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 {revision}?
77 LTTng{nbsp}{revision} bears the name _Joannès_. A Berliner Weisse style
78 beer from the http://letreflenoir.com/[Trèfle Noir] microbrewery in
79 https://en.wikipedia.org/wiki/Rouyn-Noranda[Rouyn-Noranda], the
80 https://www.beeradvocate.com/beer/profile/20537/238967/[_**Joannès**_]
81 is a tangy beer with a distinct pink dress and intense fruit flavor,
82 thanks to the presence of fresh blackcurrant grown in Témiscamingue.
84 New features and changes in LTTng{nbsp}{revision}:
86 * **Tracing control**:
87 ** You can override the name or the URL of a tracing session
88 configuration when you use man:lttng-load(1) thanks to the new
89 opt:lttng-load(1):--override-name and
90 opt:lttng-load(1):--override-url options.
91 ** The new `lttng regenerate` command replaces the now deprecated
92 `lttng metadata` command of LTTng 2.8. man:lttng-regenerate(1) can
93 also <<regenerate-statedump,generate the state dump event records>>
94 of a given tracing session on demand, a handy feature when
95 <<taking-a-snapshot,taking a snapshot>>.
96 ** You can add PMU counters by raw ID with man:lttng-add-context(1):
101 $ lttng add-context --kernel --type=perf:cpu:raw:r0013c:x86unhalted
105 The format of the raw ID is the same as used with man:perf-record(1).
106 See <<adding-context,Add context fields to a channel>> for more
109 ** The LTTng <<lttng-relayd,relay daemon>> is now supported on
110 OS{nbsp}X and macOS for a smoother integration within a trace
111 analysis workflow, regardless of the platform used.
113 * **User space tracing**:
114 ** Improved performance (tested on x86-64 and ARMv7-A
115 (https://en.wikipedia.org/wiki/Cubieboard[Cubieboard])
117 ** New helper library (`liblttng-ust-fd`) to help with
118 <<liblttng-ust-fd,applications which close file descriptors that
119 don't belong to them>>, for example, in a loop which closes file
120 descriptors after man:fork(2), or BSD's `closeall()`.
121 ** More accurate <<liblttng-ust-dl,dynamic linker instrumentation>> and
122 state dump event records, especially when a dynamically loaded
123 library manually loads its own dependencies.
124 ** New `ctf_*()` field definition macros (see man:lttng-ust(3)):
125 *** `ctf_array_hex()`
126 *** `ctf_array_network()`
127 *** `ctf_array_network_hex()`
128 *** `ctf_sequence_hex()`
129 *** `ctf_sequence_network()`
130 *** `ctf_sequence_network_hex()`
131 ** New `lttng_ust_loaded` weak symbol defined by `liblttng-ust` for
132 an application to know if the LTTng-UST shared library is loaded
140 int lttng_ust_loaded __attribute__((weak));
144 if (lttng_ust_loaded) {
145 puts("LTTng-UST is loaded!");
147 puts("LTTng-UST is not loaded!");
155 ** LTTng-UST thread names have the `-ust` suffix.
157 * **Linux kernel tracing**:
158 ** Improved performance (tested on x86-64 and ARMv7-A
159 (https://en.wikipedia.org/wiki/Cubieboard[Cubieboard])
161 ** New enumeration <<lttng-modules-tp-fields,field definition macros>>:
162 `ctf_enum()` and `ctf_user_enum()`.
163 ** IPv4, IPv6, and TCP header data is recorded in the event records
164 produced by tracepoints starting with `net_`.
165 ** Detailed system call event records: `select`, `pselect6`, `poll`,
166 `ppoll`, `epoll_wait`, `epoll_pwait`, and `epoll_ctl` on all
167 architectures supported by LTTng-modules, and `accept4` on x86-64.
168 ** New I²C instrumentation: the `extract_sensitive_payload` parameter
169 of the new `lttng-probe-i2c` LTTng module controls whether or not
170 the payloads of I²C messages are recorded in I²C event records, since
171 they may contain sensitive data (for example, keystrokes).
172 ** When the LTTng kernel modules are built into the Linux kernel image,
173 the `CONFIG_TRACEPOINTS` configuration option is automatically
180 What is LTTng? As its name suggests, the _Linux Trace Toolkit: next
181 generation_ is a modern toolkit for tracing Linux systems and
182 applications. So your first question might be:
189 As the history of software engineering progressed and led to what
190 we now take for granted--complex, numerous and
191 interdependent software applications running in parallel on
192 sophisticated operating systems like Linux--the authors of such
193 components, software developers, began feeling a natural
194 urge to have tools that would ensure the robustness and good performance
195 of their masterpieces.
197 One major achievement in this field is, inarguably, the
198 https://www.gnu.org/software/gdb/[GNU debugger (GDB)],
199 an essential tool for developers to find and fix bugs. But even the best
200 debugger won't help make your software run faster, and nowadays, faster
201 software means either more work done by the same hardware, or cheaper
202 hardware for the same work.
204 A _profiler_ is often the tool of choice to identify performance
205 bottlenecks. Profiling is suitable to identify _where_ performance is
206 lost in a given software. The profiler outputs a profile, a statistical
207 summary of observed events, which you may use to discover which
208 functions took the most time to execute. However, a profiler won't
209 report _why_ some identified functions are the bottleneck. Bottlenecks
210 might only occur when specific conditions are met, conditions that are
211 sometimes impossible to capture by a statistical profiler, or impossible
212 to reproduce with an application altered by the overhead of an
213 event-based profiler. For a thorough investigation of software
214 performance issues, a history of execution is essential, with the
215 recorded values of variables and context fields you choose, and
216 with as little influence as possible on the instrumented software. This
217 is where tracing comes in handy.
219 _Tracing_ is a technique used to understand what goes on in a running
220 software system. The software used for tracing is called a _tracer_,
221 which is conceptually similar to a tape recorder. When recording,
222 specific instrumentation points placed in the software source code
223 generate events that are saved on a giant tape: a _trace_ file. You
224 can trace user applications and the operating system at the same time,
225 opening the possibility of resolving a wide range of problems that would
226 otherwise be extremely challenging.
228 Tracing is often compared to _logging_. However, tracers and loggers are
229 two different tools, serving two different purposes. Tracers are
230 designed to record much lower-level events that occur much more
231 frequently than log messages, often in the range of thousands per
232 second, with very little execution overhead. Logging is more appropriate
233 for a very high-level analysis of less frequent events: user accesses,
234 exceptional conditions (errors and warnings, for example), database
235 transactions, instant messaging communications, and such. Simply put,
236 logging is one of the many use cases that can be satisfied with tracing.
238 The list of recorded events inside a trace file can be read manually
239 like a log file for the maximum level of detail, but it is generally
240 much more interesting to perform application-specific analyses to
241 produce reduced statistics and graphs that are useful to resolve a
242 given problem. Trace viewers and analyzers are specialized tools
245 In the end, this is what LTTng is: a powerful, open source set of
246 tools to trace the Linux kernel and user applications at the same time.
247 LTTng is composed of several components actively maintained and
248 developed by its link:/community/#where[community].
251 [[lttng-alternatives]]
252 === Alternatives to noch:{LTTng}
254 Excluding proprietary solutions, a few competing software tracers
257 * https://github.com/dtrace4linux/linux[dtrace4linux] is a port of
258 Sun Microsystems's DTrace to Linux. The cmd:dtrace tool interprets
259 user scripts and is responsible for loading code into the
260 Linux kernel for further execution and collecting the outputted data.
261 * https://en.wikipedia.org/wiki/Berkeley_Packet_Filter[eBPF] is a
262 subsystem in the Linux kernel in which a virtual machine can execute
263 programs passed from the user space to the kernel. You can attach
264 such programs to tracepoints and KProbes thanks to a system call, and
265 they can output data to the user space when executed thanks to
266 different mechanisms (pipe, VM register values, and eBPF maps, to name
268 * https://www.kernel.org/doc/Documentation/trace/ftrace.txt[ftrace]
269 is the de facto function tracer of the Linux kernel. Its user
270 interface is a set of special files in sysfs.
271 * https://perf.wiki.kernel.org/[perf] is
272 a performance analyzing tool for Linux which supports hardware
273 performance counters, tracepoints, as well as other counters and
274 types of probes. perf's controlling utility is the cmd:perf command
276 * http://linux.die.net/man/1/strace[strace]
277 is a command-line utility which records system calls made by a
278 user process, as well as signal deliveries and changes of process
279 state. strace makes use of https://en.wikipedia.org/wiki/Ptrace[ptrace]
280 to fulfill its function.
281 * http://www.sysdig.org/[sysdig], like SystemTap, uses scripts to
282 analyze Linux kernel events. You write scripts, or _chisels_ in
283 sysdig's jargon, in Lua and sysdig executes them while the system is
284 being traced or afterwards. sysdig's interface is the cmd:sysdig
285 command-line tool as well as the curses-based cmd:csysdig tool.
286 * https://sourceware.org/systemtap/[SystemTap] is a Linux kernel and
287 user space tracer which uses custom user scripts to produce plain text
288 traces. SystemTap converts the scripts to the C language, and then
289 compiles them as Linux kernel modules which are loaded to produce
290 trace data. SystemTap's primary user interface is the cmd:stap
293 The main distinctive features of LTTng is that it produces correlated
294 kernel and user space traces, as well as doing so with the lowest
295 overhead amongst other solutions. It produces trace files in the
296 http://diamon.org/ctf[CTF] format, a file format optimized
297 for the production and analyses of multi-gigabyte data.
