ust-sit-step-tp-so.txt[]
3002 To build the instrumented application:
3004 . In path:{app.c}, before including path:{tpp.h}, add the
3010 #define TRACEPOINT_DEFINE
3011 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3015 . Compile the application source file:
3024 . Build the application:
3029 $ gcc -o app app.o -ldl
3033 To run the instrumented application with tracing support:
3035 * Preload the tracepoint provider package shared object and
3036 start the application:
3041 $ LD_PRELOAD=./libtpp.so ./app
3045 To run the instrumented application without tracing support:
3047 * Start the application:
3057 The instrumented application dynamically loads the tracepoint provider
3058 package shared object.
3060 image::ust-sit+app-dlopens-tp-so+app-instrumented.png[]
3063 include::../common/ust-sit-step-tp-so.txt[]
3065 To build the instrumented application:
3067 . In path:{app.c}, before including path:{tpp.h}, add the
3073 #define TRACEPOINT_DEFINE
3074 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3078 . Compile the application source file:
3087 . Build the application:
3092 $ gcc -o app app.o -ldl
3096 To run the instrumented application:
3098 * Start the application:
3108 The application is linked with the instrumented user library.
3110 The instrumented user library is statically linked with the tracepoint
3111 provider package object file.
3113 image::ust-sit+app-linked-with-lib+lib-linked-with-tp-o+lib-instrumented.png[]
3116 include::../common/ust-sit-step-tp-o-fpic.txt[]
3118 To build the instrumented user library:
3120 . In path:{emon.c}, before including path:{tpp.h}, add the
3126 #define TRACEPOINT_DEFINE
3130 . Compile the user library source file:
3135 $ gcc -I. -fpic -c emon.c
3139 . Build the user library shared object:
3144 $ gcc -shared -o libemon.so emon.o tpp.o -llttng-ust -ldl
3148 To build the application:
3150 . Compile the application source file:
3159 . Build the application:
3164 $ gcc -o app app.o -L. -lemon
3168 To run the application:
3170 * Start the application:
3180 The application is linked with the instrumented user library.
3182 The instrumented user library is linked with the tracepoint provider
3183 package shared object.
3185 image::ust-sit+app-linked-with-lib+lib-linked-with-tp-so+lib-instrumented.png[]
3188 include::../common/ust-sit-step-tp-so.txt[]
3190 To build the instrumented user library:
3192 . In path:{emon.c}, before including path:{tpp.h}, add the
3198 #define TRACEPOINT_DEFINE
3202 . Compile the user library source file:
3207 $ gcc -I. -fpic -c emon.c
3211 . Build the user library shared object:
3216 $ gcc -shared -o libemon.so emon.o -ldl -L. -ltpp
3220 To build the application:
3222 . Compile the application source file:
3231 . Build the application:
3236 $ gcc -o app app.o -L. -lemon
3240 To run the application:
3242 * Start the application:
3252 The tracepoint provider package shared object is preloaded before the
3255 The application is linked with the instrumented user library.
3257 image::ust-sit+tp-so-preloaded+app-linked-with-lib+lib-instrumented.png[]
3260 include::../common/ust-sit-step-tp-so.txt[]
3262 To build the instrumented user library:
3264 . In path:{emon.c}, before including path:{tpp.h}, add the
3270 #define TRACEPOINT_DEFINE
3271 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3275 . Compile the user library source file:
3280 $ gcc -I. -fpic -c emon.c
3284 . Build the user library shared object:
3289 $ gcc -shared -o libemon.so emon.o -ldl
3293 To build the application:
3295 . Compile the application source file:
3304 . Build the application:
3309 $ gcc -o app app.o -L. -lemon
3313 To run the application with tracing support:
3315 * Preload the tracepoint provider package shared object and
3316 start the application:
3321 $ LD_PRELOAD=./libtpp.so ./app
3325 To run the application without tracing support:
3327 * Start the application:
3337 The application is linked with the instrumented user library.
3339 The instrumented user library dynamically loads the tracepoint provider
3340 package shared object.
3342 image::ust-sit+app-linked-with-lib+lib-dlopens-tp-so+lib-instrumented.png[]
3345 include::../common/ust-sit-step-tp-so.txt[]
3347 To build the instrumented user library:
3349 . In path:{emon.c}, before including path:{tpp.h}, add the
3355 #define TRACEPOINT_DEFINE
3356 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3360 . Compile the user library source file:
3365 $ gcc -I. -fpic -c emon.c
3369 . Build the user library shared object:
3374 $ gcc -shared -o libemon.so emon.o -ldl
3378 To build the application:
3380 . Compile the application source file:
3389 . Build the application:
3394 $ gcc -o app app.o -L. -lemon
3398 To run the application:
3400 * Start the application:
3410 The application dynamically loads the instrumented user library.
3412 The instrumented user library is linked with the tracepoint provider
3413 package shared object.
3415 image::ust-sit+app-dlopens-lib+lib-linked-with-tp-so+lib-instrumented.png[]
3418 include::../common/ust-sit-step-tp-so.txt[]
3420 To build the instrumented user library:
3422 . In path:{emon.c}, before including path:{tpp.h}, add the
3428 #define TRACEPOINT_DEFINE
3432 . Compile the user library source file:
3437 $ gcc -I. -fpic -c emon.c
3441 . Build the user library shared object:
3446 $ gcc -shared -o libemon.so emon.o -ldl -L. -ltpp
3450 To build the application:
3452 . Compile the application source file:
3461 . Build the application:
3466 $ gcc -o app app.o -ldl -L. -lemon
3470 To run the application:
3472 * Start the application:
3482 The application dynamically loads the instrumented user library.
3484 The instrumented user library dynamically loads the tracepoint provider
3485 package shared object.
3487 image::ust-sit+app-dlopens-lib+lib-dlopens-tp-so+lib-instrumented.png[]
3490 include::../common/ust-sit-step-tp-so.txt[]
3492 To build the instrumented user library:
3494 . In path:{emon.c}, before including path:{tpp.h}, add the
3500 #define TRACEPOINT_DEFINE
3501 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3505 . Compile the user library source file:
3510 $ gcc -I. -fpic -c emon.c
3514 . Build the user library shared object:
3519 $ gcc -shared -o libemon.so emon.o -ldl
3523 To build the application:
3525 . Compile the application source file:
3534 . Build the application:
3539 $ gcc -o app app.o -ldl -L. -lemon
3543 To run the application:
3545 * Start the application:
3555 The tracepoint provider package shared object is preloaded before the
3558 The application dynamically loads the instrumented user library.
3560 image::ust-sit+tp-so-preloaded+app-dlopens-lib+lib-instrumented.png[]
3563 include::../common/ust-sit-step-tp-so.txt[]
3565 To build the instrumented user library:
3567 . In path:{emon.c}, before including path:{tpp.h}, add the
3573 #define TRACEPOINT_DEFINE
3574 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3578 . Compile the user library source file:
3583 $ gcc -I. -fpic -c emon.c
3587 . Build the user library shared object:
3592 $ gcc -shared -o libemon.so emon.o -ldl
3596 To build the application:
3598 . Compile the application source file:
3607 . Build the application:
3612 $ gcc -o app app.o -L. -lemon
3616 To run the application with tracing support:
3618 * Preload the tracepoint provider package shared object and
3619 start the application:
3624 $ LD_PRELOAD=./libtpp.so ./app
3628 To run the application without tracing support:
3630 * Start the application:
3640 The application is statically linked with the tracepoint provider
3641 package object file.
3643 The application is linked with the instrumented user library.
3645 image::ust-sit+app-linked-with-tp-o+app-linked-with-lib+lib-instrumented.png[]
3648 include::../common/ust-sit-step-tp-o.txt[]
3650 To build the instrumented user library:
3652 . In path:{emon.c}, before including path:{tpp.h}, add the
3658 #define TRACEPOINT_DEFINE
3662 . Compile the user library source file:
3667 $ gcc -I. -fpic -c emon.c
3671 . Build the user library shared object:
3676 $ gcc -shared -o libemon.so emon.o
3680 To build the application:
3682 . Compile the application source file:
3691 . Build the application:
3696 $ gcc -o app app.o tpp.o -llttng-ust -ldl -L. -lemon
3700 To run the instrumented application:
3702 * Start the application:
3712 The application is statically linked with the tracepoint provider
3713 package object file.
3715 The application dynamically loads the instrumented user library.
3717 image::ust-sit+app-linked-with-tp-o+app-dlopens-lib+lib-instrumented.png[]
3720 include::../common/ust-sit-step-tp-o.txt[]
3722 To build the application:
3724 . In path:{app.c}, before including path:{tpp.h}, add the following line:
3729 #define TRACEPOINT_DEFINE
3733 . Compile the application source file:
3742 . Build the application:
3747 $ gcc -Wl,--export-dynamic -o app app.o tpp.o \
3752 The `--export-dynamic` option passed to the linker is necessary for the
3753 dynamically loaded library to ``see'' the tracepoint symbols defined in
3756 To build the instrumented user library:
3758 . Compile the user library source file:
3763 $ gcc -I. -fpic -c emon.c
3767 . Build the user library shared object:
3772 $ gcc -shared -o libemon.so emon.o
3776 To run the application:
3778 * Start the application:
3789 [[using-lttng-ust-with-daemons]]
3790 ===== Use noch:{LTTng-UST} with daemons
3792 If your instrumented application calls man:fork(2), man:clone(2),
3793 or BSD's man:rfork(2), without a following man:exec(3)-family
3794 system call, you must preload the path:{liblttng-ust-fork.so} shared
3795 object when you start the application.
3799 $ LD_PRELOAD=liblttng-ust-fork.so ./my-app
3802 If your tracepoint provider package is
3803 a shared library which you also preload, you must put both
3804 shared objects in env:LD_PRELOAD:
3808 $ LD_PRELOAD=liblttng-ust-fork.so:/path/to/tp.so ./my-app
3814 ===== Use noch:{LTTng-UST} with applications which close file descriptors that don't belong to them
3816 If your instrumented application closes one or more file descriptors
3817 which it did not open itself, you must preload the
3818 path:{liblttng-ust-fd.so} shared object when you start the application:
3822 $ LD_PRELOAD=liblttng-ust-fd.so ./my-app
3825 Typical use cases include closing all the file descriptors after
3826 man:fork(2) or man:rfork(2) and buggy applications doing
3830 [[lttng-ust-pkg-config]]
3831 ===== Use noch:{pkg-config}
3833 On some distributions, LTTng-UST ships with a
3834 https://www.freedesktop.org/wiki/Software/pkg-config/[pkg-config]
3835 metadata file. If this is your case, then you can use cmd:pkg-config to
3836 build an application on the command line:
3840 $ gcc -o my-app my-app.o tp.o $(pkg-config --cflags --libs lttng-ust)
3844 [[instrumenting-32-bit-app-on-64-bit-system]]
3845 ===== [[advanced-instrumenting-techniques]]Build a 32-bit instrumented application for a 64-bit target system
3847 In order to trace a 32-bit application running on a 64-bit system,
3848 LTTng must use a dedicated 32-bit
3849 <<lttng-consumerd,consumer daemon>>.
3851 The following steps show how to build and install a 32-bit consumer
3852 daemon, which is _not_ part of the default 64-bit LTTng build, how to
3853 build and install the 32-bit LTTng-UST libraries, and how to build and
3854 link an instrumented 32-bit application in that context.
3856 To build a 32-bit instrumented application for a 64-bit target system,
3857 assuming you have a fresh target system with no installed Userspace RCU
3860 . Download, build, and install a 32-bit version of Userspace RCU:
3865 $ cd $(mktemp -d) &&
3866 wget http://lttng.org/files/urcu/userspace-rcu-latest-0.9.tar.bz2 &&
3867 tar -xf userspace-rcu-latest-0.9.tar.bz2 &&
3868 cd userspace-rcu-0.9.* &&
3869 ./configure --libdir=/usr/local/lib32 CFLAGS=-m32 &&
3871 sudo make install &&
3876 . Using your distribution's package manager, or from source, install
3877 the following 32-bit versions of the following dependencies of
3878 LTTng-tools and LTTng-UST:
3881 * https://sourceforge.net/projects/libuuid/[libuuid]
3882 * http://directory.fsf.org/wiki/Popt[popt]
3883 * http://www.xmlsoft.org/[libxml2]
3886 . Download, build, and install a 32-bit version of the latest
3887 LTTng-UST{nbsp}{revision}:
3892 $ cd $(mktemp -d) &&
3893 wget http://lttng.org/files/lttng-ust/lttng-ust-latest-2.11.tar.bz2 &&
3894 tar -xf lttng-ust-latest-2.11.tar.bz2 &&
3895 cd lttng-ust-2.11.* &&
3896 ./configure --libdir=/usr/local/lib32 \
3897 CFLAGS=-m32 CXXFLAGS=-m32 \
3898 LDFLAGS='-L/usr/local/lib32 -L/usr/lib32' &&
3900 sudo make install &&
3907 Depending on your distribution,
3908 32-bit libraries could be installed at a different location than
3909 `/usr/lib32`. For example, Debian is known to install
3910 some 32-bit libraries in `/usr/lib/i386-linux-gnu`.
3912 In this case, make sure to set `LDFLAGS` to all the
3913 relevant 32-bit library paths, for example:
3917 $ LDFLAGS='-L/usr/lib/i386-linux-gnu -L/usr/lib32'
3921 . Download the latest LTTng-tools{nbsp}{revision}, build, and install
3922 the 32-bit consumer daemon:
3927 $ cd $(mktemp -d) &&
3928 wget http://lttng.org/files/lttng-tools/lttng-tools-latest-2.11.tar.bz2 &&
3929 tar -xf lttng-tools-latest-2.11.tar.bz2 &&
3930 cd lttng-tools-2.11.* &&
3931 ./configure --libdir=/usr/local/lib32 CFLAGS=-m32 CXXFLAGS=-m32 \
3932 LDFLAGS='-L/usr/local/lib32 -L/usr/lib32' \
3933 --disable-bin-lttng --disable-bin-lttng-crash \
3934 --disable-bin-lttng-relayd --disable-bin-lttng-sessiond &&
3936 cd src/bin/lttng-consumerd &&
3937 sudo make install &&
3942 . From your distribution or from source,
3943 <<installing-lttng,install>> the 64-bit versions of
3944 LTTng-UST and Userspace RCU.
3945 . Download, build, and install the 64-bit version of the
3946 latest LTTng-tools{nbsp}{revision}:
3951 $ cd $(mktemp -d) &&
3952 wget http://lttng.org/files/lttng-tools/lttng-tools-latest-2.11.tar.bz2 &&
3953 tar -xf lttng-tools-latest-2.11.tar.bz2 &&
3954 cd lttng-tools-2.11.* &&
3955 ./configure --with-consumerd32-libdir=/usr/local/lib32 \
3956 --with-consumerd32-bin=/usr/local/lib32/lttng/libexec/lttng-consumerd &&
3958 sudo make install &&
3963 . Pass the following options to man:gcc(1), man:g++(1), or man:clang(1)
3964 when linking your 32-bit application:
3967 -m32 -L/usr/lib32 -L/usr/local/lib32 \
3968 -Wl,-rpath,/usr/lib32,-rpath,/usr/local/lib32
3971 For example, let's rebuild the quick start example in
3972 <<tracing-your-own-user-application,Trace a user application>> as an
3973 instrumented 32-bit application:
3978 $ gcc -m32 -c -I. hello-tp.c
3979 $ gcc -m32 -c hello.c
3980 $ gcc -m32 -o hello hello.o hello-tp.o \
3981 -L/usr/lib32 -L/usr/local/lib32 \
3982 -Wl,-rpath,/usr/lib32,-rpath,/usr/local/lib32 \
3987 No special action is required to execute the 32-bit application and
3988 to trace it: use the command-line man:lttng(1) tool as usual.
3995 man:tracef(3) is a small LTTng-UST API designed for quick,
3996 man:printf(3)-like instrumentation without the burden of
3997 <<tracepoint-provider,creating>> and
3998 <<building-tracepoint-providers-and-user-application,building>>
3999 a tracepoint provider package.
4001 To use `tracef()` in your application:
4003 . In the C or C++ source files where you need to use `tracef()`,
4004 include `<lttng/tracef.h>`:
4009 #include <lttng/tracef.h>
4013 . In the application's source code, use `tracef()` like you would use
4021 tracef("my message: %d (%s)", my_integer, my_string);
4027 . Link your application with `liblttng-ust`:
4032 $ gcc -o app app.c -llttng-ust
4036 To trace the events that `tracef()` calls emit:
4038 * <<enabling-disabling-events,Create an event rule>> which matches the
4039 `lttng_ust_tracef:*` event name:
4044 $ lttng enable-event --userspace 'lttng_ust_tracef:*'
4049 .Limitations of `tracef()`
4051 The `tracef()` utility function was developed to make user space tracing
4052 super simple, albeit with notable disadvantages compared to
4053 <<defining-tracepoints,user-defined tracepoints>>:
4055 * All the emitted events have the same tracepoint provider and
4056 tracepoint names, respectively `lttng_ust_tracef` and `event`.
4057 * There is no static type checking.
4058 * The only event record field you actually get, named `msg`, is a string
4059 potentially containing the values you passed to `tracef()`
4060 using your own format string. This also means that you cannot filter
4061 events with a custom expression at run time because there are no
4063 * Since `tracef()` uses the C standard library's man:vasprintf(3)
4064 function behind the scenes to format the strings at run time, its
4065 expected performance is lower than with user-defined tracepoints,
4066 which do not require a conversion to a string.
4068 Taking this into consideration, `tracef()` is useful for some quick
4069 prototyping and debugging, but you should not consider it for any
4070 permanent and serious applicative instrumentation.
4076 ==== Use `tracelog()`
4078 The man:tracelog(3) API is very similar to <<tracef,`tracef()`>>, with
4079 the difference that it accepts an additional log level parameter.
4081 The goal of `tracelog()` is to ease the migration from logging to
4084 To use `tracelog()` in your application:
4086 . In the C or C++ source files where you need to use `tracelog()`,
4087 include `<lttng/tracelog.h>`:
4092 #include <lttng/tracelog.h>
4096 . In the application's source code, use `tracelog()` like you would use
4097 man:printf(3), except for the first parameter which is the log
4105 tracelog(TRACE_WARNING, "my message: %d (%s)",
4106 my_integer, my_string);
4112 See man:lttng-ust(3) for a list of available log level names.
4114 . Link your application with `liblttng-ust`:
4119 $ gcc -o app app.c -llttng-ust
4123 To trace the events that `tracelog()` calls emit with a log level
4124 _as severe as_ a specific log level:
4126 * <<enabling-disabling-events,Create an event rule>> which matches the
4127 `lttng_ust_tracelog:*` event name and a minimum level
4133 $ lttng enable-event --userspace 'lttng_ust_tracelog:*'
4134 --loglevel=TRACE_WARNING
4138 To trace the events that `tracelog()` calls emit with a
4139 _specific log level_:
4141 * Create an event rule which matches the `lttng_ust_tracelog:*`
4142 event name and a specific log level:
4147 $ lttng enable-event --userspace 'lttng_ust_tracelog:*'
4148 --loglevel-only=TRACE_INFO
4153 [[prebuilt-ust-helpers]]
4154 === Prebuilt user space tracing helpers
4156 The LTTng-UST package provides a few helpers in the form of preloadable
4157 shared objects which automatically instrument system functions and
4160 The helper shared objects are normally found in dir:{/usr/lib}. If you
4161 built LTTng-UST <<building-from-source,from source>>, they are probably
4162 located in dir:{/usr/local/lib}.
4164 The installed user space tracing helpers in LTTng-UST{nbsp}{revision}
4167 path:{liblttng-ust-libc-wrapper.so}::
4168 path:{liblttng-ust-pthread-wrapper.so}::
4169 <<liblttng-ust-libc-pthread-wrapper,C{nbsp}standard library
4170 memory and POSIX threads function tracing>>.
4172 path:{liblttng-ust-cyg-profile.so}::
4173 path:{liblttng-ust-cyg-profile-fast.so}::
4174 <<liblttng-ust-cyg-profile,Function entry and exit tracing>>.
4176 path:{liblttng-ust-dl.so}::
4177 <<liblttng-ust-dl,Dynamic linker tracing>>.
4179 To use a user space tracing helper with any user application:
4181 * Preload the helper shared object when you start the application:
4186 $ LD_PRELOAD=liblttng-ust-libc-wrapper.so my-app
4190 You can preload more than one helper:
4195 $ LD_PRELOAD=liblttng-ust-libc-wrapper.so:liblttng-ust-dl.so my-app
4201 [[liblttng-ust-libc-pthread-wrapper]]
4202 ==== Instrument C standard library memory and POSIX threads functions
4204 The path:{liblttng-ust-libc-wrapper.so} and
4205 path:{liblttng-ust-pthread-wrapper.so} helpers
4206 add instrumentation to some C standard library and POSIX
4210 .Functions instrumented by preloading path:{liblttng-ust-libc-wrapper.so}.
4212 |TP provider name |TP name |Instrumented function
4214 .6+|`lttng_ust_libc` |`malloc` |man:malloc(3)
4215 |`calloc` |man:calloc(3)
4216 |`realloc` |man:realloc(3)
4217 |`free` |man:free(3)
4218 |`memalign` |man:memalign(3)
4219 |`posix_memalign` |man:posix_memalign(3)
4223 .Functions instrumented by preloading path:{liblttng-ust-pthread-wrapper.so}.
4225 |TP provider name |TP name |Instrumented function
4227 .4+|`lttng_ust_pthread` |`pthread_mutex_lock_req` |man:pthread_mutex_lock(3p) (request time)
4228 |`pthread_mutex_lock_acq` |man:pthread_mutex_lock(3p) (acquire time)
4229 |`pthread_mutex_trylock` |man:pthread_mutex_trylock(3p)
4230 |`pthread_mutex_unlock` |man:pthread_mutex_unlock(3p)
4233 When you preload the shared object, it replaces the functions listed
4234 in the previous tables by wrappers which contain tracepoints and call
4235 the replaced functions.
4238 [[liblttng-ust-cyg-profile]]
4239 ==== Instrument function entry and exit
4241 The path:{liblttng-ust-cyg-profile*.so} helpers can add instrumentation
4242 to the entry and exit points of functions.
4244 man:gcc(1) and man:clang(1) have an option named
4245 https://gcc.gnu.org/onlinedocs/gcc/Instrumentation-Options.html[`-finstrument-functions`]
4246 which generates instrumentation calls for entry and exit to functions.
4247 The LTTng-UST function tracing helpers,
4248 path:{liblttng-ust-cyg-profile.so} and
4249 path:{liblttng-ust-cyg-profile-fast.so}, take advantage of this feature
4250 to add tracepoints to the two generated functions (which contain
4251 `cyg_profile` in their names, hence the helper's name).
4253 To use the LTTng-UST function tracing helper, the source files to
4254 instrument must be built using the `-finstrument-functions` compiler
4257 There are two versions of the LTTng-UST function tracing helper:
4259 * **path:{liblttng-ust-cyg-profile-fast.so}** is a lightweight variant
4260 that you should only use when it can be _guaranteed_ that the
4261 complete event stream is recorded without any lost event record.
4262 Any kind of duplicate information is left out.
4264 Assuming no event record is lost, having only the function addresses on
4265 entry is enough to create a call graph, since an event record always
4266 contains the ID of the CPU that generated it.
4268 You can use a tool like man:addr2line(1) to convert function addresses
4269 back to source file names and line numbers.
4271 * **path:{liblttng-ust-cyg-profile.so}** is a more robust variant
4272 which also works in use cases where event records might get discarded or
4273 not recorded from application startup.
4274 In these cases, the trace analyzer needs more information to be
4275 able to reconstruct the program flow.
4277 See man:lttng-ust-cyg-profile(3) to learn more about the instrumentation
4278 points of this helper.
4280 All the tracepoints that this helper provides have the
4281 log level `TRACE_DEBUG_FUNCTION` (see man:lttng-ust(3)).
4283 TIP: It's sometimes a good idea to limit the number of source files that
4284 you compile with the `-finstrument-functions` option to prevent LTTng
4285 from writing an excessive amount of trace data at run time. When using
4286 man:gcc(1), you can use the
4287 `-finstrument-functions-exclude-function-list` option to avoid
4288 instrument entries and exits of specific function names.
4293 ==== Instrument the dynamic linker
4295 The path:{liblttng-ust-dl.so} helper adds instrumentation to the
4296 man:dlopen(3) and man:dlclose(3) function calls.
4298 See man:lttng-ust-dl(3) to learn more about the instrumentation points
4303 [[java-application]]
4304 === User space Java agent
4306 You can instrument any Java application which uses one of the following
4309 * The https://docs.oracle.com/javase/7/docs/api/java/util/logging/package-summary.html[**`java.util.logging`**]
4310 (JUL) core logging facilities.
4311 * http://logging.apache.org/log4j/1.2/[**Apache log4j{nbsp}1.2**], since
4312 LTTng{nbsp}2.6. Note that Apache Log4j{nbsp}2 is not supported.
4315 .LTTng-UST Java agent imported by a Java application.
4316 image::java-app.png[]
4318 Note that the methods described below are new in LTTng{nbsp}2.8.
4319 Previous LTTng versions use another technique.
4321 NOTE: We use http://openjdk.java.net/[OpenJDK]{nbsp}8 for development
4322 and https://ci.lttng.org/[continuous integration], thus this version is
4323 directly supported. However, the LTTng-UST Java agent is also tested
4324 with OpenJDK{nbsp}7.
4329 ==== Use the LTTng-UST Java agent for `java.util.logging`
4331 To use the LTTng-UST Java agent in a Java application which uses
4332 `java.util.logging` (JUL):
4334 . In the Java application's source code, import the LTTng-UST
4335 log handler package for `java.util.logging`:
4340 import org.lttng.ust.agent.jul.LttngLogHandler;
4344 . Create an LTTng-UST JUL log handler:
4349 Handler lttngUstLogHandler = new LttngLogHandler();
4353 . Add this handler to the JUL loggers which should emit LTTng events:
4358 Logger myLogger = Logger.getLogger("some-logger");
4360 myLogger.addHandler(lttngUstLogHandler);
4364 . Use `java.util.logging` log statements and configuration as usual.
4365 The loggers with an attached LTTng-UST log handler can emit
4368 . Before exiting the application, remove the LTTng-UST log handler from
4369 the loggers attached to it and call its `close()` method:
4374 myLogger.removeHandler(lttngUstLogHandler);
4375 lttngUstLogHandler.close();
4379 This is not strictly necessary, but it is recommended for a clean
4380 disposal of the handler's resources.
4382 . Include the LTTng-UST Java agent's common and JUL-specific JAR files,
4383 path:{lttng-ust-agent-common.jar} and path:{lttng-ust-agent-jul.jar},
4385 https://docs.oracle.com/javase/tutorial/essential/environment/paths.html[class
4386 path] when you build the Java application.
4388 The JAR files are typically located in dir:{/usr/share/java}.
4390 IMPORTANT: The LTTng-UST Java agent must be
4391 <<installing-lttng,installed>> for the logging framework your
4394 .Use the LTTng-UST Java agent for `java.util.logging`.
4399 import java.io.IOException;
4400 import java.util.logging.Handler;
4401 import java.util.logging.Logger;
4402 import org.lttng.ust.agent.jul.LttngLogHandler;
4406 private static final int answer = 42;
4408 public static void main(String[] argv) throws Exception
4411 Logger logger = Logger.getLogger("jello");
4413 // Create an LTTng-UST log handler
4414 Handler lttngUstLogHandler = new LttngLogHandler();
4416 // Add the LTTng-UST log handler to our logger
4417 logger.addHandler(lttngUstLogHandler);
4420 logger.info("some info");
4421 logger.warning("some warning");
4423 logger.finer("finer information; the answer is " + answer);
4425 logger.severe("error!");
4427 // Not mandatory, but cleaner
4428 logger.removeHandler(lttngUstLogHandler);
4429 lttngUstLogHandler.close();
4438 $ javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar Test.java
4441 <<creating-destroying-tracing-sessions,Create a tracing session>>,
4442 <<enabling-disabling-events,create an event rule>> matching the
4443 `jello` JUL logger, and <<basic-tracing-session-control,start tracing>>:
4448 $ lttng enable-event --jul jello
4452 Run the compiled class:
4456 $ java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar:. Test
4459 <<basic-tracing-session-control,Stop tracing>> and inspect the
4469 In the resulting trace, an <<event,event record>> generated by a Java
4470 application using `java.util.logging` is named `lttng_jul:event` and
4471 has the following fields:
4474 Log record's message.
4480 Name of the class in which the log statement was executed.
4483 Name of the method in which the log statement was executed.
4486 Logging time (timestamp in milliseconds).
4489 Log level integer value.
4492 ID of the thread in which the log statement was executed.
4494 You can use the opt:lttng-enable-event(1):--loglevel or
4495 opt:lttng-enable-event(1):--loglevel-only option of the
4496 man:lttng-enable-event(1) command to target a range of JUL log levels
4497 or a specific JUL log level.
4502 ==== Use the LTTng-UST Java agent for Apache log4j
4504 To use the LTTng-UST Java agent in a Java application which uses
4505 Apache log4j{nbsp}1.2:
4507 . In the Java application's source code, import the LTTng-UST
4508 log appender package for Apache log4j:
4513 import org.lttng.ust.agent.log4j.LttngLogAppender;
4517 . Create an LTTng-UST log4j log appender:
4522 Appender lttngUstLogAppender = new LttngLogAppender();
4526 . Add this appender to the log4j loggers which should emit LTTng events:
4531 Logger myLogger = Logger.getLogger("some-logger");
4533 myLogger.addAppender(lttngUstLogAppender);
4537 . Use Apache log4j log statements and configuration as usual. The
4538 loggers with an attached LTTng-UST log appender can emit LTTng events.
4540 . Before exiting the application, remove the LTTng-UST log appender from
4541 the loggers attached to it and call its `close()` method:
4546 myLogger.removeAppender(lttngUstLogAppender);
4547 lttngUstLogAppender.close();
4551 This is not strictly necessary, but it is recommended for a clean
4552 disposal of the appender's resources.
4554 . Include the LTTng-UST Java agent's common and log4j-specific JAR
4555 files, path:{lttng-ust-agent-common.jar} and
4556 path:{lttng-ust-agent-log4j.jar}, in the
4557 https://docs.oracle.com/javase/tutorial/essential/environment/paths.html[class
4558 path] when you build the Java application.
4560 The JAR files are typically located in dir:{/usr/share/java}.
4562 IMPORTANT: The LTTng-UST Java agent must be
4563 <<installing-lttng,installed>> for the logging framework your
4566 .Use the LTTng-UST Java agent for Apache log4j.