299 LTTng is the result of more than 10 years of active open source
300 development by a community of passionate developers.
301 LTTng{nbsp}{revision} is currently available on major desktop and server
304 The main interface for tracing control is a single command-line tool
305 named cmd:lttng. The latter can create several tracing sessions, enable
306 and disable events on the fly, filter events efficiently with custom
307 user expressions, start and stop tracing, and much more. LTTng can
308 record the traces on the file system or send them over the network, and
309 keep them totally or partially. You can view the traces once tracing
310 becomes inactive or in real-time.
312 <<installing-lttng,Install LTTng now>> and
313 <<getting-started,start tracing>>!
319 **LTTng** is a set of software <<plumbing,components>> which interact to
320 <<instrumenting,instrument>> the Linux kernel and user applications, and
321 to <<controlling-tracing,control tracing>> (start and stop
322 tracing, enable and disable event rules, and the rest). Those
323 components are bundled into the following packages:
325 * **LTTng-tools**: Libraries and command-line interface to
327 * **LTTng-modules**: Linux kernel modules to instrument and
329 * **LTTng-UST**: Libraries and Java/Python packages to instrument and
330 trace user applications.
332 Most distributions mark the LTTng-modules and LTTng-UST packages as
333 optional when installing LTTng-tools (which is always required). In the
334 following sections, we always provide the steps to install all three,
337 * You only need to install LTTng-modules if you intend to trace the
339 * You only need to install LTTng-UST if you intend to trace user
343 .Availability of LTTng{nbsp}{revision} for major Linux distributions as of 25 July 2017.
345 |Distribution |Available in releases |Alternatives
347 |https://www.ubuntu.com/[Ubuntu]
348 |<<ubuntu,Ubuntu{nbsp}17.04 _Zesty Zapus_>>.
350 Ubuntu{nbsp}14.04 _Trusty Tahr_ and Ubuntu{nbsp}16.04 _Xenial Xerus_:
351 <<ubuntu-ppa,use the LTTng Stable{nbsp}{revision} PPA>>.
352 |<<building-from-source,Build LTTng{nbsp}{revision} from source>> for
353 other Ubuntu releases.
355 |https://getfedora.org/[Fedora]
356 |<<fedora,Fedora{nbsp}26>>.
357 |link:/docs/v2.8#doc-fedora[LTTng{nbsp}2.8 for Fedora 25].
359 <<building-from-source,Build LTTng{nbsp}{revision} from source>> for
360 other Fedora releases.
362 |https://www.debian.org/[Debian]
363 |xref:debian[Debian "stretch" (stable), Debian "buster" (testing),
364 and Debian "sid" (unstable)].
365 |<<building-from-source,Build LTTng{nbsp}{revision} from source>> for
366 other Debian releases.
368 |https://www.archlinux.org/[Arch Linux]
369 |<<arch-linux,Latest AUR packages>>.
372 |https://alpinelinux.org/[Alpine Linux]
373 |<<alpine-linux,Alpine Linux "edge">>.
374 |<<building-from-source,Build LTTng{nbsp}{revision} from source>> for
375 other Alpine Linux releases.
377 |https://www.redhat.com/[RHEL] and https://www.suse.com/[SLES]
378 |See http://packages.efficios.com/[EfficiOS Enterprise Packages].
381 |https://buildroot.org/[Buildroot]
382 |<<buildroot,Buildroot{nbsp}2017.02 and Buildroot{nbsp}2017.05>>.
383 |link:/docs/v2.8#doc-buildroot[LTTng{nbsp}2.8 for Buildroot{nbsp}2016.11].
385 <<building-from-source,Build LTTng{nbsp}{revision} from source>> for
386 other Buildroot releases.
388 |http://www.openembedded.org/wiki/Main_Page[OpenEmbedded] and
389 https://www.yoctoproject.org/[Yocto]
390 |<<oe-yocto,Yocto Project{nbsp}2.3 _Pyro_>> (`openembedded-core` layer).
391 |link:/docs/v2.8#doc-oe-yocto[LTTng{nbsp}2.8 for Yocto Project{nbsp}2.2 _Morty_]
392 (`openembedded-core` layer).
394 <<building-from-source,Build LTTng{nbsp}{revision} from source>> for
395 other OpenEmbedded releases.
400 === [[ubuntu-official-repositories]]Ubuntu
402 LTTng{nbsp}{revision} is available on Ubuntu{nbsp}17.04 _Zesty Zapus_.
403 For previous releases of Ubuntu, <<ubuntu-ppa,use the LTTng
404 Stable{nbsp}{revision} PPA>>.
406 To install LTTng{nbsp}{revision} on Ubuntu{nbsp}17.04 _Zesty Zapus_:
408 . Install the main LTTng{nbsp}{revision} packages:
413 # apt-get install lttng-tools
414 # apt-get install lttng-modules-dkms
415 # apt-get install liblttng-ust-dev
419 . **If you need to instrument and trace
420 <<java-application,Java applications>>**, install the LTTng-UST
426 # apt-get install liblttng-ust-agent-java
430 . **If you need to instrument and trace
431 <<python-application,Python{nbsp}3 applications>>**, install the
432 LTTng-UST Python agent:
437 # apt-get install python3-lttngust
443 ==== noch:{LTTng} Stable {revision} PPA
445 The https://launchpad.net/~lttng/+archive/ubuntu/stable-{revision}[LTTng
446 Stable{nbsp}{revision} PPA] offers the latest stable
447 LTTng{nbsp}{revision} packages for:
449 * Ubuntu{nbsp}14.04 _Trusty Tahr_
450 * Ubuntu{nbsp}16.04 _Xenial Xerus_
452 To install LTTng{nbsp}{revision} from the LTTng Stable{nbsp}{revision} PPA:
454 . Add the LTTng Stable{nbsp}{revision} PPA repository and update the
460 # apt-add-repository ppa:lttng/stable-2.9
465 . Install the main LTTng{nbsp}{revision} packages:
470 # apt-get install lttng-tools
471 # apt-get install lttng-modules-dkms
472 # apt-get install liblttng-ust-dev
476 . **If you need to instrument and trace
477 <<java-application,Java applications>>**, install the LTTng-UST
483 # apt-get install liblttng-ust-agent-java
487 . **If you need to instrument and trace
488 <<python-application,Python{nbsp}3 applications>>**, install the
489 LTTng-UST Python agent:
494 # apt-get install python3-lttngust
502 To install LTTng{nbsp}{revision} on Fedora{nbsp}26:
504 . Install the LTTng-tools{nbsp}{revision} and LTTng-UST{nbsp}{revision}
510 # yum install lttng-tools
511 # yum install lttng-ust
515 . Download, build, and install the latest LTTng-modules{nbsp}{revision}:
521 wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.9.tar.bz2 &&
522 tar -xf lttng-modules-latest-2.9.tar.bz2 &&
523 cd lttng-modules-2.9.* &&
525 sudo make modules_install &&
531 .Java and Python application instrumentation and tracing
533 If you need to instrument and trace <<java-application,Java
534 applications>> on Fedora, you need to build and install
535 LTTng-UST{nbsp}{revision} <<building-from-source,from source>> and pass
536 the `--enable-java-agent-jul`, `--enable-java-agent-log4j`, or
537 `--enable-java-agent-all` options to the `configure` script, depending
538 on which Java logging framework you use.
540 If you need to instrument and trace <<python-application,Python
541 applications>> on Fedora, you need to build and install
542 LTTng-UST{nbsp}{revision} from source and pass the
543 `--enable-python-agent` option to the `configure` script.
550 To install LTTng{nbsp}{revision} on Debian "stretch" (stable),
551 Debian "buster" (testing), or Debian "sid" (unstable):
553 . Install the main LTTng{nbsp}{revision} packages:
558 # apt-get install lttng-modules-dkms
559 # apt-get install liblttng-ust-dev
560 # apt-get install lttng-tools
564 . **If you need to instrument and trace <<java-application,Java
565 applications>>**, install the LTTng-UST Java agent:
570 # apt-get install liblttng-ust-agent-java
574 . **If you need to instrument and trace <<python-application,Python
575 applications>>**, install the LTTng-UST Python agent:
580 # apt-get install python3-lttngust
588 To install LTTng{nbsp}{revision} on Arch Linux using
589 https://github.com/rmarquis/pacaur[pacaur]:
591 . Install the main LTTng{nbsp}{revision} packages:
596 $ pacaur -Sy lttng-tools
597 $ pacaur -Sy lttng-ust
598 $ pacaur -Sy lttng-modules
602 . **If you need to instrument and trace <<python-application,Python
603 applications>>**, install the LTTng-UST Python agent:
608 $ pacaur -Sy python-lttngust
609 $ pacaur -Sy python2-lttngust
617 To install LTTng-tools{nbsp}{revision} and LTTng-UST{nbsp}{revision} on
620 . Make sure your system is
621 https://wiki.alpinelinux.org/wiki/Edge[configured for "edge"].
622 . Enable the _testing_ repository by uncommenting the corresponding
623 line in path:{/etc/apk/repositories}.
624 . Add the LTTng packages:
629 # apk add lttng-tools
630 # apk add lttng-ust-dev
634 To install LTTng-modules{nbsp}{revision} (Linux kernel tracing support)
635 on Alpine Linux "edge":
637 . Add the vanilla Linux kernel:
642 # apk add linux-vanilla linux-vanilla-dev
646 . Reboot with the vanilla Linux kernel.