4571 import org.apache.log4j.Appender;
4572 import org.apache.log4j.Logger;
4573 import org.lttng.ust.agent.log4j.LttngLogAppender;
4577 private static final int answer = 42;
4579 public static void main(String[] argv) throws Exception
4582 Logger logger = Logger.getLogger("jello");
4584 // Create an LTTng-UST log appender
4585 Appender lttngUstLogAppender = new LttngLogAppender();
4587 // Add the LTTng-UST log appender to our logger
4588 logger.addAppender(lttngUstLogAppender);
4591 logger.info("some info");
4592 logger.warn("some warning");
4594 logger.debug("debug information; the answer is " + answer);
4596 logger.fatal("error!");
4598 // Not mandatory, but cleaner
4599 logger.removeAppender(lttngUstLogAppender);
4600 lttngUstLogAppender.close();
4606 Build this example (`$LOG4JPATH` is the path to the Apache log4j JAR
4611 $ javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-log4j.jar:$LOG4JPATH Test.java
4614 <<creating-destroying-tracing-sessions,Create a tracing session>>,
4615 <<enabling-disabling-events,create an event rule>> matching the
4616 `jello` log4j logger, and <<basic-tracing-session-control,start tracing>>:
4621 $ lttng enable-event --log4j jello
4625 Run the compiled class:
4629 $ java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-log4j.jar:$LOG4JPATH:. Test
4632 <<basic-tracing-session-control,Stop tracing>> and inspect the
4642 In the resulting trace, an <<event,event record>> generated by a Java
4643 application using log4j is named `lttng_log4j:event` and
4644 has the following fields:
4647 Log record's message.
4653 Name of the class in which the log statement was executed.
4656 Name of the method in which the log statement was executed.
4659 Name of the file in which the executed log statement is located.
4662 Line number at which the log statement was executed.
4668 Log level integer value.
4671 Name of the Java thread in which the log statement was executed.
4673 You can use the opt:lttng-enable-event(1):--loglevel or
4674 opt:lttng-enable-event(1):--loglevel-only option of the
4675 man:lttng-enable-event(1) command to target a range of Apache log4j log levels
4676 or a specific log4j log level.
4680 [[java-application-context]]
4681 ==== Provide application-specific context fields in a Java application
4683 A Java application-specific context field is a piece of state provided
4684 by the application which <<adding-context,you can add>>, using the
4685 man:lttng-add-context(1) command, to each <<event,event record>>
4686 produced by the log statements of this application.
4688 For example, a given object might have a current request ID variable.
4689 You can create a context information retriever for this object and
4690 assign a name to this current request ID. You can then, using the
4691 man:lttng-add-context(1) command, add this context field by name to
4692 the JUL or log4j <<channel,channel>>.
4694 To provide application-specific context fields in a Java application:
4696 . In the Java application's source code, import the LTTng-UST
4697 Java agent context classes and interfaces:
4702 import org.lttng.ust.agent.context.ContextInfoManager;
4703 import org.lttng.ust.agent.context.IContextInfoRetriever;
4707 . Create a context information retriever class, that is, a class which
4708 implements the `IContextInfoRetriever` interface:
4713 class MyContextInfoRetriever implements IContextInfoRetriever
4716 public Object retrieveContextInfo(String key)
4718 if (key.equals("intCtx")) {
4720 } else if (key.equals("strContext")) {
4721 return "context value!";
4730 This `retrieveContextInfo()` method is the only member of the
4731 `IContextInfoRetriever` interface. Its role is to return the current
4732 value of a state by name to create a context field. The names of the
4733 context fields and which state variables they return depends on your
4736 All primitive types and objects are supported as context fields.
4737 When `retrieveContextInfo()` returns an object, the context field
4738 serializer calls its `toString()` method to add a string field to
4739 event records. The method can also return `null`, which means that
4740 no context field is available for the required name.
4742 . Register an instance of your context information retriever class to
4743 the context information manager singleton:
4748 IContextInfoRetriever cir = new MyContextInfoRetriever();
4749 ContextInfoManager cim = ContextInfoManager.getInstance();
4750 cim.registerContextInfoRetriever("retrieverName", cir);
4754 . Before exiting the application, remove your context information
4755 retriever from the context information manager singleton:
4760 ContextInfoManager cim = ContextInfoManager.getInstance();
4761 cim.unregisterContextInfoRetriever("retrieverName");
4765 This is not strictly necessary, but it is recommended for a clean
4766 disposal of some manager's resources.
4768 . Build your Java application with LTTng-UST Java agent support as
4769 usual, following the procedure for either the <<jul,JUL>> or
4770 <<log4j,Apache log4j>> framework.
4773 .Provide application-specific context fields in a Java application.
4778 import java.util.logging.Handler;
4779 import java.util.logging.Logger;
4780 import org.lttng.ust.agent.jul.LttngLogHandler;
4781 import org.lttng.ust.agent.context.ContextInfoManager;
4782 import org.lttng.ust.agent.context.IContextInfoRetriever;
4786 // Our context information retriever class
4787 private static class MyContextInfoRetriever
4788 implements IContextInfoRetriever
4791 public Object retrieveContextInfo(String key) {
4792 if (key.equals("intCtx")) {
4794 } else if (key.equals("strContext")) {
4795 return "context value!";
4802 private static final int answer = 42;
4804 public static void main(String args[]) throws Exception
4806 // Get the context information manager instance
4807 ContextInfoManager cim = ContextInfoManager.getInstance();
4809 // Create and register our context information retriever
4810 IContextInfoRetriever cir = new MyContextInfoRetriever();
4811 cim.registerContextInfoRetriever("myRetriever", cir);
4814 Logger logger = Logger.getLogger("jello");
4816 // Create an LTTng-UST log handler
4817 Handler lttngUstLogHandler = new LttngLogHandler();
4819 // Add the LTTng-UST log handler to our logger
4820 logger.addHandler(lttngUstLogHandler);
4823 logger.info("some info");
4824 logger.warning("some warning");
4826 logger.finer("finer information; the answer is " + answer);
4828 logger.severe("error!");
4830 // Not mandatory, but cleaner
4831 logger.removeHandler(lttngUstLogHandler);
4832 lttngUstLogHandler.close();
4833 cim.unregisterContextInfoRetriever("myRetriever");
4842 $ javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar Test.java
4845 <<creating-destroying-tracing-sessions,Create a tracing session>>
4846 and <<enabling-disabling-events,create an event rule>> matching the
4852 $ lttng enable-event --jul jello
4855 <<adding-context,Add the application-specific context fields>> to the
4860 $ lttng add-context --jul --type='$app.myRetriever:intCtx'
4861 $ lttng add-context --jul --type='$app.myRetriever:strContext'
4864 <<basic-tracing-session-control,Start tracing>>:
4871 Run the compiled class:
4875 $ java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar:. Test
4878 <<basic-tracing-session-control,Stop tracing>> and inspect the
4890 [[python-application]]
4891 === User space Python agent
4893 You can instrument a Python{nbsp}2 or Python{nbsp}3 application which
4895 https://docs.python.org/3/library/logging.html[`logging`] package.
4897 Each log statement emits an LTTng event once the
4898 application module imports the
4899 <<lttng-ust-agents,LTTng-UST Python agent>> package.
4902 .A Python application importing the LTTng-UST Python agent.
4903 image::python-app.png[]
4905 To use the LTTng-UST Python agent:
4907 . In the Python application's source code, import the LTTng-UST Python
4917 The LTTng-UST Python agent automatically adds its logging handler to the
4918 root logger at import time.
4920 Any log statement that the application executes before this import does
4921 not emit an LTTng event.
4923 IMPORTANT: The LTTng-UST Python agent must be
4924 <<installing-lttng,installed>>.
4926 . Use log statements and logging configuration as usual.
4927 Since the LTTng-UST Python agent adds a handler to the _root_
4928 logger, you can trace any log statement from any logger.
4930 .Use the LTTng-UST Python agent.
4941 logging.basicConfig()
4942 logger = logging.getLogger('my-logger')
4945 logger.debug('debug message')
4946 logger.info('info message')
4947 logger.warn('warn message')
4948 logger.error('error message')
4949 logger.critical('critical message')
4953 if __name__ == '__main__':
4957 NOTE: `logging.basicConfig()`, which adds to the root logger a basic
4958 logging handler which prints to the standard error stream, is not
4959 strictly required for LTTng-UST tracing to work, but in versions of
4960 Python preceding{nbsp}3.2, you could see a warning message which indicates
4961 that no handler exists for the logger `my-logger`.
4963 <<creating-destroying-tracing-sessions,Create a tracing session>>,
4964 <<enabling-disabling-events,create an event rule>> matching the
4965 `my-logger` Python logger, and <<basic-tracing-session-control,start
4971 $ lttng enable-event --python my-logger
4975 Run the Python script:
4982 <<basic-tracing-session-control,Stop tracing>> and inspect the recorded
4992 In the resulting trace, an <<event,event record>> generated by a Python
4993 application is named `lttng_python:event` and has the following fields:
4996 Logging time (string).
4999 Log record's message.
5005 Name of the function in which the log statement was executed.
5008 Line number at which the log statement was executed.
5011 Log level integer value.
5014 ID of the Python thread in which the log statement was executed.
5017 Name of the Python thread in which the log statement was executed.
5019 You can use the opt:lttng-enable-event(1):--loglevel or
5020 opt:lttng-enable-event(1):--loglevel-only option of the
5021 man:lttng-enable-event(1) command to target a range of Python log levels
5022 or a specific Python log level.
5024 When an application imports the LTTng-UST Python agent, the agent tries
5025 to register to a <<lttng-sessiond,session daemon>>. Note that you must
5026 <<start-sessiond,start the session daemon>> _before_ you run the Python
5027 application. If a session daemon is found, the agent tries to register
5028 to it during five seconds, after which the application continues
5029 without LTTng tracing support. You can override this timeout value with
5030 the env:LTTNG_UST_PYTHON_REGISTER_TIMEOUT environment variable
5033 If the session daemon stops while a Python application with an imported
5034 LTTng-UST Python agent runs, the agent retries to connect and to
5035 register to a session daemon every three seconds. You can override this
5036 delay with the env:LTTNG_UST_PYTHON_REGISTER_RETRY_DELAY environment
5041 [[proc-lttng-logger-abi]]
5044 The `lttng-tracer` Linux kernel module, part of
5045 <<lttng-modules,LTTng-modules>>, creates the special LTTng logger files
5046 path:{/proc/lttng-logger} and path:{/dev/lttng-logger} (since
5047 LTTng{nbsp}2.11) when it's loaded. Any application can write text data
5048 to any of those files to emit an LTTng event.
5051 .An application writes to the LTTng logger file to emit an LTTng event.
5052 image::lttng-logger.png[]
5054 The LTTng logger is the quickest method--not the most efficient,
5055 however--to add instrumentation to an application. It is designed
5056 mostly to instrument shell scripts:
5060 $ echo "Some message, some $variable" > /dev/lttng-logger
5063 Any event that the LTTng logger emits is named `lttng_logger` and
5064 belongs to the Linux kernel <<domain,tracing domain>>. However, unlike
5065 other instrumentation points in the kernel tracing domain, **any Unix
5066 user** can <<enabling-disabling-events,create an event rule>> which
5067 matches its event name, not only the root user or users in the
5068 <<tracing-group,tracing group>>.
5070 To use the LTTng logger:
5072 * From any application, write text data to the path:{/dev/lttng-logger}
5075 The `msg` field of `lttng_logger` event records contains the
5078 NOTE: The maximum message length of an LTTng logger event is
5079 1024{nbsp}bytes. Writing more than this makes the LTTng logger emit more
5080 than one event to contain the remaining data.
5082 You should not use the LTTng logger to trace a user application which
5083 can be instrumented in a more efficient way, namely:
5085 * <<c-application,C and $$C++$$ applications>>.
5086 * <<java-application,Java applications>>.
5087 * <<python-application,Python applications>>.
5089 .Use the LTTng logger.
5094 echo 'Hello, World!' > /dev/lttng-logger
5096 df --human-readable --print-type / > /dev/lttng-logger
5099 <<creating-destroying-tracing-sessions,Create a tracing session>>,
5100 <<enabling-disabling-events,create an event rule>> matching the
5101 `lttng_logger` Linux kernel tracepoint, and
5102 <<basic-tracing-session-control,start tracing>>:
5107 $ lttng enable-event --kernel lttng_logger
5111 Run the Bash script:
5118 <<basic-tracing-session-control,Stop tracing>> and inspect the recorded
5129 [[instrumenting-linux-kernel]]
5130 === LTTng kernel tracepoints
5132 NOTE: This section shows how to _add_ instrumentation points to the
5133 Linux kernel. The kernel's subsystems are already thoroughly
5134 instrumented at strategic places for LTTng when you
5135 <<installing-lttng,install>> the <<lttng-modules,LTTng-modules>>
5139 There are two methods to instrument the Linux kernel:
5141 . <<linux-add-lttng-layer,Add an LTTng layer>> over an existing ftrace
5142 tracepoint which uses the `TRACE_EVENT()` API.
5144 Choose this if you want to instrumentation a Linux kernel tree with an
5145 instrumentation point compatible with ftrace, perf, and SystemTap.
5147 . Use an <<linux-lttng-tracepoint-event,LTTng-only approach>> to
5148 instrument an out-of-tree kernel module.
5150 Choose this if you don't need ftrace, perf, or SystemTap support.
5154 [[linux-add-lttng-layer]]
5155 ==== [[instrumenting-linux-kernel-itself]][[mainline-trace-event]][[lttng-adaptation-layer]]Add an LTTng layer to an existing ftrace tracepoint
5157 This section shows how to add an LTTng layer to existing ftrace
5158 instrumentation using the `TRACE_EVENT()` API.
5160 This section does not document the `TRACE_EVENT()` macro. You can
5161 read the following articles to learn more about this API:
5163 * http://lwn.net/Articles/379903/[Using the TRACE_EVENT() macro (Part{nbsp}1)]
5164 * http://lwn.net/Articles/381064/[Using the TRACE_EVENT() macro (Part{nbsp}2)]
5165 * http://lwn.net/Articles/383362/[Using the TRACE_EVENT() macro (Part{nbsp}3)]
5167 The following procedure assumes that your ftrace tracepoints are
5168 correctly defined in their own header and that they are created in
5169 one source file using the `CREATE_TRACE_POINTS` definition.
5171 To add an LTTng layer over an existing ftrace tracepoint:
5173 . Make sure the following kernel configuration options are
5179 * `CONFIG_HIGH_RES_TIMERS`
5180 * `CONFIG_TRACEPOINTS`
5183 . Build the Linux source tree with your custom ftrace tracepoints.
5184 . Boot the resulting Linux image on your target system.
5186 Confirm that the tracepoints exist by looking for their names in the
5187 dir:{/sys/kernel/debug/tracing/events/subsys} directory, where `subsys`
5188 is your subsystem's name.
5190 . Get a copy of the latest LTTng-modules{nbsp}{revision}:
5195 $ cd $(mktemp -d) &&
5196 wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.11.tar.bz2 &&
5197 tar -xf lttng-modules-latest-2.11.tar.bz2 &&
5198 cd lttng-modules-2.11.*
5202 . In dir:{instrumentation/events/lttng-module}, relative to the root
5203 of the LTTng-modules source tree, create a header file named
5204 +__subsys__.h+ for your custom subsystem +__subsys__+ and write your
5205 LTTng-modules tracepoint definitions using the LTTng-modules
5208 Start with this template:
5212 .path:{instrumentation/events/lttng-module/my_subsys.h}
5215 #define TRACE_SYSTEM my_subsys
5217 #if !defined(_LTTNG_MY_SUBSYS_H) || defined(TRACE_HEADER_MULTI_READ)
5218 #define _LTTNG_MY_SUBSYS_H
5220 #include "../../../probes/lttng-tracepoint-event.h"
5221 #include <linux/tracepoint.h>
5223 LTTNG_TRACEPOINT_EVENT(
5225 * Format is identical to TRACE_EVENT()'s version for the three
5226 * following macro parameters:
5229 TP_PROTO(int my_int, const char *my_string),
5230 TP_ARGS(my_int, my_string),
5232 /* LTTng-modules specific macros */
5234 ctf_integer(int, my_int_field, my_int)
5235 ctf_string(my_bar_field, my_bar)
5239 #endif /* !defined(_LTTNG_MY_SUBSYS_H) || defined(TRACE_HEADER_MULTI_READ) */
5241 #include "../../../probes/define_trace.h"
5245 The entries in the `TP_FIELDS()` section are the list of fields for the
5246 LTTng tracepoint. This is similar to the `TP_STRUCT__entry()` part of
5247 ftrace's `TRACE_EVENT()` macro.