647 . Download, build, and install the latest LTTng-modules{nbsp}{revision}:
653 wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.9.tar.bz2 &&
654 tar -xf lttng-modules-latest-2.9.tar.bz2 &&
655 cd lttng-modules-2.9.* &&
657 sudo make modules_install &&
663 [[enterprise-distributions]]
664 === RHEL, SUSE, and other enterprise distributions
666 To install LTTng on enterprise Linux distributions, such as Red Hat
667 Enterprise Linux (RHEL) and SUSE Linux Enterprise Server (SUSE), please
668 see http://packages.efficios.com/[EfficiOS Enterprise Packages].
674 To install LTTng{nbsp}{revision} on Buildroot{nbsp}2017.02 or
675 Buildroot{nbsp}2017.05:
677 . Launch the Buildroot configuration tool:
686 . In **Kernel**, check **Linux kernel**.
687 . In **Toolchain**, check **Enable WCHAR support**.
688 . In **Target packages**{nbsp}→ **Debugging, profiling and benchmark**,
689 check **lttng-modules** and **lttng-tools**.
690 . In **Target packages**{nbsp}→ **Libraries**{nbsp}→
691 **Other**, check **lttng-libust**.
695 === OpenEmbedded and Yocto
697 LTTng{nbsp}{revision} recipes are available in the
698 http://layers.openembedded.org/layerindex/branch/master/layer/openembedded-core/[`openembedded-core`]
699 layer for Yocto Project{nbsp}2.3 _Pyro_ under the following names:
705 With BitBake, the simplest way to include LTTng recipes in your target
706 image is to add them to `IMAGE_INSTALL_append` in path:{conf/local.conf}:
709 IMAGE_INSTALL_append = " lttng-tools lttng-modules lttng-ust"
714 . Select a machine and an image recipe.
715 . Click **Edit image recipe**.
716 . Under the **All recipes** tab, search for **lttng**.
717 . Check the desired LTTng recipes.
720 .Java and Python application instrumentation and tracing
722 If you need to instrument and trace <<java-application,Java
723 applications>> on Yocto/OpenEmbedded, you need to build and install
724 LTTng-UST{nbsp}{revision} <<building-from-source,from source>> and pass
725 the `--enable-java-agent-jul`, `--enable-java-agent-log4j`, or
726 `--enable-java-agent-all` options to the `configure` script, depending
727 on which Java logging framework you use.
729 If you need to instrument and trace <<python-application,Python
730 applications>> on Yocto/OpenEmbedded, you need to build and install
731 LTTng-UST{nbsp}{revision} from source and pass the
732 `--enable-python-agent` option to the `configure` script.
736 [[building-from-source]]
737 === Build from source
739 To build and install LTTng{nbsp}{revision} from source:
741 . Using your distribution's package manager, or from source, install
742 the following dependencies of LTTng-tools and LTTng-UST:
745 * https://sourceforge.net/projects/libuuid/[libuuid]
746 * http://directory.fsf.org/wiki/Popt[popt]
747 * http://liburcu.org/[Userspace RCU]
748 * http://www.xmlsoft.org/[libxml2]
751 . Download, build, and install the latest LTTng-modules{nbsp}{revision}:
757 wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.9.tar.bz2 &&
758 tar -xf lttng-modules-latest-2.9.tar.bz2 &&
759 cd lttng-modules-2.9.* &&
761 sudo make modules_install &&
766 . Download, build, and install the latest LTTng-UST{nbsp}{revision}:
772 wget http://lttng.org/files/lttng-ust/lttng-ust-latest-2.9.tar.bz2 &&
773 tar -xf lttng-ust-latest-2.9.tar.bz2 &&
774 cd lttng-ust-2.9.* &&
784 .Java and Python application tracing
786 If you need to instrument and trace <<java-application,Java
787 applications>>, pass the `--enable-java-agent-jul`,
788 `--enable-java-agent-log4j`, or `--enable-java-agent-all` options to the
789 `configure` script, depending on which Java logging framework you use.
791 If you need to instrument and trace <<python-application,Python
792 applications>>, pass the `--enable-python-agent` option to the
793 `configure` script. You can set the `PYTHON` environment variable to the
794 path to the Python interpreter for which to install the LTTng-UST Python
802 By default, LTTng-UST libraries are installed to
803 dir:{/usr/local/lib}, which is the de facto directory in which to
804 keep self-compiled and third-party libraries.
806 When <<building-tracepoint-providers-and-user-application,linking an
807 instrumented user application with `liblttng-ust`>>:
809 * Append `/usr/local/lib` to the env:LD_LIBRARY_PATH environment
811 * Pass the `-L/usr/local/lib` and `-Wl,-rpath,/usr/local/lib` options to
812 man:gcc(1), man:g++(1), or man:clang(1).
816 . Download, build, and install the latest LTTng-tools{nbsp}{revision}:
822 wget http://lttng.org/files/lttng-tools/lttng-tools-latest-2.9.tar.bz2 &&
823 tar -xf lttng-tools-latest-2.9.tar.bz2 &&
824 cd lttng-tools-2.9.* &&
832 TIP: The https://github.com/eepp/vlttng[vlttng tool] can do all the
833 previous steps automatically for a given version of LTTng and confine
834 the installed files in a specific directory. This can be useful to test
835 LTTng without installing it on your system.
841 This is a short guide to get started quickly with LTTng kernel and user
844 Before you follow this guide, make sure to <<installing-lttng,install>>
847 This tutorial walks you through the steps to:
849 . <<tracing-the-linux-kernel,Trace the Linux kernel>>.
850 . <<tracing-your-own-user-application,Trace a user application>> written
852 . <<viewing-and-analyzing-your-traces,View and analyze the
856 [[tracing-the-linux-kernel]]
857 === Trace the Linux kernel
859 The following command lines start with the `#` prompt because you need
860 root privileges to trace the Linux kernel. You can also trace the kernel
861 as a regular user if your Unix user is a member of the
862 <<tracing-group,tracing group>>.
864 . Create a <<tracing-session,tracing session>> which writes its traces
865 to dir:{/tmp/my-kernel-trace}:
870 # lttng create my-kernel-session --output=/tmp/my-kernel-trace
874 . List the available kernel tracepoints and system calls:
879 # lttng list --kernel
880 # lttng list --kernel --syscall
884 . Create <<event,event rules>> which match the desired instrumentation
885 point names, for example the `sched_switch` and `sched_process_fork`
886 tracepoints, and the man:open(2) and man:close(2) system calls:
891 # lttng enable-event --kernel sched_switch,sched_process_fork
892 # lttng enable-event --kernel --syscall open,close
896 You can also create an event rule which matches _all_ the Linux kernel
897 tracepoints (this will generate a lot of data when tracing):
902 # lttng enable-event --kernel --all
906 . <<basic-tracing-session-control,Start tracing>>:
915 . Do some operation on your system for a few seconds. For example,
916 load a website, or list the files of a directory.
917 . <<basic-tracing-session-control,Stop tracing>> and destroy the
928 The man:lttng-destroy(1) command does not destroy the trace data; it
929 only destroys the state of the tracing session.
931 . For the sake of this example, make the recorded trace accessible to
937 # chown -R $(whoami) /tmp/my-kernel-trace
941 See <<viewing-and-analyzing-your-traces,View and analyze the
942 recorded events>> to view the recorded events.
945 [[tracing-your-own-user-application]]
946 === Trace a user application
948 This section steps you through a simple example to trace a
949 _Hello world_ program written in C.
951 To create the traceable user application:
953 . Create the tracepoint provider header file, which defines the
954 tracepoints and the events they can generate:
960 #undef TRACEPOINT_PROVIDER
961 #define TRACEPOINT_PROVIDER hello_world
963 #undef TRACEPOINT_INCLUDE
964 #define TRACEPOINT_INCLUDE "./hello-tp.h"
966 #if !defined(_HELLO_TP_H) || defined(TRACEPOINT_HEADER_MULTI_READ)
969 #include <lttng/tracepoint.h>
979 ctf_string(my_string_field, my_string_arg)
980 ctf_integer(int, my_integer_field, my_integer_arg)
984 #endif /* _HELLO_TP_H */
986 #include <lttng/tracepoint-event.h>
990 . Create the tracepoint provider package source file:
996 #define TRACEPOINT_CREATE_PROBES
997 #define TRACEPOINT_DEFINE
999 #include "hello-tp.h"
1003 . Build the tracepoint provider package:
1008 $ gcc -c -I. hello-tp.c
1012 . Create the _Hello World_ application source file:
1019 #include "hello-tp.h"
1021 int main(int argc, char *argv[])
1025 puts("Hello, World!\nPress Enter to continue...");
1028 * The following getchar() call is only placed here for the purpose
1029 * of this demonstration, to pause the application in order for
1030 * you to have time to list its tracepoints. It is not
1036 * A tracepoint() call.
1038 * Arguments, as defined in hello-tp.h:
1040 * 1. Tracepoint provider name (required)
1041 * 2. Tracepoint name (required)
1042 * 3. my_integer_arg (first user-defined argument)
1043 * 4. my_string_arg (second user-defined argument)
1045 * Notice the tracepoint provider and tracepoint names are
1046 * NOT strings: they are in fact parts of variables that the
1047 * macros in hello-tp.h create.
1049 tracepoint(hello_world, my_first_tracepoint, 23, "hi there!");
1051 for (x = 0; x < argc; ++x) {
1052 tracepoint(hello_world, my_first_tracepoint, x, argv[x]);
1055 puts("Quitting now!");
1056 tracepoint(hello_world, my_first_tracepoint, x * x, "x^2");
1063 . Build the application:
1072 . Link the application with the tracepoint provider package,
1073 `liblttng-ust`, and `libdl`:
1078 $ gcc -o hello hello.o hello-tp.o -llttng-ust -ldl
1082 Here's the whole build process:
1085 .User space tracing tutorial's build steps.
1086 image::ust-flow.png[]
1088 To trace the user application:
1090 . Run the application with a few arguments:
1095 $ ./hello world and beyond
1104 Press Enter to continue...
1108 . Start an LTTng <<lttng-sessiond,session daemon>>:
1113 $ lttng-sessiond --daemonize
1117 Note that a session daemon might already be running, for example as
1118 a service that the distribution's service manager started.
1120 . List the available user space tracepoints:
1125 $ lttng list --userspace
1129 You see the `hello_world:my_first_tracepoint` tracepoint listed
1130 under the `./hello` process.
1132 . Create a <<tracing-session,tracing session>>:
1137 $ lttng create my-user-space-session
1141 . Create an <<event,event rule>> which matches the
1142 `hello_world:my_first_tracepoint` event name:
1147 $ lttng enable-event --userspace hello_world:my_first_tracepoint
1151 . <<basic-tracing-session-control,Start tracing>>:
1160 . Go back to the running `hello` application and press Enter. The
1161 program executes all `tracepoint()` instrumentation points and exits.
1162 . <<basic-tracing-session-control,Stop tracing>> and destroy the
1173 The man:lttng-destroy(1) command does not destroy the trace data; it
1174 only destroys the state of the tracing session.
1176 By default, LTTng saves the traces in
1177 +$LTTNG_HOME/lttng-traces/__name__-__date__-__time__+,
1178 where +__name__+ is the tracing session name. The
1179 env:LTTNG_HOME environment variable defaults to `$HOME` if not set.
1181 See <<viewing-and-analyzing-your-traces,View and analyze the
1182 recorded events>> to view the recorded events.
1185 [[viewing-and-analyzing-your-traces]]
1186 === View and analyze the recorded events
1188 Once you have completed the <<tracing-the-linux-kernel,Trace the Linux
1189 kernel>> and <<tracing-your-own-user-application,Trace a user
1190 application>> tutorials, you can inspect the recorded events.
1192 Many tools are available to read LTTng traces:
1194 * **cmd:babeltrace** is a command-line utility which converts trace
1195 formats; it supports the format that LTTng produces, CTF, as well as a
1196 basic text output which can be ++grep++ed. The cmd:babeltrace command
1197 is part of the http://diamon.org/babeltrace[Babeltrace] project.
1198 * Babeltrace also includes
1199 **https://www.python.org/[Python] bindings** so
1200 that you can easily open and read an LTTng trace with your own script,
1201 benefiting from the power of Python.
1202 * http://tracecompass.org/[**Trace Compass**]
1203 is a graphical user interface for viewing and analyzing any type of
1204 logs or traces, including LTTng's.
1205 * https://github.com/lttng/lttng-analyses[**LTTng analyses**] is a
1206 project which includes many high-level analyses of LTTng kernel
1207 traces, like scheduling statistics, interrupt frequency distribution,
1208 top CPU usage, and more.
1210 NOTE: This section assumes that the traces recorded during the previous
1211 tutorials were saved to their default location, in the
1212 dir:{$LTTNG_HOME/lttng-traces} directory. The env:LTTNG_HOME
1213 environment variable defaults to `$HOME` if not set.
1216 [[viewing-and-analyzing-your-traces-bt]]
1217 ==== Use the cmd:babeltrace command-line tool
1219 The simplest way to list all the recorded events of a trace is to pass
1220 its path to cmd:babeltrace with no options:
1224 $ babeltrace ~/lttng-traces/my-user-space-session*
1227 cmd:babeltrace finds all traces recursively within the given path and
1228 prints all their events, merging them in chronological order.
1230 You can pipe the output of cmd:babeltrace into a tool like man:grep(1) for
1235 $ babeltrace /tmp/my-kernel-trace | grep _switch
1238 You can pipe the output of cmd:babeltrace into a tool like man:wc(1) to
1239 count the recorded events:
1243 $ babeltrace /tmp/my-kernel-trace | grep _open | wc --lines
1247 [[viewing-and-analyzing-your-traces-bt-python]]
1248 ==== Use the Babeltrace Python bindings
1250 The <<viewing-and-analyzing-your-traces-bt,text output of cmd:babeltrace>>
1251 is useful to isolate events by simple matching using man:grep(1) and
1252 similar utilities. However, more elaborate filters, such as keeping only
1253 event records with a field value falling within a specific range, are
1254 not trivial to write using a shell. Moreover, reductions and even the
1255 most basic computations involving multiple event records are virtually
1256 impossible to implement.
1258 Fortunately, Babeltrace ships with Python 3 bindings which makes it easy
1259 to read the event records of an LTTng trace sequentially and compute the
1260 desired information.
1262 The following script accepts an LTTng Linux kernel trace path as its
1263 first argument and prints the short names of the top 5 running processes
1264 on CPU 0 during the whole trace:
1269 from collections import Counter
1275 if len(sys.argv) != 2:
1276 msg = 'Usage: python3 {} TRACEPATH'.format(sys.argv[0])
1277 print(msg, file=sys.stderr)
1280 # A trace collection contains one or more traces
1281 col = babeltrace.TraceCollection()
1283 # Add the trace provided by the user (LTTng traces always have
1285 if col.add_trace(sys.argv[1], 'ctf') is None:
1286 raise RuntimeError('Cannot add trace')
1288 # This counter dict contains execution times:
1290 # task command name -> total execution time (ns)
1291 exec_times = Counter()
1293 # This contains the last `sched_switch` timestamp
1297 for event in col.events:
1298 # Keep only `sched_switch` events
1299 if event.name != 'sched_switch':
1302 # Keep only events which happened on CPU 0
1303 if event['cpu_id'] != 0:
1307 cur_ts = event.timestamp
1313 # Previous task command (short) name
1314 prev_comm = event['prev_comm']
1316 # Initialize entry in our dict if not yet done
1317 if prev_comm not in exec_times:
1318 exec_times[prev_comm] = 0
1320 # Compute previous command execution time
1321 diff = cur_ts - last_ts
1323 # Update execution time of this command
1324 exec_times[prev_comm] += diff
1326 # Update last timestamp
1330 for name, ns in exec_times.most_common(5):
1332 print('{:20}{} s'.format(name, s))
1337 if __name__ == '__main__':
1338 sys.exit(0 if top5proc() else 1)
1345 $ python3 top5proc.py /tmp/my-kernel-trace/kernel
1351 swapper/0 48.607245889 s
1352 chromium 7.192738188 s
1353 pavucontrol 0.709894415 s
1354 Compositor 0.660867933 s
1355 Xorg.bin 0.616753786 s
1358 Note that `swapper/0` is the "idle" process of CPU 0 on Linux; since we
1359 weren't using the CPU that much when tracing, its first position in the
1364 == [[understanding-lttng]]Core concepts
1366 From a user's perspective, the LTTng system is built on a few concepts,
1367 or objects, on which the <<lttng-cli,cmd:lttng command-line tool>>
1368 operates by sending commands to the <<lttng-sessiond,session daemon>>.
1369 Understanding how those objects relate to eachother is key in mastering
1372 The core concepts are:
1374 * <<tracing-session,Tracing session>>
1375 * <<domain,Tracing domain>>
1376 * <<channel,Channel and ring buffer>>
1377 * <<"event","Instrumentation point, event rule, event, and event record">>
1383 A _tracing session_ is a stateful dialogue between you and
1384 a <<lttng-sessiond,session daemon>>. You can
1385 <<creating-destroying-tracing-sessions,create a new tracing
1386 session>> with the `lttng create` command.
1388 Anything that you do when you control LTTng tracers happens within a
1389 tracing session. In particular, a tracing session:
1392 * Has its own set of trace files.
1393 * Has its own state of activity (started or stopped).
1394 * Has its own <<tracing-session-mode,mode>> (local, network streaming,
1396 * Has its own <<channel,channels>> which have their own
1397 <<event,event rules>>.
1400 .A _tracing session_ contains <<channel,channels>> that are members of <<domain,tracing domains>> and contain <<event,event rules>>.
1401 image::concepts.png[]
1403 Those attributes and objects are completely isolated between different
1406 A tracing session is analogous to a cash machine session:
1407 the operations you do on the banking system through the cash machine do
1408 not alter the data of other users of the same system. In the case of
1409 the cash machine, a session lasts as long as your bank card is inside.
1410 In the case of LTTng, a tracing session lasts from the `lttng create`
1411 command to the `lttng destroy` command.
1414 .Each Unix user has its own set of tracing sessions.
1415 image::many-sessions.png[]
1418 [[tracing-session-mode]]
1419 ==== Tracing session mode
1421 LTTng can send the generated trace data to different locations. The
1422 _tracing session mode_ dictates where to send it. The following modes
1423 are available in LTTng{nbsp}{revision}:
1426 LTTng writes the traces to the file system of the machine being traced
1429 Network streaming mode::
1430 LTTng sends the traces over the network to a
1431 <<lttng-relayd,relay daemon>> running on a remote system.
1434 LTTng does not write the traces by default. Instead, you can request
1435 LTTng to <<taking-a-snapshot,take a snapshot>>, that is, a copy of the
1436 current tracing buffers, and to write it to the target's file system
1437 or to send it over the network to a <<lttng-relayd,relay daemon>>
1438 running on a remote system.
1441 This mode is similar to the network streaming mode, but a live
1442 trace viewer can connect to the distant relay daemon to
1443 <<lttng-live,view event records as LTTng generates them>> by
1450 A _tracing domain_ is a namespace for event sources. A tracing domain
1451 has its own properties and features.