5249 See <<lttng-modules-tp-fields,Tracepoint fields macros>> for a
5250 complete description of the available `ctf_*()` macros.
5252 . Create the LTTng-modules probe's kernel module C source file,
5253 +probes/lttng-probe-__subsys__.c+, where +__subsys__+ is your
5258 .path:{probes/lttng-probe-my-subsys.c}
5260 #include <linux/module.h>
5261 #include "../lttng-tracer.h"
5264 * Build-time verification of mismatch between mainline
5265 * TRACE_EVENT() arguments and the LTTng-modules adaptation
5266 * layer LTTNG_TRACEPOINT_EVENT() arguments.
5268 #include <trace/events/my_subsys.h>
5270 /* Create LTTng tracepoint probes */
5271 #define LTTNG_PACKAGE_BUILD
5272 #define CREATE_TRACE_POINTS
5273 #define TRACE_INCLUDE_PATH ../instrumentation/events/lttng-module
5275 #include "../instrumentation/events/lttng-module/my_subsys.h"
5277 MODULE_LICENSE("GPL and additional rights");
5278 MODULE_AUTHOR("Your name <your-email>");
5279 MODULE_DESCRIPTION("LTTng my_subsys probes");
5280 MODULE_VERSION(__stringify(LTTNG_MODULES_MAJOR_VERSION) "."
5281 __stringify(LTTNG_MODULES_MINOR_VERSION) "."
5282 __stringify(LTTNG_MODULES_PATCHLEVEL_VERSION)
5283 LTTNG_MODULES_EXTRAVERSION);
5287 . Edit path:{probes/KBuild} and add your new kernel module object
5288 next to the existing ones:
5292 .path:{probes/KBuild}
5296 obj-m += lttng-probe-module.o
5297 obj-m += lttng-probe-power.o
5299 obj-m += lttng-probe-my-subsys.o
5305 . Build and install the LTTng kernel modules:
5310 $ make KERNELDIR=/path/to/linux
5311 # make modules_install && depmod -a
5315 Replace `/path/to/linux` with the path to the Linux source tree where
5316 you defined and used tracepoints with ftrace's `TRACE_EVENT()` macro.
5318 Note that you can also use the
5319 <<lttng-tracepoint-event-code,`LTTNG_TRACEPOINT_EVENT_CODE()` macro>>
5320 instead of `LTTNG_TRACEPOINT_EVENT()` to use custom local variables and
5321 C code that need to be executed before the event fields are recorded.
5323 The best way to learn how to use the previous LTTng-modules macros is to
5324 inspect the existing LTTng-modules tracepoint definitions in the
5325 dir:{instrumentation/events/lttng-module} header files. Compare them
5326 with the Linux kernel mainline versions in the
5327 dir:{include/trace/events} directory of the Linux source tree.
5331 [[lttng-tracepoint-event-code]]
5332 ===== Use custom C code to access the data for tracepoint fields
5334 Although we recommended to always use the
5335 <<lttng-adaptation-layer,`LTTNG_TRACEPOINT_EVENT()`>> macro to describe
5336 the arguments and fields of an LTTng-modules tracepoint when possible,
5337 sometimes you need a more complex process to access the data that the
5338 tracer records as event record fields. In other words, you need local
5339 variables and multiple C{nbsp}statements instead of simple
5340 argument-based expressions that you pass to the
5341 <<lttng-modules-tp-fields,`ctf_*()` macros of `TP_FIELDS()`>>.
5343 You can use the `LTTNG_TRACEPOINT_EVENT_CODE()` macro instead of
5344 `LTTNG_TRACEPOINT_EVENT()` to declare custom local variables and define
5345 a block of C{nbsp}code to be executed before LTTng records the fields.
5346 The structure of this macro is:
5349 .`LTTNG_TRACEPOINT_EVENT_CODE()` macro syntax.
5351 LTTNG_TRACEPOINT_EVENT_CODE(
5353 * Format identical to the LTTNG_TRACEPOINT_EVENT()
5354 * version for the following three macro parameters:
5357 TP_PROTO(int my_int, const char *my_string),
5358 TP_ARGS(my_int, my_string),
5360 /* Declarations of custom local variables */
5363 unsigned long b = 0;
5364 const char *name = "(undefined)";
5365 struct my_struct *my_struct;
5369 * Custom code which uses both tracepoint arguments
5370 * (in TP_ARGS()) and local variables (in TP_locvar()).
5372 * Local variables are actually members of a structure pointed
5373 * to by the special variable tp_locvar.
5377 tp_locvar->a = my_int + 17;
5378 tp_locvar->my_struct = get_my_struct_at(tp_locvar->a);
5379 tp_locvar->b = my_struct_compute_b(tp_locvar->my_struct);
5380 tp_locvar->name = my_struct_get_name(tp_locvar->my_struct);
5381 put_my_struct(tp_locvar->my_struct);
5390 * Format identical to the LTTNG_TRACEPOINT_EVENT()
5391 * version for this, except that tp_locvar members can be
5392 * used in the argument expression parameters of
5393 * the ctf_*() macros.
5396 ctf_integer(unsigned long, my_struct_b, tp_locvar->b)
5397 ctf_integer(int, my_struct_a, tp_locvar->a)
5398 ctf_string(my_string_field, my_string)
5399 ctf_string(my_struct_name, tp_locvar->name)
5404 IMPORTANT: The C code defined in `TP_code()` must not have any side
5405 effects when executed. In particular, the code must not allocate
5406 memory or get resources without deallocating this memory or putting
5407 those resources afterwards.
5410 [[instrumenting-linux-kernel-tracing]]
5411 ==== Load and unload a custom probe kernel module
5413 You must load a <<lttng-adaptation-layer,created LTTng-modules probe
5414 kernel module>> in the kernel before it can emit LTTng events.
5416 To load the default probe kernel modules and a custom probe kernel
5419 * Use the opt:lttng-sessiond(8):--extra-kmod-probes option to give extra
5420 probe modules to load when starting a root <<lttng-sessiond,session
5424 .Load the `my_subsys`, `usb`, and the default probe modules.
5428 # lttng-sessiond --extra-kmod-probes=my_subsys,usb
5433 You only need to pass the subsystem name, not the whole kernel module
5436 To load _only_ a given custom probe kernel module:
5438 * Use the opt:lttng-sessiond(8):--kmod-probes option to give the probe
5439 modules to load when starting a root session daemon:
5442 .Load only the `my_subsys` and `usb` probe modules.
5446 # lttng-sessiond --kmod-probes=my_subsys,usb
5451 To confirm that a probe module is loaded:
5458 $ lsmod | grep lttng_probe_usb
5462 To unload the loaded probe modules:
5464 * Kill the session daemon with `SIGTERM`:
5469 # pkill lttng-sessiond
5473 You can also use man:modprobe(8)'s `--remove` option if the session
5474 daemon terminates abnormally.
5477 [[controlling-tracing]]
5480 Once an application or a Linux kernel is
5481 <<instrumenting,instrumented>> for LTTng tracing,
5484 This section is divided in topics on how to use the various
5485 <<plumbing,components of LTTng>>, in particular the <<lttng-cli,cmd:lttng
5486 command-line tool>>, to _control_ the LTTng daemons and tracers.
5488 NOTE: In the following subsections, we refer to an man:lttng(1) command
5489 using its man page name. For example, instead of _Run the `create`
5490 command to..._, we use _Run the man:lttng-create(1) command to..._.
5494 === Start a session daemon
5496 In some situations, you need to run a <<lttng-sessiond,session daemon>>
5497 (man:lttng-sessiond(8)) _before_ you can use the man:lttng(1)
5500 You will see the following error when you run a command while no session
5504 Error: No session daemon is available
5507 The only command that automatically runs a session daemon is
5508 man:lttng-create(1), which you use to
5509 <<creating-destroying-tracing-sessions,create a tracing session>>. While
5510 this is most of the time the first operation that you do, sometimes it's
5511 not. Some examples are:
5513 * <<list-instrumentation-points,List the available instrumentation points>>.
5514 * <<saving-loading-tracing-session,Load a tracing session configuration>>.
5516 [[tracing-group]] Each Unix user must have its own running session
5517 daemon to trace user applications. The session daemon that the root user
5518 starts is the only one allowed to control the LTTng kernel tracer. Users
5519 that are part of the _tracing group_ can control the root session
5520 daemon. The default tracing group name is `tracing`; you can set it to
5521 something else with the opt:lttng-sessiond(8):--group option when you
5522 start the root session daemon.
5524 To start a user session daemon:
5526 * Run man:lttng-sessiond(8):
5531 $ lttng-sessiond --daemonize
5535 To start the root session daemon:
5537 * Run man:lttng-sessiond(8) as the root user:
5542 # lttng-sessiond --daemonize
5546 In both cases, remove the opt:lttng-sessiond(8):--daemonize option to
5547 start the session daemon in foreground.
5549 To stop a session daemon, use man:kill(1) on its process ID (standard
5552 Note that some Linux distributions could manage the LTTng session daemon
5553 as a service. In this case, you should use the service manager to
5554 start, restart, and stop session daemons.
5557 [[creating-destroying-tracing-sessions]]
5558 === Create and destroy a tracing session
5560 Almost all the LTTng control operations happen in the scope of
5561 a <<tracing-session,tracing session>>, which is the dialogue between the
5562 <<lttng-sessiond,session daemon>> and you.
5564 To create a tracing session with a generated name:
5566 * Use the man:lttng-create(1) command:
5575 The created tracing session's name is `auto` followed by the
5578 To create a tracing session with a specific name:
5580 * Use the optional argument of the man:lttng-create(1) command:
5585 $ lttng create my-session
5589 Replace `my-session` with the specific tracing session name.
5591 LTTng appends the creation date to the created tracing session's name.
5593 LTTng writes the traces of a tracing session in
5594 +$LTTNG_HOME/lttng-trace/__name__+ by default, where +__name__+ is the
5595 name of the tracing session. Note that the env:LTTNG_HOME environment
5596 variable defaults to `$HOME` if not set.
5598 To output LTTng traces to a non-default location:
5600 * Use the opt:lttng-create(1):--output option of the man:lttng-create(1) command:
5605 $ lttng create my-session --output=/tmp/some-directory
5609 You may create as many tracing sessions as you wish.
5611 To list all the existing tracing sessions for your Unix user:
5613 * Use the man:lttng-list(1) command:
5622 When you create a tracing session, it is set as the _current tracing
5623 session_. The following man:lttng(1) commands operate on the current
5624 tracing session when you don't specify one:
5626 [role="list-3-cols"]
5627 * man:lttng-add-context(1)
5628 * man:lttng-destroy(1)
5629 * man:lttng-disable-channel(1)
5630 * man:lttng-disable-event(1)
5631 * man:lttng-disable-rotation(1)
5632 * man:lttng-enable-channel(1)
5633 * man:lttng-enable-event(1)
5634 * man:lttng-enable-rotation(1)
5636 * man:lttng-regenerate(1)
5637 * man:lttng-rotate(1)
5639 * man:lttng-snapshot(1)
5640 * man:lttng-start(1)
5641 * man:lttng-status(1)
5643 * man:lttng-track(1)
5644 * man:lttng-untrack(1)
5647 To change the current tracing session:
5649 * Use the man:lttng-set-session(1) command:
5654 $ lttng set-session new-session
5658 Replace `new-session` by the name of the new current tracing session.
5660 When you are done tracing in a given tracing session, you can destroy
5661 it. This operation frees the resources taken by the tracing session
5662 to destroy; it does not destroy the trace data that LTTng wrote for
5663 this tracing session.
5665 To destroy the current tracing session:
5667 * Use the man:lttng-destroy(1) command:
5676 The man:lttng-destroy(1) command also runs the man:lttng-stop(1)
5677 command implicitly (see <<basic-tracing-session-control,Start and stop a
5678 tracing session>>). You need to stop tracing to make LTTng flush the
5679 remaining trace data and make the trace readable.
5682 [[list-instrumentation-points]]
5683 === List the available instrumentation points
5685 The <<lttng-sessiond,session daemon>> can query the running instrumented
5686 user applications and the Linux kernel to get a list of available
5687 instrumentation points. For the Linux kernel <<domain,tracing domain>>,
5688 they are tracepoints and system calls. For the user space tracing
5689 domain, they are tracepoints. For the other tracing domains, they are
5692 To list the available instrumentation points:
5694 * Use the man:lttng-list(1) command with the requested tracing domain's
5698 * opt:lttng-list(1):--kernel: Linux kernel tracepoints (your Unix user
5699 must be a root user, or it must be a member of the
5700 <<tracing-group,tracing group>>).
5701 * opt:lttng-list(1):--kernel with opt:lttng-list(1):--syscall: Linux
5702 kernel system calls (your Unix user must be a root user, or it must be
5703 a member of the tracing group).
5704 * opt:lttng-list(1):--userspace: user space tracepoints.
5705 * opt:lttng-list(1):--jul: `java.util.logging` loggers.
5706 * opt:lttng-list(1):--log4j: Apache log4j loggers.
5707 * opt:lttng-list(1):--python: Python loggers.
5710 .List the available user space tracepoints.
5714 $ lttng list --userspace
5718 .List the available Linux kernel system call tracepoints.
5722 $ lttng list --kernel --syscall
5727 [[enabling-disabling-events]]
5728 === Create and enable an event rule
5730 Once you <<creating-destroying-tracing-sessions,create a tracing
5731 session>>, you can create <<event,event rules>> with the
5732 man:lttng-enable-event(1) command.
5734 You specify each condition with a command-line option. The available
5735 condition arguments are shown in the following table.
5737 [role="growable",cols="asciidoc,asciidoc,default"]
5738 .Condition command-line arguments for the man:lttng-enable-event(1) command.
5740 |Argument |Description |Applicable tracing domains
5746 . +--probe=__ADDR__+
5747 . +--function=__ADDR__+
5748 . +--userspace-probe=__PATH__:__SYMBOL__+
5749 . +--userspace-probe=sdt:__PATH__:__PROVIDER__:__NAME__+
5752 Instead of using the default _tracepoint_ instrumentation type, use:
5754 . A Linux system call (entry and exit).
5755 . A Linux https://lwn.net/Articles/132196/[kprobe] (symbol or address).
5756 . The entry and return points of a Linux function (symbol or address).
5757 . The entry point of a user application or library function (path to
5758 application/library and symbol).
5759 . A https://www.sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps[SystemTap
5760 Statically Defined Tracing] (USDT) probe (path to application/library,
5761 provider and probe names).
5765 |First positional argument.
5768 Tracepoint or system call name.
5770 With the opt:lttng-enable-event(1):--probe,
5771 opt:lttng-enable-event(1):--function, and
5772 opt:lttng-enable-event(1):--userspace-probe options, this is a custom
5773 name given to the event rule. With the JUL, log4j, and Python domains,
5774 this is a logger name.
5776 With a tracepoint, logger, or system call name, you can use the special
5777 `*` globbing character to match anything (for example, `sched_*`,
5785 . +--loglevel=__LEVEL__+
5786 . +--loglevel-only=__LEVEL__+
5789 . Match only tracepoints or log statements with a logging level at
5790 least as severe as +__LEVEL__+.
5791 . Match only tracepoints or log statements with a logging level
5792 equal to +__LEVEL__+.
5794 See man:lttng-enable-event(1) for the list of available logging level
5797 |User space, JUL, log4j, and Python.
5799 |+--exclude=__EXCLUSIONS__+
5802 When you use a `*` character at the end of the tracepoint or logger
5803 name (first positional argument), exclude the specific names in the
5804 comma-delimited list +__EXCLUSIONS__+.
5807 User space, JUL, log4j, and Python.
5809 |+--filter=__EXPR__+
5812 Match only events which satisfy the expression +__EXPR__+.
5814 See man:lttng-enable-event(1) to learn more about the syntax of a
5821 You attach an event rule to a <<channel,channel>> on creation. If you do
5822 not specify the channel with the opt:lttng-enable-event(1):--channel
5823 option, and if the event rule to create is the first in its
5824 <<domain,tracing domain>> for a given tracing session, then LTTng
5825 creates a _default channel_ for you. This default channel is reused in
5826 subsequent invocations of the man:lttng-enable-event(1) command for the
5827 same tracing domain.
5829 An event rule is always enabled at creation time.
5831 The following examples show how you can combine the previous
5832 command-line options to create simple to more complex event rules.
5834 .Create an event rule targetting a Linux kernel tracepoint (default channel).
5838 $ lttng enable-event --kernel sched_switch
5842 .Create an event rule matching four Linux kernel system calls (default channel).
5846 $ lttng enable-event --kernel --syscall open,write,read,close
5850 .Create event rules matching tracepoints with filter expressions (default channel).
5854 $ lttng enable-event --kernel sched_switch --filter='prev_comm == "bash"'
5859 $ lttng enable-event --kernel --all \
5860 --filter='$ctx.tid == 1988 || $ctx.tid == 1534'
5865 $ lttng enable-event --jul my_logger \
5866 --filter='$app.retriever:cur_msg_id > 3'
5869 IMPORTANT: Make sure to always quote the filter string when you
5870 use man:lttng(1) from a shell.
5873 .Create an event rule matching any user space tracepoint of a given tracepoint provider with a log level range (default channel).
5877 $ lttng enable-event --userspace my_app:'*' --loglevel=TRACE_INFO
5880 IMPORTANT: Make sure to always quote the wildcard character when you
5881 use man:lttng(1) from a shell.
5884 .Create an event rule matching multiple Python loggers with a wildcard and with exclusions (default channel).
5888 $ lttng enable-event --python my-app.'*' \
5889 --exclude='my-app.module,my-app.hello'
5893 .Create an event rule matching any Apache log4j logger with a specific log level (default channel).
5897 $ lttng enable-event --log4j --all --loglevel-only=LOG4J_WARN
5901 .Create an event rule attached to a specific channel matching a specific user space tracepoint provider and tracepoint.
5905 $ lttng enable-event --userspace my_app:my_tracepoint --channel=my-channel
5909 .Create an event rule matching the `malloc` function entry in path:{/usr/lib/libc.so.6}:
5913 $ lttng enable-event --kernel --userspace-probe=/usr/lib/libc.so.6:malloc \
5918 .Create an event rule matching the `server`/`accept_request` https://www.sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps[USDT probe] in path:{/usr/bin/serv}:
5922 $ lttng enable-event --kernel --userspace-probe=sdt:serv:server:accept_request \
5923 server_accept_request
5927 The event rules of a given channel form a whitelist: as soon as an
5928 emitted event passes one of them, LTTng can record the event. For
5929 example, an event named `my_app:my_tracepoint` emitted from a user space
5930 tracepoint with a `TRACE_ERROR` log level passes both of the following
5935 $ lttng enable-event --userspace my_app:my_tracepoint
5936 $ lttng enable-event --userspace my_app:my_tracepoint \
5937 --loglevel=TRACE_INFO
5940 The second event rule is redundant: the first one includes
5944 [[disable-event-rule]]
5945 === Disable an event rule
5947 To disable an event rule that you <<enabling-disabling-events,created>>
5948 previously, use the man:lttng-disable-event(1) command. This command
5949 disables _all_ the event rules (of a given tracing domain and channel)
5950 which match an instrumentation point. The other conditions are not
5951 supported as of LTTng{nbsp}{revision}.
5953 The LTTng tracer does not record an emitted event which passes
5954 a _disabled_ event rule.
5956 .Disable an event rule matching a Python logger (default channel).
5960 $ lttng disable-event --python my-logger
5964 .Disable an event rule matching all `java.util.logging` loggers (default channel).
5968 $ lttng disable-event --jul '*'
5972 .Disable _all_ the event rules of the default channel.
5974 The opt:lttng-disable-event(1):--all-events option is not, like the
5975 opt:lttng-enable-event(1):--all option of man:lttng-enable-event(1), the
5976 equivalent of the event name `*` (wildcard): it disables _all_ the event
5977 rules of a given channel.
5981 $ lttng disable-event --jul --all-events
5985 NOTE: You cannot delete an event rule once you create it.
5989 === Get the status of a tracing session
5991 To get the status of the current tracing session, that is, its
5992 parameters, its channels, event rules, and their attributes:
5994 * Use the man:lttng-status(1) command:
6004 To get the status of any tracing session:
6006 * Use the man:lttng-list(1) command with the tracing session's name:
6011 $ lttng list my-session
6015 Replace `my-session` with the desired tracing session's name.
6018 [[basic-tracing-session-control]]
6019 === Start and stop a tracing session
6021 Once you <<creating-destroying-tracing-sessions,create a tracing
6023 <<enabling-disabling-events,create one or more event rules>>,
6024 you can start and stop the tracers for this tracing session.
6026 To start tracing in the current tracing session:
6028 * Use the man:lttng-start(1) command:
6037 LTTng is very flexible: you can launch user applications before
6038 or after the you start the tracers. The tracers only record the events
6039 if they pass enabled event rules and if they occur while the tracers are
6042 To stop tracing in the current tracing session:
6044 * Use the man:lttng-stop(1) command:
6053 If there were <<channel-overwrite-mode-vs-discard-mode,lost event
6054 records>> or lost sub-buffers since the last time you ran
6055 man:lttng-start(1), warnings are printed when you run the
6056 man:lttng-stop(1) command.
6058 IMPORTANT: You need to stop tracing to make LTTng flush the remaining
6059 trace data and make the trace readable. Note that the
6060 man:lttng-destroy(1) command (see
6061 <<creating-destroying-tracing-sessions,Create and destroy a tracing
6062 session>>) also runs the man:lttng-stop(1) command implicitly.
6065 [[enabling-disabling-channels]]
6066 === Create a channel
6068 Once you create a tracing session, you can create a <<channel,channel>>
6069 with the man:lttng-enable-channel(1) command.
6071 Note that LTTng automatically creates a default channel when, for a
6072 given <<domain,tracing domain>>, no channels exist and you
6073 <<enabling-disabling-events,create>> the first event rule. This default
6074 channel is named `channel0` and its attributes are set to reasonable
6075 values. Therefore, you only need to create a channel when you need
6076 non-default attributes.
6078 You specify each non-default channel attribute with a command-line
6079 option when you use the man:lttng-enable-channel(1) command. The
6080 available command-line options are:
6082 [role="growable",cols="asciidoc,asciidoc"]
6083 .Command-line options for the man:lttng-enable-channel(1) command.
6085 |Option |Description
6091 <<channel-overwrite-mode-vs-discard-mode,event record loss mode>> instead
6092 of the default _discard_ mode.
6094 |`--buffers-pid` (user space tracing domain only)
6097 Use the per-process <<channel-buffering-schemes,buffering scheme>>
6098 instead of the default per-user buffering scheme.
6100 |+--subbuf-size=__SIZE__+
6103 Allocate sub-buffers of +__SIZE__+ bytes (power of two), for each CPU,
6104 either for each Unix user (default), or for each instrumented process.
6106 See <<channel-subbuf-size-vs-subbuf-count,Sub-buffer count and size>>.
6108 |+--num-subbuf=__COUNT__+
6111 Allocate +__COUNT__+ sub-buffers (power of two), for each CPU, either
6112 for each Unix user (default), or for each instrumented process.
6114 See <<channel-subbuf-size-vs-subbuf-count,Sub-buffer count and size>>.
6116 |+--tracefile-size=__SIZE__+
6119 Set the maximum size of each trace file that this channel writes within
6120 a stream to +__SIZE__+ bytes instead of no maximum.
6122 See <<tracefile-rotation,Trace file count and size>>.
6124 |+--tracefile-count=__COUNT__+
6127 Limit the number of trace files that this channel creates to
6128 +__COUNT__+ channels instead of no limit.
6130 See <<tracefile-rotation,Trace file count and size>>.
6132 |+--switch-timer=__PERIODUS__+
6135 Set the <<channel-switch-timer,switch timer period>>
6136 to +__PERIODUS__+{nbsp}µs.
6138 |+--read-timer=__PERIODUS__+
6141 Set the <<channel-read-timer,read timer period>>
6142 to +__PERIODUS__+{nbsp}µs.
6144 |[[opt-blocking-timeout]]+--blocking-timeout=__TIMEOUTUS__+
6147 Set the timeout of user space applications which load LTTng-UST
6148 in blocking mode to +__TIMEOUTUS__+:
6151 Never block (non-blocking mode).
6154 Block forever until space is available in a sub-buffer to record
6157 __n__, a positive value::
6158 Wait for at most __n__ µs when trying to write into a sub-buffer.
6160 Note that, for this option to have any effect on an instrumented
6161 user space application, you need to run the application with a set
6162 env:LTTNG_UST_ALLOW_BLOCKING environment variable.
6164 |+--output=__TYPE__+ (Linux kernel tracing domain only)
6167 Set the channel's output type to +__TYPE__+, either `mmap` or `splice`.
6171 You can only create a channel in the Linux kernel and user space
6172 <<domain,tracing domains>>: other tracing domains have their own channel
6173 created on the fly when <<enabling-disabling-events,creating event
6178 Because of a current LTTng limitation, you must create all channels
6179 _before_ you <<basic-tracing-session-control,start tracing>> in a given
6180 tracing session, that is, before the first time you run
6183 Since LTTng automatically creates a default channel when you use the
6184 man:lttng-enable-event(1) command with a specific tracing domain, you
6185 cannot, for example, create a Linux kernel event rule, start tracing,
6186 and then create a user space event rule, because no user space channel
6187 exists yet and it's too late to create one.
6189 For this reason, make sure to configure your channels properly
6190 before starting the tracers for the first time!
6193 The following examples show how you can combine the previous
6194 command-line options to create simple to more complex channels.
6196 .Create a Linux kernel channel with default attributes.
6200 $ lttng enable-channel --kernel my-channel
6204 .Create a user space channel with four sub-buffers or 1{nbsp}MiB each, per CPU, per instrumented process.
6208 $ lttng enable-channel --userspace --num-subbuf=4 --subbuf-size=1M \
6209 --buffers-pid my-channel
6213 .[[blocking-timeout-example]]Create a default user space channel with an infinite blocking timeout.
6215 <<creating-destroying-tracing-sessions,Create a tracing-session>>,
6216 create the channel, <<enabling-disabling-events,create an event rule>>,
6217 and <<basic-tracing-session-control,start tracing>>:
6222 $ lttng enable-channel --userspace --blocking-timeout=inf blocking-channel
6223 $ lttng enable-event --userspace --channel=blocking-channel --all
6227 Run an application instrumented with LTTng-UST and allow it to block:
6231 $ LTTNG_UST_ALLOW_BLOCKING=1 my-app
6235 .Create a Linux kernel channel which rotates eight trace files of 4{nbsp}MiB each for each stream
6239 $ lttng enable-channel --kernel --tracefile-count=8 \
6240 --tracefile-size=4194304 my-channel
6244 .Create a user space channel in overwrite (or _flight recorder_) mode.
6248 $ lttng enable-channel --userspace --overwrite my-channel
6252 You can <<enabling-disabling-events,create>> the same event rule in
6253 two different channels:
6257 $ lttng enable-event --userspace --channel=my-channel app:tp
6258 $ lttng enable-event --userspace --channel=other-channel app:tp
6261 If both channels are enabled, when a tracepoint named `app:tp` is
6262 reached, LTTng records two events, one for each channel.
6266 === Disable a channel
6268 To disable a specific channel that you <<enabling-disabling-channels,created>>
6269 previously, use the man:lttng-disable-channel(1) command.
6271 .Disable a specific Linux kernel channel.
6275 $ lttng disable-channel --kernel my-channel
6279 The state of a channel precedes the individual states of event rules
6280 attached to it: event rules which belong to a disabled channel, even if
6281 they are enabled, are also considered disabled.
6285 === Add context fields to a channel
6287 Event record fields in trace files provide important information about
6288 events that occured previously, but sometimes some external context may
6289 help you solve a problem faster. Examples of context fields are:
6291 * The **process ID**, **thread ID**, **process name**, and
6292 **process priority** of the thread in which the event occurs.
6293 * The **hostname** of the system on which the event occurs.
6294 * The Linux kernel and user call stacks (since
6296 * The current values of many possible **performance counters** using
6298 ** CPU cycles, stalled cycles, idle cycles, and the other cycle types.
6300 ** Branch instructions, misses, and loads.
6302 * Any context defined at the application level (supported for the
6303 JUL and log4j <<domain,tracing domains>>).
6305 To get the full list of available context fields, see
6306 `lttng add-context --list`. Some context fields are reserved for a
6307 specific <<domain,tracing domain>> (Linux kernel or user space).
6309 You add context fields to <<channel,channels>>. All the events
6310 that a channel with added context fields records contain those fields.
6312 To add context fields to one or all the channels of a given tracing
6315 * Use the man:lttng-add-context(1) command.
6317 .Add context fields to all the channels of the current tracing session.
6319 The following command line adds the virtual process identifier and
6320 the per-thread CPU cycles count fields to all the user space channels
6321 of the current tracing session.
6325 $ lttng add-context --userspace --type=vpid --type=perf:thread:cpu-cycles
6329 .Add performance counter context fields by raw ID
6331 See man:lttng-add-context(1) for the exact format of the context field
6332 type, which is partly compatible with the format used in
6337 $ lttng add-context --userspace --type=perf:thread:raw:r0110:test
6338 $ lttng add-context --kernel --type=perf:cpu:raw:r0013c:x86unhalted
6342 .Add context fields to a specific channel.
6344 The following command line adds the thread identifier and user call
6345 stack context fields to the Linux kernel channel named `my-channel` in
6346 the current tracing session.
6350 $ lttng add-context --kernel --channel=my-channel \
6351 --type=tid --type=callstack-user
6355 .Add an application-specific context field to a specific channel.
6357 The following command line adds the `cur_msg_id` context field of the
6358 `retriever` context retriever for all the instrumented
6359 <<java-application,Java applications>> recording <<event,event records>>
6360 in the channel named `my-channel`:
6364 $ lttng add-context --kernel --channel=my-channel \
6365 --type='$app:retriever:cur_msg_id'
6368 IMPORTANT: Make sure to always quote the `$` character when you
6369 use man:lttng-add-context(1) from a shell.
6372 NOTE: You cannot remove context fields from a channel once you add it.
6377 === Track process IDs
6379 It's often useful to allow only specific process IDs (PIDs) to emit
6380 events. For example, you may wish to record all the system calls made by
6381 a given process (Ă la http://linux.die.net/man/1/strace[strace]).
6383 The man:lttng-track(1) and man:lttng-untrack(1) commands serve this
6384 purpose. Both commands operate on a whitelist of process IDs. You _add_
6385 entries to this whitelist with the man:lttng-track(1) command and remove
6386 entries with the man:lttng-untrack(1) command. Any process which has one
6387 of the PIDs in the whitelist is allowed to emit LTTng events which pass
6388 an enabled <<event,event rule>>.
6390 NOTE: The PID tracker tracks the _numeric process IDs_. Should a
6391 process with a given tracked ID exit and another process be given this
6392 ID, then the latter would also be allowed to emit events.
6394 .Track and untrack process IDs.