1453 There are currently five available tracing domains:
1457 * `java.util.logging` (JUL)
1461 You must specify a tracing domain when using some commands to avoid
1462 ambiguity. For example, since all the domains support named tracepoints
1463 as event sources (instrumentation points that you manually insert in the
1464 source code), you need to specify a tracing domain when
1465 <<enabling-disabling-events,creating an event rule>> because all the
1466 tracing domains could have tracepoints with the same names.
1468 Some features are reserved to specific tracing domains. Dynamic function
1469 entry and return instrumentation points, for example, are currently only
1470 supported in the Linux kernel tracing domain, but support for other
1471 tracing domains could be added in the future.
1473 You can create <<channel,channels>> in the Linux kernel and user space
1474 tracing domains. The other tracing domains have a single default
1479 === Channel and ring buffer
1481 A _channel_ is an object which is responsible for a set of ring buffers.
1482 Each ring buffer is divided into multiple sub-buffers. When an LTTng
1483 tracer emits an event, it can record it to one or more
1484 sub-buffers. The attributes of a channel determine what to do when
1485 there's no space left for a new event record because all sub-buffers
1486 are full, where to send a full sub-buffer, and other behaviours.
1488 A channel is always associated to a <<domain,tracing domain>>. The
1489 `java.util.logging` (JUL), log4j, and Python tracing domains each have
1490 a default channel which you cannot configure.
1492 A channel also owns <<event,event rules>>. When an LTTng tracer emits
1493 an event, it records it to the sub-buffers of all
1494 the enabled channels with a satisfied event rule, as long as those
1495 channels are part of active <<tracing-session,tracing sessions>>.
1498 [[channel-buffering-schemes]]
1499 ==== Per-user vs. per-process buffering schemes
1501 A channel has at least one ring buffer _per CPU_. LTTng always
1502 records an event to the ring buffer associated to the CPU on which it
1505 Two _buffering schemes_ are available when you
1506 <<enabling-disabling-channels,create a channel>> in the
1507 user space <<domain,tracing domain>>:
1509 Per-user buffering::
1510 Allocate one set of ring buffers--one per CPU--shared by all the
1511 instrumented processes of each Unix user.
1515 .Per-user buffering scheme.
1516 image::per-user-buffering.png[]
1519 Per-process buffering::
1520 Allocate one set of ring buffers--one per CPU--for each
1521 instrumented process.
1525 .Per-process buffering scheme.
1526 image::per-process-buffering.png[]
1529 The per-process buffering scheme tends to consume more memory than the
1530 per-user option because systems generally have more instrumented
1531 processes than Unix users running instrumented processes. However, the
1532 per-process buffering scheme ensures that one process having a high
1533 event throughput won't fill all the shared sub-buffers of the same
1536 The Linux kernel tracing domain has only one available buffering scheme
1537 which is to allocate a single set of ring buffers for the whole system.
1538 This scheme is similar to the per-user option, but with a single, global
1539 user "running" the kernel.
1542 [[channel-overwrite-mode-vs-discard-mode]]
1543 ==== Overwrite vs. discard event loss modes
1545 When an event occurs, LTTng records it to a specific sub-buffer (yellow
1546 arc in the following animation) of a specific channel's ring buffer.
1547 When there's no space left in a sub-buffer, the tracer marks it as
1548 consumable (red) and another, empty sub-buffer starts receiving the
1549 following event records. A <<lttng-consumerd,consumer daemon>>
1550 eventually consumes the marked sub-buffer (returns to white).
1553 [role="docsvg-channel-subbuf-anim"]
1558 In an ideal world, sub-buffers are consumed faster than they are filled,
1559 as is the case in the previous animation. In the real world,
1560 however, all sub-buffers can be full at some point, leaving no space to
1561 record the following events.
1563 By design, LTTng is a _non-blocking_ tracer: when no empty sub-buffer is
1564 available, it is acceptable to lose event records when the alternative
1565 would be to cause substantial delays in the instrumented application's
1566 execution. LTTng privileges performance over integrity; it aims at
1567 perturbing the traced system as little as possible in order to make
1568 tracing of subtle race conditions and rare interrupt cascades possible.
1570 When it comes to losing event records because no empty sub-buffer is
1571 available, the channel's _event loss mode_ determines what to do. The
1572 available event loss modes are:
1575 Drop the newest event records until a the tracer
1576 releases a sub-buffer.
1579 Clear the sub-buffer containing the oldest event records and start
1580 writing the newest event records there.
1582 This mode is sometimes called _flight recorder mode_ because it's
1584 https://en.wikipedia.org/wiki/Flight_recorder[flight recorder]:
1585 always keep a fixed amount of the latest data.
1587 Which mechanism you should choose depends on your context: prioritize
1588 the newest or the oldest event records in the ring buffer?
1590 Beware that, in overwrite mode, the tracer abandons a whole sub-buffer
1591 as soon as a there's no space left for a new event record, whereas in
1592 discard mode, the tracer only discards the event record that doesn't
1595 In discard mode, LTTng increments a count of lost event records when an
1596 event record is lost and saves this count to the trace. In overwrite
1597 mode, since LTTng 2.8, LTTng increments a count of lost sub-buffers when
1598 a sub-buffer is lost and saves this count to the trace. In this mode,
1599 the exact number of lost event records in those lost sub-buffers is not
1600 saved to the trace. Trace analyses can use the trace's saved discarded
1601 event record and sub-buffer counts to decide whether or not to perform
1602 the analyses even if trace data is known to be missing.
1604 There are a few ways to decrease your probability of losing event
1606 <<channel-subbuf-size-vs-subbuf-count,Sub-buffer count and size>> shows
1607 how you can fine-une the sub-buffer count and size of a channel to
1608 virtually stop losing event records, though at the cost of greater
1612 [[channel-subbuf-size-vs-subbuf-count]]
1613 ==== Sub-buffer count and size
1615 When you <<enabling-disabling-channels,create a channel>>, you can
1616 set its number of sub-buffers and their size.
1618 Note that there is noticeable CPU overhead introduced when
1619 switching sub-buffers (marking a full one as consumable and switching
1620 to an empty one for the following events to be recorded). Knowing this,
1621 the following list presents a few practical situations along with how
1622 to configure the sub-buffer count and size for them:
1624 * **High event throughput**: In general, prefer bigger sub-buffers to
1625 lower the risk of losing event records.
1627 Having bigger sub-buffers also ensures a lower
1628 <<channel-switch-timer,sub-buffer switching frequency>>.
1630 The number of sub-buffers is only meaningful if you create the channel
1631 in overwrite mode: in this case, if a sub-buffer overwrite happens, the
1632 other sub-buffers are left unaltered.
1634 * **Low event throughput**: In general, prefer smaller sub-buffers
1635 since the risk of losing event records is low.
1637 Because events occur less frequently, the sub-buffer switching frequency
1638 should remain low and thus the tracer's overhead should not be a
1641 * **Low memory system**: If your target system has a low memory
1642 limit, prefer fewer first, then smaller sub-buffers.
1644 Even if the system is limited in memory, you want to keep the
1645 sub-buffers as big as possible to avoid a high sub-buffer switching
1648 Note that LTTng uses http://diamon.org/ctf/[CTF] as its trace format,
1649 which means event data is very compact. For example, the average
1650 LTTng kernel event record weights about 32{nbsp}bytes. Thus, a
1651 sub-buffer size of 1{nbsp}MiB is considered big.
1653 The previous situations highlight the major trade-off between a few big
1654 sub-buffers and more, smaller sub-buffers: sub-buffer switching
1655 frequency vs. how much data is lost in overwrite mode. Assuming a
1656 constant event throughput and using the overwrite mode, the two
1657 following configurations have the same ring buffer total size:
1660 [role="docsvg-channel-subbuf-size-vs-count-anim"]
1665 * **2 sub-buffers of 4{nbsp}MiB each**: Expect a very low sub-buffer
1666 switching frequency, but if a sub-buffer overwrite happens, half of
1667 the event records so far (4{nbsp}MiB) are definitely lost.
1668 * **8 sub-buffers of 1{nbsp}MiB each**: Expect 4{nbsp}times the tracer's
1669 overhead as the previous configuration, but if a sub-buffer
1670 overwrite happens, only the eighth of event records so far are
1673 In discard mode, the sub-buffers count parameter is pointless: use two
1674 sub-buffers and set their size according to the requirements of your
1678 [[channel-switch-timer]]
1679 ==== Switch timer period
1681 The _switch timer period_ is an important configurable attribute of
1682 a channel to ensure periodic sub-buffer flushing.
1684 When the _switch timer_ expires, a sub-buffer switch happens. You can
1685 set the switch timer period attribute when you
1686 <<enabling-disabling-channels,create a channel>> to ensure that event
1687 data is consumed and committed to trace files or to a distant relay
1688 daemon periodically in case of a low event throughput.
1691 [role="docsvg-channel-switch-timer"]
1696 This attribute is also convenient when you use big sub-buffers to cope
1697 with a sporadic high event throughput, even if the throughput is
1701 [[channel-read-timer]]
1702 ==== Read timer period
1704 By default, the LTTng tracers use a notification mechanism to signal a
1705 full sub-buffer so that a consumer daemon can consume it. When such
1706 notifications must be avoided, for example in real-time applications,
1707 you can use the channel's _read timer_ instead. When the read timer
1708 fires, the <<lttng-consumerd,consumer daemon>> checks for full,
1709 consumable sub-buffers.
1712 [[tracefile-rotation]]
1713 ==== Trace file count and size
1715 By default, trace files can grow as large as needed. You can set the
1716 maximum size of each trace file that a channel writes when you
1717 <<enabling-disabling-channels,create a channel>>. When the size of
1718 a trace file reaches the channel's fixed maximum size, LTTng creates
1719 another file to contain the next event records. LTTng appends a file
1720 count to each trace file name in this case.
1722 If you set the trace file size attribute when you create a channel, the
1723 maximum number of trace files that LTTng creates is _unlimited_ by
1724 default. To limit them, you can also set a maximum number of trace
1725 files. When the number of trace files reaches the channel's fixed
1726 maximum count, the oldest trace file is overwritten. This mechanism is
1727 called _trace file rotation_.
1731 === Instrumentation point, event rule, event, and event record
1733 An _event rule_ is a set of conditions which must be **all** satisfied
1734 for LTTng to record an occuring event.
1736 You set the conditions when you <<enabling-disabling-events,create
1739 You always attach an event rule to <<channel,channel>> when you create
1742 When an event passes the conditions of an event rule, LTTng records it
1743 in one of the attached channel's sub-buffers.
1745 The available conditions, as of LTTng{nbsp}{revision}, are:
1747 * The event rule _is enabled_.
1748 * The instrumentation point's type _is{nbsp}T_.
1749 * The instrumentation point's name (sometimes called _event name_)
1750 _matches{nbsp}N_, but _is not{nbsp}E_.
1751 * The instrumentation point's log level _is as severe as{nbsp}L_, or
1752 _is exactly{nbsp}L_.
1753 * The fields of the event's payload _satisfy_ a filter
1754 expression{nbsp}__F__.
1756 As you can see, all the conditions but the dynamic filter are related to
1757 the event rule's status or to the instrumentation point, not to the
1758 occurring events. This is why, without a filter, checking if an event
1759 passes an event rule is not a dynamic task: when you create or modify an
1760 event rule, all the tracers of its tracing domain enable or disable the
1761 instrumentation points themselves once. This is possible because the
1762 attributes of an instrumentation point (type, name, and log level) are
1763 defined statically. In other words, without a dynamic filter, the tracer
1764 _does not evaluate_ the arguments of an instrumentation point unless it
1765 matches an enabled event rule.
1767 Note that, for LTTng to record an event, the <<channel,channel>> to
1768 which a matching event rule is attached must also be enabled, and the
1769 tracing session owning this channel must be active.
1772 .Logical path from an instrumentation point to an event record.
1773 image::event-rule.png[]
1775 .Event, event record, or event rule?
1777 With so many similar terms, it's easy to get confused.
1779 An **event** is the consequence of the execution of an _instrumentation
1780 point_, like a tracepoint that you manually place in some source code,
1781 or a Linux kernel KProbe. An event is said to _occur_ at a specific
1782 time. Different actions can be taken upon the occurrence of an event,
1783 like record the event's payload to a buffer.
1785 An **event record** is the representation of an event in a sub-buffer. A
1786 tracer is responsible for capturing the payload of an event, current
1787 context variables, the event's ID, and the event's timestamp. LTTng
1788 can append this sub-buffer to a trace file.
1790 An **event rule** is a set of conditions which must all be satisfied for
1791 LTTng to record an occuring event. Events still occur without
1792 satisfying event rules, but LTTng does not record them.
1797 == Components of noch:{LTTng}
1799 The second _T_ in _LTTng_ stands for _toolkit_: it would be wrong
1800 to call LTTng a simple _tool_ since it is composed of multiple
1801 interacting components. This section describes those components,
1802 explains their respective roles, and shows how they connect together to
1803 form the LTTng ecosystem.
1805 The following diagram shows how the most important components of LTTng
1806 interact with user applications, the Linux kernel, and you:
1809 .Control and trace data paths between LTTng components.
1810 image::plumbing.png[]
1812 The LTTng project incorporates:
1814 * **LTTng-tools**: Libraries and command-line interface to
1815 control tracing sessions.
1816 ** <<lttng-sessiond,Session daemon>> (man:lttng-sessiond(8)).
1817 ** <<lttng-consumerd,Consumer daemon>> (man:lttng-consumerd(8)).
1818 ** <<lttng-relayd,Relay daemon>> (man:lttng-relayd(8)).
1819 ** <<liblttng-ctl-lttng,Tracing control library>> (`liblttng-ctl`).
1820 ** <<lttng-cli,Tracing control command-line tool>> (man:lttng(1)).
1821 * **LTTng-UST**: Libraries and Java/Python packages to trace user
1823 ** <<lttng-ust,User space tracing library>> (`liblttng-ust`) and its
1824 headers to instrument and trace any native user application.
1825 ** <<prebuilt-ust-helpers,Preloadable user space tracing helpers>>:
1826 *** `liblttng-ust-libc-wrapper`
1827 *** `liblttng-ust-pthread-wrapper`
1828 *** `liblttng-ust-cyg-profile`
1829 *** `liblttng-ust-cyg-profile-fast`
1830 *** `liblttng-ust-dl`
1831 ** User space tracepoint provider source files generator command-line
1832 tool (man:lttng-gen-tp(1)).
1833 ** <<lttng-ust-agents,LTTng-UST Java agent>> to instrument and trace
1834 Java applications using `java.util.logging` or
1835 Apache log4j 1.2 logging.
1836 ** <<lttng-ust-agents,LTTng-UST Python agent>> to instrument
1837 Python applications using the standard `logging` package.
1838 * **LTTng-modules**: <<lttng-modules,Linux kernel modules>> to trace
1840 ** LTTng kernel tracer module.
1841 ** Tracing ring buffer kernel modules.
1842 ** Probe kernel modules.
1843 ** LTTng logger kernel module.
1847 === Tracing control command-line interface
1850 .The tracing control command-line interface.
1851 image::plumbing-lttng-cli.png[]
1853 The _man:lttng(1) command-line tool_ is the standard user interface to
1854 control LTTng <<tracing-session,tracing sessions>>. The cmd:lttng tool
1855 is part of LTTng-tools.
1857 The cmd:lttng tool is linked with
1858 <<liblttng-ctl-lttng,`liblttng-ctl`>> to communicate with
1859 one or more <<lttng-sessiond,session daemons>> behind the scenes.
1861 The cmd:lttng tool has a Git-like interface:
1865 $ lttng <GENERAL OPTIONS> <COMMAND> <COMMAND OPTIONS>
1868 The <<controlling-tracing,Tracing control>> section explores the
1869 available features of LTTng using the cmd:lttng tool.
1872 [[liblttng-ctl-lttng]]
1873 === Tracing control library
1876 .The tracing control library.
1877 image::plumbing-liblttng-ctl.png[]
1879 The _LTTng control library_, `liblttng-ctl`, is used to communicate
1880 with a <<lttng-sessiond,session daemon>> using a C API that hides the
1881 underlying protocol's details. `liblttng-ctl` is part of LTTng-tools.
1883 The <<lttng-cli,cmd:lttng command-line tool>>
1884 is linked with `liblttng-ctl`.
1886 You can use `liblttng-ctl` in C or $$C++$$ source code by including its
1891 #include <lttng/lttng.h>
1894 Some objects are referenced by name (C string), such as tracing
1895 sessions, but most of them require to create a handle first using
1896 `lttng_create_handle()`.
1898 The best available developer documentation for `liblttng-ctl` is, as of
1899 LTTng{nbsp}{revision}, its installed header files. Every function and
1900 structure is thoroughly documented.
1904 === User space tracing library
1907 .The user space tracing library.
1908 image::plumbing-liblttng-ust.png[]
1910 The _user space tracing library_, `liblttng-ust` (see man:lttng-ust(3)),
1911 is the LTTng user space tracer. It receives commands from a
1912 <<lttng-sessiond,session daemon>>, for example to
1913 enable and disable specific instrumentation points, and writes event
1914 records to ring buffers shared with a
1915 <<lttng-consumerd,consumer daemon>>.
1916 `liblttng-ust` is part of LTTng-UST.
1918 Public C header files are installed beside `liblttng-ust` to
1919 instrument any <<c-application,C or $$C++$$ application>>.
1921 <<lttng-ust-agents,LTTng-UST agents>>, which are regular Java and Python
1922 packages, use their own library providing tracepoints which is
1923 linked with `liblttng-ust`.
1925 An application or library does not have to initialize `liblttng-ust`
1926 manually: its constructor does the necessary tasks to properly register
1927 to a session daemon. The initialization phase also enables the
1928 instrumentation points matching the <<event,event rules>> that you
1932 [[lttng-ust-agents]]
1933 === User space tracing agents
1936 .The user space tracing agents.
1937 image::plumbing-lttng-ust-agents.png[]
1939 The _LTTng-UST Java and Python agents_ are regular Java and Python
1940 packages which add LTTng tracing capabilities to the
1941 native logging frameworks. The LTTng-UST agents are part of LTTng-UST.
1943 In the case of Java, the
1944 https://docs.oracle.com/javase/7/docs/api/java/util/logging/package-summary.html[`java.util.logging`
1945 core logging facilities] and
1946 https://logging.apache.org/log4j/1.2/[Apache log4j 1.2] are supported.
1947 Note that Apache Log4{nbsp}2 is not supported.
1949 In the case of Python, the standard
1950 https://docs.python.org/3/library/logging.html[`logging`] package
1951 is supported. Both Python 2 and Python 3 modules can import the
1952 LTTng-UST Python agent package.
1954 The applications using the LTTng-UST agents are in the
1955 `java.util.logging` (JUL),
1956 log4j, and Python <<domain,tracing domains>>.