6396 For the sake of the following example, assume the target system has
6397 16{nbsp}possible PIDs.
6400 <<creating-destroying-tracing-sessions,create a tracing session>>,
6401 the whitelist contains all the possible PIDs:
6404 .All PIDs are tracked.
6405 image::track-all.png[]
6407 When the whitelist is full and you use the man:lttng-track(1) command to
6408 specify some PIDs to track, LTTng first clears the whitelist, then it
6409 tracks the specific PIDs. After:
6413 $ lttng track --pid=3,4,7,10,13
6419 .PIDs 3, 4, 7, 10, and 13 are tracked.
6420 image::track-3-4-7-10-13.png[]
6422 You can add more PIDs to the whitelist afterwards:
6426 $ lttng track --pid=1,15,16
6432 .PIDs 1, 15, and 16 are added to the whitelist.
6433 image::track-1-3-4-7-10-13-15-16.png[]
6435 The man:lttng-untrack(1) command removes entries from the PID tracker's
6436 whitelist. Given the previous example, the following command:
6440 $ lttng untrack --pid=3,7,10,13
6443 leads to this whitelist:
6446 .PIDs 3, 7, 10, and 13 are removed from the whitelist.
6447 image::track-1-4-15-16.png[]
6449 LTTng can track all possible PIDs again using the
6450 opt:lttng-track(1):--all option:
6454 $ lttng track --pid --all
6457 The result is, again:
6460 .All PIDs are tracked.
6461 image::track-all.png[]
6464 .Track only specific PIDs
6466 A very typical use case with PID tracking is to start with an empty
6467 whitelist, then <<basic-tracing-session-control,start the tracers>>, and
6468 then add PIDs manually while tracers are active. You can accomplish this
6469 by using the opt:lttng-untrack(1):--all option of the
6470 man:lttng-untrack(1) command to clear the whitelist after you
6471 <<creating-destroying-tracing-sessions,create a tracing session>>:
6475 $ lttng untrack --pid --all
6481 .No PIDs are tracked.
6482 image::untrack-all.png[]
6484 If you trace with this whitelist configuration, the tracer records no
6485 events for this <<domain,tracing domain>> because no processes are
6486 tracked. You can use the man:lttng-track(1) command as usual to track
6487 specific PIDs, for example:
6491 $ lttng track --pid=6,11
6497 .PIDs 6 and 11 are tracked.
6498 image::track-6-11.png[]
6503 [[saving-loading-tracing-session]]
6504 === Save and load tracing session configurations
6506 Configuring a <<tracing-session,tracing session>> can be long. Some of
6507 the tasks involved are:
6509 * <<enabling-disabling-channels,Create channels>> with
6510 specific attributes.
6511 * <<adding-context,Add context fields>> to specific channels.
6512 * <<enabling-disabling-events,Create event rules>> with specific log
6513 level and filter conditions.
6515 If you use LTTng to solve real world problems, chances are you have to
6516 record events using the same tracing session setup over and over,
6517 modifying a few variables each time in your instrumented program
6518 or environment. To avoid constant tracing session reconfiguration,
6519 the man:lttng(1) command-line tool can save and load tracing session
6520 configurations to/from XML files.
6522 To save a given tracing session configuration:
6524 * Use the man:lttng-save(1) command:
6529 $ lttng save my-session
6533 Replace `my-session` with the name of the tracing session to save.
6535 LTTng saves tracing session configurations to
6536 dir:{$LTTNG_HOME/.lttng/sessions} by default. Note that the
6537 env:LTTNG_HOME environment variable defaults to `$HOME` if not set. Use
6538 the opt:lttng-save(1):--output-path option to change this destination
6541 LTTng saves all configuration parameters, for example:
6543 * The tracing session name.
6544 * The trace data output path.
6545 * The channels with their state and all their attributes.
6546 * The context fields you added to channels.
6547 * The event rules with their state, log level and filter conditions.
6549 To load a tracing session:
6551 * Use the man:lttng-load(1) command:
6556 $ lttng load my-session
6560 Replace `my-session` with the name of the tracing session to load.
6562 When LTTng loads a configuration, it restores your saved tracing session
6563 as if you just configured it manually.
6565 See man:lttng-load(1) for the complete list of command-line options. You
6566 can also save and load many sessions at a time, and decide in which
6567 directory to output the XML files.
6570 [[sending-trace-data-over-the-network]]
6571 === Send trace data over the network
6573 LTTng can send the recorded trace data to a remote system over the
6574 network instead of writing it to the local file system.
6576 To send the trace data over the network:
6578 . On the _remote_ system (which can also be the target system),
6579 start an LTTng <<lttng-relayd,relay daemon>> (man:lttng-relayd(8)):
6588 . On the _target_ system, create a tracing session configured to
6589 send trace data over the network:
6594 $ lttng create my-session --set-url=net://remote-system
6598 Replace `remote-system` by the host name or IP address of the
6599 remote system. See man:lttng-create(1) for the exact URL format.
6601 . On the target system, use the man:lttng(1) command-line tool as usual.
6602 When tracing is active, the target's consumer daemon sends sub-buffers
6603 to the relay daemon running on the remote system instead of flushing
6604 them to the local file system. The relay daemon writes the received
6605 packets to the local file system.
6607 The relay daemon writes trace files to
6608 +$LTTNG_HOME/lttng-traces/__hostname__/__session__+ by default, where
6609 +__hostname__+ is the host name of the target system and +__session__+
6610 is the tracing session name. Note that the env:LTTNG_HOME environment
6611 variable defaults to `$HOME` if not set. Use the
6612 opt:lttng-relayd(8):--output option of man:lttng-relayd(8) to write
6613 trace files to another base directory.
6618 === View events as LTTng emits them (noch:{LTTng} live)
6620 LTTng live is a network protocol implemented by the <<lttng-relayd,relay
6621 daemon>> (man:lttng-relayd(8)) to allow compatible trace viewers to
6622 display events as LTTng emits them on the target system while tracing is
6625 The relay daemon creates a _tee_: it forwards the trace data to both
6626 the local file system and to connected live viewers:
6629 .The relay daemon creates a _tee_, forwarding the trace data to both trace files and a connected live viewer.
6634 . On the _target system_, create a <<tracing-session,tracing session>>
6640 $ lttng create my-session --live
6644 This spawns a local relay daemon.
6646 . Start the live viewer and configure it to connect to the relay
6647 daemon. For example, with http://diamon.org/babeltrace[Babeltrace]:
6652 $ babeltrace --input-format=lttng-live \
6653 net://localhost/host/hostname/my-session
6660 * `hostname` with the host name of the target system.
6661 * `my-session` with the name of the tracing session to view.
6664 . Configure the tracing session as usual with the man:lttng(1)
6665 command-line tool, and <<basic-tracing-session-control,start tracing>>.
6667 You can list the available live tracing sessions with Babeltrace:
6671 $ babeltrace --input-format=lttng-live net://localhost
6674 You can start the relay daemon on another system. In this case, you need
6675 to specify the relay daemon's URL when you create the tracing session
6676 with the opt:lttng-create(1):--set-url option. You also need to replace
6677 `localhost` in the procedure above with the host name of the system on
6678 which the relay daemon is running.
6680 See man:lttng-create(1) and man:lttng-relayd(8) for the complete list of
6681 command-line options.
6685 [[taking-a-snapshot]]
6686 === Take a snapshot of the current sub-buffers of a tracing session
6688 The normal behavior of LTTng is to append full sub-buffers to growing
6689 trace data files. This is ideal to keep a full history of the events
6690 that occurred on the target system, but it can
6691 represent too much data in some situations. For example, you may wish
6692 to trace your application continuously until some critical situation
6693 happens, in which case you only need the latest few recorded
6694 events to perform the desired analysis, not multi-gigabyte trace files.
6696 With the man:lttng-snapshot(1) command, you can take a snapshot of the
6697 current sub-buffers of a given <<tracing-session,tracing session>>.
6698 LTTng can write the snapshot to the local file system or send it over
6702 .A snapshot is a copy of the current sub-buffers, which are not cleared after the operation.
6703 image::snapshot.png[]
6705 If you wish to create unmanaged, self-contained, non-overlapping
6706 trace chunk archives instead of a simple copy of the current
6707 sub-buffers, see the <<session-rotation,tracing session rotation>>
6708 feature (available since LTTng{nbsp}2.11).
6712 . Create a tracing session in _snapshot mode_:
6717 $ lttng create my-session --snapshot
6721 The <<channel-overwrite-mode-vs-discard-mode,event record loss mode>> of
6722 <<channel,channels>> created in this mode is automatically set to
6723 _overwrite_ (flight recorder mode).
6725 . Configure the tracing session as usual with the man:lttng(1)
6726 command-line tool, and <<basic-tracing-session-control,start tracing>>.
6728 . **Optional**: When you need to take a snapshot,
6729 <<basic-tracing-session-control,stop tracing>>.
6731 You can take a snapshot when the tracers are active, but if you stop
6732 them first, you are sure that the data in the sub-buffers does not
6733 change before you actually take the snapshot.
6740 $ lttng snapshot record --name=my-first-snapshot
6744 LTTng writes the current sub-buffers of all the current tracing
6745 session's channels to trace files on the local file system. Those trace
6746 files have `my-first-snapshot` in their name.
6748 There is no difference between the format of a normal trace file and the
6749 format of a snapshot: viewers of LTTng traces also support LTTng
6752 By default, LTTng writes snapshot files to the path shown by
6753 `lttng snapshot list-output`. You can change this path or decide to send
6754 snapshots over the network using either:
6756 . An output path or URL that you specify when you
6757 <<creating-destroying-tracing-sessions,create the tracing session>>.
6758 . A snapshot output path or URL that you add using
6759 `lttng snapshot add-output`.
6760 . An output path or URL that you provide directly to the
6761 `lttng snapshot record` command.
6763 Method{nbsp}3 overrides method{nbsp}2, which overrides method 1. When
6764 you specify a URL, a relay daemon must listen on a remote system (see
6765 <<sending-trace-data-over-the-network,Send trace data over the
6770 [[session-rotation]]
6771 === Archive the current trace chunk (rotate a tracing session)
6773 The <<taking-a-snapshot,snapshot user guide>> shows how you can dump
6774 a tracing session's current sub-buffers to the file system or send them
6775 over the network. When you take a snapshot, LTTng does not clear the
6776 tracing session's ring buffers: if you take another snapshot immediately
6777 after, both snapshots could contain overlapping trace data.
6779 Inspired by https://en.wikipedia.org/wiki/Log_rotation[log rotation],
6780 _tracing session rotation_ is a feature which appends the content of the
6781 ring buffers to what's already on the file system or sent over the
6782 network since the tracing session's creation or since the last
6783 rotation, and then clears those ring buffers to avoid trace data
6786 What LTTng is about to write when performing a tracing session rotation
6787 is called the _current trace chunk_. When this current trace chunk is
6788 written to the file system or sent over the network, it becomes a _trace
6789 chunk archive_. Therefore, a tracing session rotation _archives_ the
6790 current trace chunk.
6793 .A tracing session rotation operation _archives_ the current trace chunk.
6794 image::rotation.png[]
6796 A trace chunk archive is a self-contained LTTng trace which LTTng
6797 doesn't manage anymore: you can read it, modify it, move it, or remove
6800 There are two methods to perform a tracing session rotation: immediately
6801 or with a rotation schedule.
6803 To perform an immediate tracing session rotation:
6805 . <<creating-destroying-tracing-sessions,Create a tracing session>>
6806 in _normal mode_ or _network streaming mode_
6807 (only those two creation modes support tracing session rotation):
6812 $ lttng create my-session
6816 . <<enabling-disabling-events,Create one or more event rules>>
6817 and <<basic-tracing-session-control,start tracing>>:
6822 $ lttng enable-event --kernel sched_'*'
6827 . When needed, immediately rotate the current tracing session:
6836 The cmd:lttng-rotate command prints the path to the created trace
6837 chunk archive. See man:lttng-rotate(1) to learn about the format
6838 of trace chunk archive directory names.
6840 You can perform other immediate rotations while the tracing session is
6841 active. It is guaranteed that all the trace chunk archives do not
6842 contain overlapping trace data. You can also perform an immediate
6843 rotation once you have <<basic-tracing-session-control,stopped>> the
6846 . When you are done tracing,
6847 <<creating-destroying-tracing-sessions,destroy the current tracing
6857 The tracing session destruction operation creates one last trace
6858 chunk archive from the current trace chunk.
6860 A tracing session rotation schedule is a planned rotation which LTTng
6861 performs automatically based on one of the following conditions:
6863 * A timer with a configured period times out.
6865 * The total size of the flushed part of the current trace chunk
6866 becomes greater than or equal to a configured value.
6868 To schedule a tracing session rotation, set a _rotation schedule_:
6870 . <<creating-destroying-tracing-sessions,Create a tracing session>>
6871 in _normal mode_ or _network streaming mode_
6872 (only those two creation modes support tracing session rotation):
6877 $ lttng create my-session
6881 . <<enabling-disabling-events,Create one or more event rules>>:
6886 $ lttng enable-event --kernel sched_'*'
6890 . Set a tracing session rotation schedule:
6895 $ lttng enable-rotation --timer=10s
6899 In this example, we set a rotation schedule so that LTTng performs a
6900 tracing session rotation every ten seconds.
6902 See man:lttng-enable-rotation(1) to learn more about other ways to set a
6905 . <<basic-tracing-session-control,Start tracing>>:
6914 LTTng performs tracing session rotations automatically while the tracing
6915 session is active thanks to the rotation schedule.
6917 . When you are done tracing,
6918 <<creating-destroying-tracing-sessions,destroy the current tracing
6928 The tracing session destruction operation creates one last trace chunk
6929 archive from the current trace chunk.
6931 You can use man:lttng-disable-rotation(1) to unset a tracing session
6934 NOTE: man:lttng-rotate(1) and man:lttng-enable-rotation(1) list
6935 limitations regarding those two commands.
6940 === Use the machine interface
6942 With any command of the man:lttng(1) command-line tool, you can set the
6943 opt:lttng(1):--mi option to `xml` (before the command name) to get an
6944 XML machine interface output, for example:
6948 $ lttng --mi=xml enable-event --kernel --syscall open
6951 A schema definition (XSD) is
6952 https://github.com/lttng/lttng-tools/blob/stable-2.11/src/common/mi-lttng-3.0.xsd[available]
6953 to ease the integration with external tools as much as possible.
6957 [[metadata-regenerate]]
6958 === Regenerate the metadata of an LTTng trace
6960 An LTTng trace, which is a http://diamon.org/ctf[CTF] trace, has both
6961 data stream files and a metadata file. This metadata file contains,
6962 amongst other things, information about the offset of the clock sources
6963 used to timestamp <<event,event records>> when tracing.
6965 If, once a <<tracing-session,tracing session>> is
6966 <<basic-tracing-session-control,started>>, a major
6967 https://en.wikipedia.org/wiki/Network_Time_Protocol[NTP] correction
6968 happens, the trace's clock offset also needs to be updated. You
6969 can use the `metadata` item of the man:lttng-regenerate(1) command
6972 The main use case of this command is to allow a system to boot with
6973 an incorrect wall time and trace it with LTTng before its wall time
6974 is corrected. Once the system is known to be in a state where its
6975 wall time is correct, it can run `lttng regenerate metadata`.
6977 To regenerate the metadata of an LTTng trace:
6979 * Use the `metadata` item of the man:lttng-regenerate(1) command:
6984 $ lttng regenerate metadata
6990 `lttng regenerate metadata` has the following limitations:
6992 * Tracing session <<creating-destroying-tracing-sessions,created>>
6994 * User space <<channel,channels>>, if any, are using
6995 <<channel-buffering-schemes,per-user buffering>>.