1958 Both agents use the same mechanism to trace the log statements. When an
1959 agent is initialized, it creates a log handler that attaches to the root
1960 logger. The agent also registers to a <<lttng-sessiond,session daemon>>.
1961 When the application executes a log statement, it is passed to the
1962 agent's log handler by the root logger. The agent's log handler calls a
1963 native function in a tracepoint provider package shared library linked
1964 with <<lttng-ust,`liblttng-ust`>>, passing the formatted log message and
1965 other fields, like its logger name and its log level. This native
1966 function contains a user space instrumentation point, hence tracing the
1969 The log level condition of an
1970 <<event,event rule>> is considered when tracing
1971 a Java or a Python application, and it's compatible with the standard
1972 JUL, log4j, and Python log levels.
1976 === LTTng kernel modules
1979 .The LTTng kernel modules.
1980 image::plumbing-lttng-modules.png[]
1982 The _LTTng kernel modules_ are a set of Linux kernel modules
1983 which implement the kernel tracer of the LTTng project. The LTTng
1984 kernel modules are part of LTTng-modules.
1986 The LTTng kernel modules include:
1988 * A set of _probe_ modules.
1990 Each module attaches to a specific subsystem
1991 of the Linux kernel using its tracepoint instrument points. There are
1992 also modules to attach to the entry and return points of the Linux
1993 system call functions.
1995 * _Ring buffer_ modules.
1997 A ring buffer implementation is provided as kernel modules. The LTTng
1998 kernel tracer writes to the ring buffer; a
1999 <<lttng-consumerd,consumer daemon>> reads from the ring buffer.
2001 * The _LTTng kernel tracer_ module.
2002 * The _LTTng logger_ module.
2004 The LTTng logger module implements the special path:{/proc/lttng-logger}
2005 file so that any executable can generate LTTng events by opening and
2006 writing to this file.
2008 See <<proc-lttng-logger-abi,LTTng logger>>.
2010 Generally, you do not have to load the LTTng kernel modules manually
2011 (using man:modprobe(8), for example): a root <<lttng-sessiond,session
2012 daemon>> loads the necessary modules when starting. If you have extra
2013 probe modules, you can specify to load them to the session daemon on
2016 The LTTng kernel modules are installed in
2017 +/usr/lib/modules/__release__/extra+ by default, where +__release__+ is
2018 the kernel release (see `uname --kernel-release`).
2025 .The session daemon.
2026 image::plumbing-sessiond.png[]
2028 The _session daemon_, man:lttng-sessiond(8), is a daemon responsible for
2029 managing tracing sessions and for controlling the various components of
2030 LTTng. The session daemon is part of LTTng-tools.
2032 The session daemon sends control requests to and receives control
2035 * The <<lttng-ust,user space tracing library>>.
2037 Any instance of the user space tracing library first registers to
2038 a session daemon. Then, the session daemon can send requests to
2039 this instance, such as:
2042 ** Get the list of tracepoints.
2043 ** Share an <<event,event rule>> so that the user space tracing library
2044 can enable or disable tracepoints. Amongst the possible conditions
2045 of an event rule is a filter expression which `liblttng-ust` evalutes
2046 when an event occurs.
2047 ** Share <<channel,channel>> attributes and ring buffer locations.
2050 The session daemon and the user space tracing library use a Unix
2051 domain socket for their communication.
2053 * The <<lttng-ust-agents,user space tracing agents>>.
2055 Any instance of a user space tracing agent first registers to
2056 a session daemon. Then, the session daemon can send requests to
2057 this instance, such as:
2060 ** Get the list of loggers.
2061 ** Enable or disable a specific logger.
2064 The session daemon and the user space tracing agent use a TCP connection
2065 for their communication.
2067 * The <<lttng-modules,LTTng kernel tracer>>.
2068 * The <<lttng-consumerd,consumer daemon>>.
2070 The session daemon sends requests to the consumer daemon to instruct
2071 it where to send the trace data streams, amongst other information.
2073 * The <<lttng-relayd,relay daemon>>.
2075 The session daemon receives commands from the
2076 <<liblttng-ctl-lttng,tracing control library>>.
2078 The root session daemon loads the appropriate
2079 <<lttng-modules,LTTng kernel modules>> on startup. It also spawns
2080 a <<lttng-consumerd,consumer daemon>> as soon as you create
2081 an <<event,event rule>>.
2083 The session daemon does not send and receive trace data: this is the
2084 role of the <<lttng-consumerd,consumer daemon>> and
2085 <<lttng-relayd,relay daemon>>. It does, however, generate the
2086 http://diamon.org/ctf/[CTF] metadata stream.
2088 Each Unix user can have its own session daemon instance. The
2089 tracing sessions managed by different session daemons are completely
2092 The root user's session daemon is the only one which is
2093 allowed to control the LTTng kernel tracer, and its spawned consumer
2094 daemon is the only one which is allowed to consume trace data from the
2095 LTTng kernel tracer. Note, however, that any Unix user which is a member
2096 of the <<tracing-group,tracing group>> is allowed
2097 to create <<channel,channels>> in the
2098 Linux kernel <<domain,tracing domain>>, and thus to trace the Linux
2101 The <<lttng-cli,cmd:lttng command-line tool>> automatically starts a
2102 session daemon when using its `create` command if none is currently
2103 running. You can also start the session daemon manually.
2110 .The consumer daemon.
2111 image::plumbing-consumerd.png[]
2113 The _consumer daemon_, man:lttng-consumerd(8), is a daemon which shares
2114 ring buffers with user applications or with the LTTng kernel modules to
2115 collect trace data and send it to some location (on disk or to a
2116 <<lttng-relayd,relay daemon>> over the network). The consumer daemon
2117 is part of LTTng-tools.
2119 You do not start a consumer daemon manually: a consumer daemon is always
2120 spawned by a <<lttng-sessiond,session daemon>> as soon as you create an
2121 <<event,event rule>>, that is, before you start tracing. When you kill
2122 its owner session daemon, the consumer daemon also exits because it is
2123 the session daemon's child process. Command-line options of
2124 man:lttng-sessiond(8) target the consumer daemon process.
2126 There are up to two running consumer daemons per Unix user, whereas only
2127 one session daemon can run per user. This is because each process can be
2128 either 32-bit or 64-bit: if the target system runs a mixture of 32-bit
2129 and 64-bit processes, it is more efficient to have separate
2130 corresponding 32-bit and 64-bit consumer daemons. The root user is an
2131 exception: it can have up to _three_ running consumer daemons: 32-bit
2132 and 64-bit instances for its user applications, and one more
2133 reserved for collecting kernel trace data.
2141 image::plumbing-relayd.png[]
2143 The _relay daemon_, man:lttng-relayd(8), is a daemon acting as a bridge
2144 between remote session and consumer daemons, local trace files, and a
2145 remote live trace viewer. The relay daemon is part of LTTng-tools.
2147 The main purpose of the relay daemon is to implement a receiver of
2148 <<sending-trace-data-over-the-network,trace data over the network>>.
2149 This is useful when the target system does not have much file system
2150 space to record trace files locally.
2152 The relay daemon is also a server to which a
2153 <<lttng-live,live trace viewer>> can
2154 connect. The live trace viewer sends requests to the relay daemon to
2155 receive trace data as the target system emits events. The
2156 communication protocol is named _LTTng live_; it is used over TCP
2159 Note that you can start the relay daemon on the target system directly.
2160 This is the setup of choice when the use case is to view events as
2161 the target system emits them without the need of a remote system.
2165 == [[using-lttng]]Instrumentation
2167 There are many examples of tracing and monitoring in our everyday life:
2169 * You have access to real-time and historical weather reports and
2170 forecasts thanks to weather stations installed around the country.
2171 * You know your heart is safe thanks to an electrocardiogram.
2172 * You make sure not to drive your car too fast and to have enough fuel
2173 to reach your destination thanks to gauges visible on your dashboard.
2175 All the previous examples have something in common: they rely on
2176 **instruments**. Without the electrodes attached to the surface of your
2177 body's skin, cardiac monitoring is futile.
2179 LTTng, as a tracer, is no different from those real life examples. If
2180 you're about to trace a software system or, in other words, record its
2181 history of execution, you better have **instrumentation points** in the
2182 subject you're tracing, that is, the actual software.
2184 Various ways were developed to instrument a piece of software for LTTng
2185 tracing. The most straightforward one is to manually place
2186 instrumentation points, called _tracepoints_, in the software's source
2187 code. It is also possible to add instrumentation points dynamically in
2188 the Linux kernel <<domain,tracing domain>>.
2190 If you're only interested in tracing the Linux kernel, your
2191 instrumentation needs are probably already covered by LTTng's built-in
2192 <<lttng-modules,Linux kernel tracepoints>>. You may also wish to trace a
2193 user application which is already instrumented for LTTng tracing.
2194 In such cases, you can skip this whole section and read the topics of
2195 the <<controlling-tracing,Tracing control>> section.
2197 Many methods are available to instrument a piece of software for LTTng
2200 * <<c-application,User space instrumentation for C and $$C++$$
2202 * <<prebuilt-ust-helpers,Prebuilt user space tracing helpers>>.
2203 * <<java-application,User space Java agent>>.
2204 * <<python-application,User space Python agent>>.
2205 * <<proc-lttng-logger-abi,LTTng logger>>.
2206 * <<instrumenting-linux-kernel,LTTng kernel tracepoints>>.