7000 [[regenerate-statedump]]
7001 === Regenerate the state dump of a tracing session
7003 The LTTng kernel and user space tracers generate state dump
7004 <<event,event records>> when the application starts or when you
7005 <<basic-tracing-session-control,start a tracing session>>. An analysis
7006 can use the state dump event records to set an initial state before it
7007 builds the rest of the state from the following event records.
7008 http://tracecompass.org/[Trace Compass] is a notable example of an
7009 application which uses the state dump of an LTTng trace.
7011 When you <<taking-a-snapshot,take a snapshot>>, it's possible that the
7012 state dump event records are not included in the snapshot because they
7013 were recorded to a sub-buffer that has been consumed or overwritten
7016 You can use the `lttng regenerate statedump` command to emit the state
7017 dump event records again.
7019 To regenerate the state dump of the current tracing session, provided
7020 create it in snapshot mode, before you take a snapshot:
7022 . Use the `statedump` item of the man:lttng-regenerate(1) command:
7027 $ lttng regenerate statedump
7031 . <<basic-tracing-session-control,Stop the tracing session>>:
7040 . <<taking-a-snapshot,Take a snapshot>>:
7045 $ lttng snapshot record --name=my-snapshot
7049 Depending on the event throughput, you should run steps 1 and 2
7050 as closely as possible.
7052 NOTE: To record the state dump events, you need to
7053 <<enabling-disabling-events,create event rules>> which enable them.
7054 LTTng-UST state dump tracepoints start with `lttng_ust_statedump:`.
7055 LTTng-modules state dump tracepoints start with `lttng_statedump_`.
7059 [[persistent-memory-file-systems]]
7060 === Record trace data on persistent memory file systems
7062 https://en.wikipedia.org/wiki/Non-volatile_random-access_memory[Non-volatile random-access memory]
7063 (NVRAM) is random-access memory that retains its information when power
7064 is turned off (non-volatile). Systems with such memory can store data
7065 structures in RAM and retrieve them after a reboot, without flushing
7066 to typical _storage_.
7068 Linux supports NVRAM file systems thanks to either
7069 http://pramfs.sourceforge.net/[PRAMFS] or
7070 https://www.kernel.org/doc/Documentation/filesystems/dax.txt[DAX]{nbsp}+{nbsp}http://lkml.iu.edu/hypermail/linux/kernel/1504.1/03463.html[pmem]
7071 (requires Linux{nbsp}4.1+).
7073 This section does not describe how to operate such file systems;
7074 we assume that you have a working persistent memory file system.
7076 When you create a <<tracing-session,tracing session>>, you can specify
7077 the path of the shared memory holding the sub-buffers. If you specify a
7078 location on an NVRAM file system, then you can retrieve the latest
7079 recorded trace data when the system reboots after a crash.
7081 To record trace data on a persistent memory file system and retrieve the
7082 trace data after a system crash:
7084 . Create a tracing session with a sub-buffer shared memory path located
7085 on an NVRAM file system:
7090 $ lttng create my-session --shm-path=/path/to/shm
7094 . Configure the tracing session as usual with the man:lttng(1)
7095 command-line tool, and <<basic-tracing-session-control,start tracing>>.
7097 . After a system crash, use the man:lttng-crash(1) command-line tool to
7098 view the trace data recorded on the NVRAM file system:
7103 $ lttng-crash /path/to/shm
7107 The binary layout of the ring buffer files is not exactly the same as
7108 the trace files layout. This is why you need to use man:lttng-crash(1)
7109 instead of your preferred trace viewer directly.
7111 To convert the ring buffer files to LTTng trace files:
7113 * Use the opt:lttng-crash(1):--extract option of man:lttng-crash(1):
7118 $ lttng-crash --extract=/path/to/trace /path/to/shm
7124 [[notif-trigger-api]]
7125 === Get notified when a channel's buffer usage is too high or too low
7127 With LTTng's $$C/C++$$ notification and trigger API, your user
7128 application can get notified when the buffer usage of one or more
7129 <<channel,channels>> becomes too low or too high. You can use this API
7130 and enable or disable <<event,event rules>> during tracing to avoid
7131 <<channel-overwrite-mode-vs-discard-mode,discarded event records>>.
7133 .Have a user application get notified when an LTTng channel's buffer usage is too high.
7135 In this example, we create and build an application which gets notified
7136 when the buffer usage of a specific LTTng channel is higher than
7137 75{nbsp}%. We only print that it is the case in the example, but we
7138 could as well use the API of <<liblttng-ctl-lttng,`liblttng-ctl`>> to
7139 disable event rules when this happens.
7141 . Create the application's C source file:
7149 #include <lttng/domain.h>
7150 #include <lttng/action/action.h>
7151 #include <lttng/action/notify.h>
7152 #include <lttng/condition/condition.h>
7153 #include <lttng/condition/buffer-usage.h>
7154 #include <lttng/condition/evaluation.h>
7155 #include <lttng/notification/channel.h>
7156 #include <lttng/notification/notification.h>
7157 #include <lttng/trigger/trigger.h>
7158 #include <lttng/endpoint.h>
7160 int main(int argc, char *argv[])
7162 int exit_status = 0;
7163 struct lttng_notification_channel *notification_channel;
7164 struct lttng_condition *condition;
7165 struct lttng_action *action;
7166 struct lttng_trigger *trigger;
7167 const char *tracing_session_name;
7168 const char *channel_name;
7171 tracing_session_name = argv[1];
7172 channel_name = argv[2];
7175 * Create a notification channel. A notification channel
7176 * connects the user application to the LTTng session daemon.
7177 * This notification channel can be used to listen to various
7178 * types of notifications.
7180 notification_channel = lttng_notification_channel_create(
7181 lttng_session_daemon_notification_endpoint);
7184 * Create a "high buffer usage" condition. In this case, the
7185 * condition is reached when the buffer usage is greater than or
7186 * equal to 75 %. We create the condition for a specific tracing
7187 * session name, channel name, and for the user space tracing
7190 * The "low buffer usage" condition type also exists.
7192 condition = lttng_condition_buffer_usage_high_create();
7193 lttng_condition_buffer_usage_set_threshold_ratio(condition, .75);
7194 lttng_condition_buffer_usage_set_session_name(
7195 condition, tracing_session_name);
7196 lttng_condition_buffer_usage_set_channel_name(condition,
7198 lttng_condition_buffer_usage_set_domain_type(condition,
7202 * Create an action (get a notification) to take when the
7203 * condition created above is reached.
7205 action = lttng_action_notify_create();
7208 * Create a trigger. A trigger associates a condition to an
7209 * action: the action is executed when the condition is reached.
7211 trigger = lttng_trigger_create(condition, action);
7213 /* Register the trigger to LTTng. */
7214 lttng_register_trigger(trigger);
7217 * Now that we have registered a trigger, a notification will be
7218 * emitted everytime its condition is met. To receive this
7219 * notification, we must subscribe to notifications that match
7220 * the same condition.
7222 lttng_notification_channel_subscribe(notification_channel,
7226 * Notification loop. You can put this in a dedicated thread to
7227 * avoid blocking the main thread.
7230 struct lttng_notification *notification;
7231 enum lttng_notification_channel_status status;
7232 const struct lttng_evaluation *notification_evaluation;
7233 const struct lttng_condition *notification_condition;
7234 double buffer_usage;
7236 /* Receive the next notification. */
7237 status = lttng_notification_channel_get_next_notification(
7238 notification_channel, ¬ification);
7241 case LTTNG_NOTIFICATION_CHANNEL_STATUS_OK:
7243 case LTTNG_NOTIFICATION_CHANNEL_STATUS_NOTIFICATIONS_DROPPED:
7245 * The session daemon can drop notifications if
7246 * a monitoring application is not consuming the
7247 * notifications fast enough.
7250 case LTTNG_NOTIFICATION_CHANNEL_STATUS_CLOSED:
7252 * The notification channel has been closed by the
7253 * session daemon. This is typically caused by a session
7254 * daemon shutting down.
7258 /* Unhandled conditions or errors. */
7264 * A notification provides, amongst other things:
7266 * * The condition that caused this notification to be
7268 * * The condition evaluation, which provides more
7269 * specific information on the evaluation of the
7272 * The condition evaluation provides the buffer usage
7273 * value at the moment the condition was reached.
7275 notification_condition = lttng_notification_get_condition(
7277 notification_evaluation = lttng_notification_get_evaluation(
7280 /* We're subscribed to only one condition. */
7281 assert(lttng_condition_get_type(notification_condition) ==
7282 LTTNG_CONDITION_TYPE_BUFFER_USAGE_HIGH);
7285 * Get the exact sampled buffer usage from the
7286 * condition evaluation.
7288 lttng_evaluation_buffer_usage_get_usage_ratio(
7289 notification_evaluation, &buffer_usage);
7292 * At this point, instead of printing a message, we
7293 * could do something to reduce the channel's buffer
7294 * usage, like disable specific events.
7296 printf("Buffer usage is %f %% in tracing session \"%s\", "
7297 "user space channel \"%s\".\n", buffer_usage * 100,
7298 tracing_session_name, channel_name);
7299 lttng_notification_destroy(notification);
7303 lttng_action_destroy(action);
7304 lttng_condition_destroy(condition);
7305 lttng_trigger_destroy(trigger);
7306 lttng_notification_channel_destroy(notification_channel);
7312 . Build the `notif-app` application, linking it to `liblttng-ctl`:
7317 $ gcc -o notif-app notif-app.c -llttng-ctl
7321 . <<creating-destroying-tracing-sessions,Create a tracing session>>,
7322 <<enabling-disabling-events,create an event rule>> matching all the
7323 user space tracepoints, and
7324 <<basic-tracing-session-control,start tracing>>:
7329 $ lttng create my-session
7330 $ lttng enable-event --userspace --all
7335 If you create the channel manually with the man:lttng-enable-channel(1)
7336 command, you can control how frequently are the current values of the
7337 channel's properties sampled to evaluate user conditions with the
7338 opt:lttng-enable-channel(1):--monitor-timer option.
7340 . Run the `notif-app` application. This program accepts the
7341 <<tracing-session,tracing session>> name and the user space channel
7342 name as its two first arguments. The channel which LTTng automatically
7343 creates with the man:lttng-enable-event(1) command above is named
7349 $ ./notif-app my-session channel0
7353 . In another terminal, run an application with a very high event
7354 throughput so that the 75{nbsp}% buffer usage condition is reached.
7356 In the first terminal, the application should print lines like this:
7359 Buffer usage is 81.45197 % in tracing session "my-session", user space
7363 If you don't see anything, try modifying the condition in
7364 path:{notif-app.c} to a lower value (0.1, for example), rebuilding it
7365 (step{nbsp}2) and running it again (step{nbsp}4).
7372 [[lttng-modules-ref]]
7373 === noch:{LTTng-modules}
7377 [[lttng-tracepoint-enum]]
7378 ==== `LTTNG_TRACEPOINT_ENUM()` usage
7380 Use the `LTTNG_TRACEPOINT_ENUM()` macro to define an enumeration:
7384 LTTNG_TRACEPOINT_ENUM(name, TP_ENUM_VALUES(entries))
7389 * `name` with the name of the enumeration (C identifier, unique
7390 amongst all the defined enumerations).
7391 * `entries` with a list of enumeration entries.
7393 The available enumeration entry macros are:
7395 +ctf_enum_value(__name__, __value__)+::
7396 Entry named +__name__+ mapped to the integral value +__value__+.
7398 +ctf_enum_range(__name__, __begin__, __end__)+::
7399 Entry named +__name__+ mapped to the range of integral values between
7400 +__begin__+ (included) and +__end__+ (included).
7402 +ctf_enum_auto(__name__)+::
7403 Entry named +__name__+ mapped to the integral value following the
7404 last mapping's value.
7406 The last value of a `ctf_enum_value()` entry is its +__value__+
7409 The last value of a `ctf_enum_range()` entry is its +__end__+ parameter.
7411 If `ctf_enum_auto()` is the first entry in the list, its integral
7414 Use the `ctf_enum()` <<lttng-modules-tp-fields,field definition macro>>
7415 to use a defined enumeration as a tracepoint field.
7417 .Define an enumeration with `LTTNG_TRACEPOINT_ENUM()`.
7421 LTTNG_TRACEPOINT_ENUM(
7424 ctf_enum_auto("AUTO: EXPECT 0")
7425 ctf_enum_value("VALUE: 23", 23)
7426 ctf_enum_value("VALUE: 27", 27)
7427 ctf_enum_auto("AUTO: EXPECT 28")
7428 ctf_enum_range("RANGE: 101 TO 303", 101, 303)
7429 ctf_enum_auto("AUTO: EXPECT 304")
7437 [[lttng-modules-tp-fields]]
7438 ==== Tracepoint fields macros (for `TP_FIELDS()`)
7440 [[tp-fast-assign]][[tp-struct-entry]]The available macros to define
7441 tracepoint fields, which must be listed within `TP_FIELDS()` in
7442 `LTTNG_TRACEPOINT_EVENT()`, are:
7444 [role="func-desc growable",cols="asciidoc,asciidoc"]
7445 .Available macros to define LTTng-modules tracepoint fields
7447 |Macro |Description and parameters
7450 +ctf_integer(__t__, __n__, __e__)+
7452 +ctf_integer_nowrite(__t__, __n__, __e__)+
7454 +ctf_user_integer(__t__, __n__, __e__)+
7456 +ctf_user_integer_nowrite(__t__, __n__, __e__)+
7458 Standard integer, displayed in base{nbsp}10.
7461 Integer C type (`int`, `long`, `size_t`, ...).
7467 Argument expression.
7470 +ctf_integer_hex(__t__, __n__, __e__)+
7472 +ctf_user_integer_hex(__t__, __n__, __e__)+
7474 Standard integer, displayed in base{nbsp}16.
7483 Argument expression.
7485 |+ctf_integer_oct(__t__, __n__, __e__)+
7487 Standard integer, displayed in base{nbsp}8.
7496 Argument expression.
7499 +ctf_integer_network(__t__, __n__, __e__)+
7501 +ctf_user_integer_network(__t__, __n__, __e__)+
7503 Integer in network byte order (big-endian), displayed in base{nbsp}10.
7512 Argument expression.
7515 +ctf_integer_network_hex(__t__, __n__, __e__)+
7517 +ctf_user_integer_network_hex(__t__, __n__, __e__)+
7519 Integer in network byte order, displayed in base{nbsp}16.
7528 Argument expression.
7531 +ctf_enum(__N__, __t__, __n__, __e__)+
7533 +ctf_enum_nowrite(__N__, __t__, __n__, __e__)+
7535 +ctf_user_enum(__N__, __t__, __n__, __e__)+
7537 +ctf_user_enum_nowrite(__N__, __t__, __n__, __e__)+
7542 Name of a <<lttng-tracepoint-enum,previously defined enumeration>>.
7545 Integer C type (`int`, `long`, `size_t`, ...).
7551 Argument expression.
7554 +ctf_string(__n__, __e__)+
7556 +ctf_string_nowrite(__n__, __e__)+
7558 +ctf_user_string(__n__, __e__)+
7560 +ctf_user_string_nowrite(__n__, __e__)+
7562 Null-terminated string; undefined behavior if +__e__+ is `NULL`.
7568 Argument expression.
7571 +ctf_array(__t__, __n__, __e__, __s__)+
7573 +ctf_array_nowrite(__t__, __n__, __e__, __s__)+
7575 +ctf_user_array(__t__, __n__, __e__, __s__)+
7577 +ctf_user_array_nowrite(__t__, __n__, __e__, __s__)+
7579 Statically-sized array of integers.
7582 Array element C type.
7588 Argument expression.
7594 +ctf_array_bitfield(__t__, __n__, __e__, __s__)+
7596 +ctf_array_bitfield_nowrite(__t__, __n__, __e__, __s__)+
7598 +ctf_user_array_bitfield(__t__, __n__, __e__, __s__)+
7600 +ctf_user_array_bitfield_nowrite(__t__, __n__, __e__, __s__)+
7602 Statically-sized array of bits.
7604 The type of +__e__+ must be an integer type. +__s__+ is the number
7605 of elements of such type in +__e__+, not the number of bits.
7608 Array element C type.
7614 Argument expression.
7620 +ctf_array_text(__t__, __n__, __e__, __s__)+
7622 +ctf_array_text_nowrite(__t__, __n__, __e__, __s__)+
7624 +ctf_user_array_text(__t__, __n__, __e__, __s__)+
7626 +ctf_user_array_text_nowrite(__t__, __n__, __e__, __s__)+
7628 Statically-sized array, printed as text.
7630 The string does not need to be null-terminated.
7633 Array element C type (always `char`).
7639 Argument expression.
7645 +ctf_sequence(__t__, __n__, __e__, __T__, __E__)+
7647 +ctf_sequence_nowrite(__t__, __n__, __e__, __T__, __E__)+
7649 +ctf_user_sequence(__t__, __n__, __e__, __T__, __E__)+
7651 +ctf_user_sequence_nowrite(__t__, __n__, __e__, __T__, __E__)+
7653 Dynamically-sized array of integers.
7655 The type of +__E__+ must be unsigned.
7658 Array element C type.
7664 Argument expression.
7667 Length expression C type.
7673 +ctf_sequence_hex(__t__, __n__, __e__, __T__, __E__)+
7675 +ctf_user_sequence_hex(__t__, __n__, __e__, __T__, __E__)+
7677 Dynamically-sized array of integers, displayed in base{nbsp}16.
7679 The type of +__E__+ must be unsigned.
7682 Array element C type.
7688 Argument expression.
7691 Length expression C type.
7696 |+ctf_sequence_network(__t__, __n__, __e__, __T__, __E__)+
7698 Dynamically-sized array of integers in network byte order (big-endian),
7699 displayed in base{nbsp}10.
7701 The type of +__E__+ must be unsigned.
7704 Array element C type.
7710 Argument expression.
7713 Length expression C type.
7719 +ctf_sequence_bitfield(__t__, __n__, __e__, __T__, __E__)+
7721 +ctf_sequence_bitfield_nowrite(__t__, __n__, __e__, __T__, __E__)+
7723 +ctf_user_sequence_bitfield(__t__, __n__, __e__, __T__, __E__)+
7725 +ctf_user_sequence_bitfield_nowrite(__t__, __n__, __e__, __T__, __E__)+
7727 Dynamically-sized array of bits.
7729 The type of +__e__+ must be an integer type. +__s__+ is the number
7730 of elements of such type in +__e__+, not the number of bits.
7732 The type of +__E__+ must be unsigned.
7735 Array element C type.
7741 Argument expression.
7744 Length expression C type.
7750 +ctf_sequence_text(__t__, __n__, __e__, __T__, __E__)+
7752 +ctf_sequence_text_nowrite(__t__, __n__, __e__, __T__, __E__)+
7754 +ctf_user_sequence_text(__t__, __n__, __e__, __T__, __E__)+
7756 +ctf_user_sequence_text_nowrite(__t__, __n__, __e__, __T__, __E__)+
7758 Dynamically-sized array, displayed as text.
7760 The string does not need to be null-terminated.
7762 The type of +__E__+ must be unsigned.
7764 The behaviour is undefined if +__e__+ is `NULL`.
7767 Sequence element C type (always `char`).
7773 Argument expression.
7776 Length expression C type.
7782 Use the `_user` versions when the argument expression, `e`, is
7783 a user space address. In the cases of `ctf_user_integer*()` and
7784 `ctf_user_float*()`, `&e` must be a user space address, thus `e` must
7787 The `_nowrite` versions omit themselves from the session trace, but are
7788 otherwise identical. This means the `_nowrite` fields won't be written
7789 in the recorded trace. Their primary purpose is to make some
7790 of the event context available to the
7791 <<enabling-disabling-events,event filters>> without having to
7792 commit the data to sub-buffers.
7798 Terms related to LTTng and to tracing in general:
7801 The http://diamon.org/babeltrace[Babeltrace] project, which includes:
7803 * The cmd:babeltrace (Babeltrace{nbsp}1) or cmd:babeltrace2
7804 (Babeltrace{nbsp}2) command.
7805 * Libraries with a C{nbsp}API.
7806 * Python{nbsp}3 bindings.
7807 * Plugins (Babeltrace{nbsp}2).
7809 [[def-buffering-scheme]]<<channel-buffering-schemes,buffering scheme>>::
7810 A layout of <<def-sub-buffer,sub-buffers>> applied to a given channel.
7812 [[def-channel]]<<channel,channel>>::
7813 An entity which is responsible for a set of
7814 <<def-ring-buffer,ring buffers>>.
7816 <<def-event-rule,Event rules>> are always attached to a specific
7820 A source of time for a <<def-tracer,tracer>>.
7822 [[def-consumer-daemon]]<<lttng-consumerd,consumer daemon>>::
7823 A process which is responsible for consuming the full
7824 <<def-sub-buffer,sub-buffers>> and write them to a file system or
7825 send them over the network.
7827 [[def-current-trace-chunk]]current trace chunk::
7828 A <<def-trace-chunk,trace chunk>> which includes the current content
7829 of all the <<def-tracing-session-rotation,tracing session>>'s
7830 <<def-sub-buffer,sub-buffers>> and the stream files produced since the
7831 latest event amongst:
7833 * The creation of the <<def-tracing-session,tracing session>>.
7834 * The last tracing session rotation, if any.
7836 <<channel-overwrite-mode-vs-discard-mode,discard mode>>::
7837 The <<def-event-record-loss-mode,event record loss mode>> in which
7838 the <<def-tracer,tracer>> _discards_ new event records when there's no
7839 <<def-sub-buffer,sub-buffer>> space left to store them.
7841 [[def-event]]event::
7842 The consequence of the execution of an
7843 <<def-instrumentation-point,instrumentation point>>, like a
7844 <<def-tracepoint,tracepoint>> that you manually place in some source
7845 code, or a Linux kernel kprobe.
7847 An event is said to _occur_ at a specific time. <<def-lttng,LTTng>> can
7848 take various actions upon the occurrence of an event, like record the
7849 event's payload to a <<def-sub-buffer,sub-buffer>>.
7851 [[def-event-name]]event name::
7852 The name of an <<def-event,event>>, which is also the name of the
7853 <<def-event-record,event record>>.
7855 This is also called the _instrumentation point name_.
7857 [[def-event-record]]event record::
7858 A record, in a <<def-trace,trace>>, of the payload of an
7859 <<def-event,event>> which occured.
7861 [[def-event-record-loss-mode]]<<channel-overwrite-mode-vs-discard-mode,event record loss mode>>::
7862 The mechanism by which event records of a given
7863 <<def-channel,channel>> are lost (not recorded) when there is no
7864 <<def-sub-buffer,sub-buffer>> space left to store them.
7866 [[def-event-rule]]<<event,event rule>>::
7867 Set of conditions which must be satisfied for one or more occuring
7868 <<def-event,events>> to be recorded.
7870 `java.util.logging`::
7872 https://docs.oracle.com/javase/7/docs/api/java/util/logging/package-summary.html[core logging facilities].
7874 <<instrumenting,instrumentation>>::
7875 The use of <<def-lttng,LTTng>> probes to make a piece of software
7878 [[def-instrumentation-point]]instrumentation point::
7879 A point in the execution path of a piece of software that, when
7880 reached by this execution, can emit an <<def-event,event>>.
7882 instrumentation point name::
7883 See _<<def-event-name,event name>>_.
7886 A http://logging.apache.org/log4j/1.2/[logging library] for Java
7887 developed by the Apache Software Foundation.
7890 Level of severity of a log statement or user space
7891 <<def-instrumentation-point,instrumentation point>>.
7893 [[def-lttng]]LTTng::
7894 The _Linux Trace Toolkit: next generation_ project.
7896 <<lttng-cli,cmd:lttng>>::
7897 A command-line tool provided by the <<def-lttng-tools,LTTng-tools>>
7898 project which you can use to send and receive control messages to and
7899 from a <<def-session-daemon,session daemon>>.
7902 The https://github.com/lttng/lttng-analyses[LTTng analyses] project,
7903 which is a set of analyzing programs that you can use to obtain a
7904 higher level view of an <<def-lttng,LTTng>> <<def-trace,trace>>.
7906 cmd:lttng-consumerd::
7907 The name of the <<def-consumer-daemon,consumer daemon>> program.
7910 A utility provided by the <<def-lttng-tools,LTTng-tools>> project
7911 which can convert <<def-ring-buffer,ring buffer>> files (usually
7912 <<persistent-memory-file-systems,saved on a persistent memory file
7913 system>>) to <<def-trace,trace>> files.
7915 See man:lttng-crash(1).
7917 LTTng Documentation::
7920 <<lttng-live,LTTng live>>::
7921 A communication protocol between the <<lttng-relayd,relay daemon>> and
7922 live viewers which makes it possible to see <<def-event-record,event
7923 records>> "live", as they are received by the
7924 <<def-relay-daemon,relay daemon>>.
7926 <<lttng-modules,LTTng-modules>>::
7927 The https://github.com/lttng/lttng-modules[LTTng-modules] project,
7928 which contains the Linux kernel modules to make the Linux kernel
7929 <<def-instrumentation-point,instrumentation points>> available for
7930 <<def-lttng,LTTng>> tracing.
7933 The name of the <<def-relay-daemon,relay daemon>> program.
7935 cmd:lttng-sessiond::
7936 The name of the <<def-session-daemon,session daemon>> program.
7938 [[def-lttng-tools]]LTTng-tools::
7939 The https://github.com/lttng/lttng-tools[LTTng-tools] project, which
7940 contains the various programs and libraries used to
7941 <<controlling-tracing,control tracing>>.
7943 [[def-lttng-ust]]<<lttng-ust,LTTng-UST>>::
7944 The https://github.com/lttng/lttng-ust[LTTng-UST] project, which
7945 contains libraries to instrument
7946 <<def-user-application,user applications>>.
7948 <<lttng-ust-agents,LTTng-UST Java agent>>::
7949 A Java package provided by the <<def-lttng-ust,LTTng-UST>> project to
7950 allow the LTTng instrumentation of `java.util.logging` and Apache
7951 log4j{nbsp}1.2 logging statements.
7953 <<lttng-ust-agents,LTTng-UST Python agent>>::
7954 A Python package provided by the <<def-lttng-ust,LTTng-UST>> project
7955 to allow the <<def-lttng,LTTng>> instrumentation of Python logging
7958 <<channel-overwrite-mode-vs-discard-mode,overwrite mode>>::
7959 The <<def-event-record-loss-mode,event record loss mode>> in which new
7960 <<def-event-record,event records>> _overwrite_ older event records
7961 when there's no <<def-sub-buffer,sub-buffer>> space left to store
7964 <<channel-buffering-schemes,per-process buffering>>::
7965 A <<def-buffering-scheme,buffering scheme>> in which each instrumented
7966 process has its own <<def-sub-buffer,sub-buffers>> for a given user
7967 space <<def-channel,channel>>.
7969 <<channel-buffering-schemes,per-user buffering>>::
7970 A <<def-buffering-scheme,buffering scheme>> in which all the processes
7971 of a Unix user share the same <<def-sub-buffer,sub-buffers>> for a
7972 given user space <<def-channel,channel>>.
7974 [[def-relay-daemon]]<<lttng-relayd,relay daemon>>::
7975 A process which is responsible for receiving the <<def-trace,trace>>
7976 data which a distant <<def-consumer-daemon,consumer daemon>> sends.
7978 [[def-ring-buffer]]ring buffer::
7979 A set of <<def-sub-buffer,sub-buffers>>.
7982 See _<<def-tracing-session-rotation,tracing session rotation>>_.
7984 [[def-session-daemon]]<<lttng-sessiond,session daemon>>::
7985 A process which receives control commands from you and orchestrates
7986 the <<def-tracer,tracers>> and various <<def-lttng,LTTng>> daemons.
7988 <<taking-a-snapshot,snapshot>>::
7989 A copy of the current data of all the <<def-sub-buffer,sub-buffers>>
7990 of a given <<def-tracing-session,tracing session>>, saved as
7991 <<def-trace,trace>> files.
7993 [[def-sub-buffer]]sub-buffer::
7994 One part of an <<def-lttng,LTTng>> <<def-ring-buffer,ring buffer>>
7995 which contains <<def-event-record,event records>>.
7998 The time information attached to an <<def-event,event>> when it is
8001 [[def-trace]]trace (_noun_)::
8004 * One http://diamon.org/ctf/[CTF] metadata stream file.
8005 * One or more CTF data stream files which are the concatenations of one
8006 or more flushed <<def-sub-buffer,sub-buffers>>.
8008 [[def-trace-verb]]trace (_verb_)::
8009 The action of recording the <<def-event,events>> emitted by an
8010 application or by a system, or to initiate such recording by
8011 controlling a <<def-tracer,tracer>>.
8013 [[def-trace-chunk]]trace chunk::
8014 A self-contained <<def-trace,trace>> which is part of a
8015 <<def-tracing-session,tracing session>>. Each
8016 <<def-tracing-session-rotation, tracing session rotation>> produces a
8017 <<def-trace-chunk-archive,trace chunk archive>>.
8019 [[def-trace-chunk-archive]]trace chunk archive::
8020 The result of a <<def-tracing-session-rotation, tracing session rotation>>.
8022 <<def-lttng,LTTng>> does not manage any trace chunk archive, even if its
8023 containing <<def-tracing-session,tracing session>> is still active: you
8024 are free to read it, modify it, move it, or remove it.
8027 The http://tracecompass.org[Trace Compass] project and application.
8029 [[def-tracepoint]]tracepoint::
8030 An instrumentation point using the tracepoint mechanism of the Linux
8031 kernel or of <<def-lttng-ust,LTTng-UST>>.
8033 tracepoint definition::
8034 The definition of a single <<def-tracepoint,tracepoint>>.
8037 The name of a <<def-tracepoint,tracepoint>>.
8039 [[def-tracepoint-provider]]tracepoint provider::
8040 A set of functions providing <<def-tracepoint,tracepoints>> to an
8041 instrumented <<def-user-application,user application>>.
8043 Not to be confused with a <<def-tracepoint-provider-package,tracepoint
8044 provider package>>: many tracepoint providers can exist within a
8045 tracepoint provider package.
8047 [[def-tracepoint-provider-package]]tracepoint provider package::
8048 One or more <<def-tracepoint-provider,tracepoint providers>> compiled
8049 as an https://en.wikipedia.org/wiki/Object_file[object file] or as a
8050 link:https://en.wikipedia.org/wiki/Library_(computing)#Shared_libraries[shared
8053 [[def-tracer]]tracer::
8054 A software which records emitted <<def-event,events>>.
8056 <<domain,tracing domain>>::
8057 A namespace for <<def-event,event>> sources.
8059 <<tracing-group,tracing group>>::
8060 The Unix group in which a Unix user can be to be allowed to
8061 <<def-trace-verb,trace>> the Linux kernel.
8063 [[def-tracing-session]]<<tracing-session,tracing session>>::
8064 A stateful dialogue between you and a <<lttng-sessiond,session daemon>>.
8066 [[def-tracing-session-rotation]]<<session-rotation,tracing session rotation>>::
8067 The action of archiving the
8068 <<def-current-trace-chunk,current trace chunk>> of a
8069 <<def-tracing-session,tracing session>>.
8071 [[def-user-application]]user application::
8072 An application running in user space, as opposed to a Linux kernel
8073 module, for example.