2210 === [[cxx-application]]User space instrumentation for C and $$C++$$ applications
2212 The procedure to instrument a C or $$C++$$ user application with
2213 the <<lttng-ust,LTTng user space tracing library>>, `liblttng-ust`, is:
2215 . <<tracepoint-provider,Create the source files of a tracepoint provider
2217 . <<probing-the-application-source-code,Add tracepoints to
2218 the application's source code>>.
2219 . <<building-tracepoint-providers-and-user-application,Build and link
2220 a tracepoint provider package and the user application>>.
2222 If you need quick, man:printf(3)-like instrumentation, you can skip
2223 those steps and use <<tracef,`tracef()`>> or <<tracelog,`tracelog()`>>
2226 IMPORTANT: You need to <<installing-lttng,install>> LTTng-UST to
2227 instrument a user application with `liblttng-ust`.
2230 [[tracepoint-provider]]
2231 ==== Create the source files of a tracepoint provider package
2233 A _tracepoint provider_ is a set of compiled functions which provide
2234 **tracepoints** to an application, the type of instrumentation point
2235 supported by LTTng-UST. Those functions can emit events with
2236 user-defined fields and serialize those events as event records to one
2237 or more LTTng-UST <<channel,channel>> sub-buffers. The `tracepoint()`
2238 macro, which you <<probing-the-application-source-code,insert in a user
2239 application's source code>>, calls those functions.
2241 A _tracepoint provider package_ is an object file (`.o`) or a shared
2242 library (`.so`) which contains one or more tracepoint providers.
2243 Its source files are:
2245 * One or more <<tpp-header,tracepoint provider header>> (`.h`).
2246 * A <<tpp-source,tracepoint provider package source>> (`.c`).
2248 A tracepoint provider package is dynamically linked with `liblttng-ust`,
2249 the LTTng user space tracer, at run time.
2252 .User application linked with `liblttng-ust` and containing a tracepoint provider.
2253 image::ust-app.png[]
2255 NOTE: If you need quick, man:printf(3)-like instrumentation, you can
2256 skip creating and using a tracepoint provider and use
2257 <<tracef,`tracef()`>> or <<tracelog,`tracelog()`>> instead.
2261 ===== Create a tracepoint provider header file template
2263 A _tracepoint provider header file_ contains the tracepoint
2264 definitions of a tracepoint provider.
2266 To create a tracepoint provider header file:
2268 . Start from this template:
2272 .Tracepoint provider header file template (`.h` file extension).
2274 #undef TRACEPOINT_PROVIDER
2275 #define TRACEPOINT_PROVIDER provider_name
2277 #undef TRACEPOINT_INCLUDE
2278 #define TRACEPOINT_INCLUDE "./tp.h"
2280 #if !defined(_TP_H) || defined(TRACEPOINT_HEADER_MULTI_READ)
2283 #include <lttng/tracepoint.h>
2286 * Use TRACEPOINT_EVENT(), TRACEPOINT_EVENT_CLASS(),
2287 * TRACEPOINT_EVENT_INSTANCE(), and TRACEPOINT_LOGLEVEL() here.
2292 #include <lttng/tracepoint-event.h>
2298 * `provider_name` with the name of your tracepoint provider.
2299 * `"tp.h"` with the name of your tracepoint provider header file.
2301 . Below the `#include <lttng/tracepoint.h>` line, put your
2302 <<defining-tracepoints,tracepoint definitions>>.
2304 Your tracepoint provider name must be unique amongst all the possible
2305 tracepoint provider names used on the same target system. We
2306 suggest to include the name of your project or company in the name,
2307 for example, `org_lttng_my_project_tpp`.
2309 TIP: [[lttng-gen-tp]]You can use the man:lttng-gen-tp(1) tool to create
2310 this boilerplate for you. When using cmd:lttng-gen-tp, all you need to
2311 write are the <<defining-tracepoints,tracepoint definitions>>.
2314 [[defining-tracepoints]]
2315 ===== Create a tracepoint definition
2317 A _tracepoint definition_ defines, for a given tracepoint:
2319 * Its **input arguments**. They are the macro parameters that the
2320 `tracepoint()` macro accepts for this particular tracepoint
2321 in the user application's source code.
2322 * Its **output event fields**. They are the sources of event fields
2323 that form the payload of any event that the execution of the
2324 `tracepoint()` macro emits for this particular tracepoint.
2326 You can create a tracepoint definition by using the
2327 `TRACEPOINT_EVENT()` macro below the `#include <lttng/tracepoint.h>`
2329 <<tpp-header,tracepoint provider header file template>>.
2331 The syntax of the `TRACEPOINT_EVENT()` macro is:
2334 .`TRACEPOINT_EVENT()` macro syntax.
2337 /* Tracepoint provider name */
2340 /* Tracepoint name */
2343 /* Input arguments */
2348 /* Output event fields */
2357 * `provider_name` with your tracepoint provider name.
2358 * `tracepoint_name` with your tracepoint name.
2359 * `arguments` with the <<tpp-def-input-args,input arguments>>.
2360 * `fields` with the <<tpp-def-output-fields,output event field>>
2363 This tracepoint emits events named `provider_name:tracepoint_name`.
2366 .Event name's length limitation
2368 The concatenation of the tracepoint provider name and the
2369 tracepoint name must not exceed **254 characters**. If it does, the
2370 instrumented application compiles and runs, but LTTng throws multiple
2371 warnings and you could experience serious issues.
2374 [[tpp-def-input-args]]The syntax of the `TP_ARGS()` macro is:
2377 .`TP_ARGS()` macro syntax.
2386 * `type` with the C type of the argument.
2387 * `arg_name` with the argument name.
2389 You can repeat `type` and `arg_name` up to 10 times to have
2390 more than one argument.
2392 .`TP_ARGS()` usage with three arguments.
2404 The `TP_ARGS()` and `TP_ARGS(void)` forms are valid to create a
2405 tracepoint definition with no input arguments.
2407 [[tpp-def-output-fields]]The `TP_FIELDS()` macro contains a list of
2408 `ctf_*()` macros. Each `ctf_*()` macro defines one event field. See
2409 man:lttng-ust(3) for a complete description of the available `ctf_*()`
2410 macros. A `ctf_*()` macro specifies the type, size, and byte order of
2413 Each `ctf_*()` macro takes an _argument expression_ parameter. This is a
2414 C expression that the tracer evalutes at the `tracepoint()` macro site
2415 in the application's source code. This expression provides a field's
2416 source of data. The argument expression can include input argument names
2417 listed in the `TP_ARGS()` macro.
2419 Each `ctf_*()` macro also takes a _field name_ parameter. Field names
2420 must be unique within a given tracepoint definition.
2422 Here's a complete tracepoint definition example:
2424 .Tracepoint definition.
2426 The following tracepoint definition defines a tracepoint which takes
2427 three input arguments and has four output event fields.
2431 #include "my-custom-structure.h"
2437 const struct my_custom_structure*, my_custom_structure,
2442 ctf_string(query_field, query)
2443 ctf_float(double, ratio_field, ratio)
2444 ctf_integer(int, recv_size, my_custom_structure->recv_size)
2445 ctf_integer(int, send_size, my_custom_structure->send_size)
2450 You can refer to this tracepoint definition with the `tracepoint()`
2451 macro in your application's source code like this:
2455 tracepoint(my_provider, my_tracepoint,
2456 my_structure, some_ratio, the_query);
2460 NOTE: The LTTng tracer only evaluates tracepoint arguments at run time
2461 if they satisfy an enabled <<event,event rule>>.
2464 [[using-tracepoint-classes]]
2465 ===== Use a tracepoint class
2467 A _tracepoint class_ is a class of tracepoints which share the same
2468 output event field definitions. A _tracepoint instance_ is one
2469 instance of such a defined tracepoint class, with its own tracepoint
2472 The <<defining-tracepoints,`TRACEPOINT_EVENT()` macro>> is actually a
2473 shorthand which defines both a tracepoint class and a tracepoint
2474 instance at the same time.
2476 When you build a tracepoint provider package, the C or $$C++$$ compiler
2477 creates one serialization function for each **tracepoint class**. A
2478 serialization function is responsible for serializing the event fields
2479 of a tracepoint to a sub-buffer when tracing.
2481 For various performance reasons, when your situation requires multiple
2482 tracepoint definitions with different names, but with the same event
2483 fields, we recommend that you manually create a tracepoint class
2484 and instantiate as many tracepoint instances as needed. One positive
2485 effect of such a design, amongst other advantages, is that all
2486 tracepoint instances of the same tracepoint class reuse the same
2487 serialization function, thus reducing
2488 https://en.wikipedia.org/wiki/Cache_pollution[cache pollution].
2490 .Use a tracepoint class and tracepoint instances.
2492 Consider the following three tracepoint definitions:
2504 ctf_integer(int, userid, userid)
2505 ctf_integer(size_t, len, len)
2517 ctf_integer(int, userid, userid)
2518 ctf_integer(size_t, len, len)
2530 ctf_integer(int, userid, userid)
2531 ctf_integer(size_t, len, len)
2536 In this case, we create three tracepoint classes, with one implicit
2537 tracepoint instance for each of them: `get_account`, `get_settings`, and
2538 `get_transaction`. However, they all share the same event field names
2539 and types. Hence three identical, yet independent serialization
2540 functions are created when you build the tracepoint provider package.
2542 A better design choice is to define a single tracepoint class and three
2543 tracepoint instances:
2547 /* The tracepoint class */
2548 TRACEPOINT_EVENT_CLASS(
2549 /* Tracepoint provider name */
2552 /* Tracepoint class name */
2555 /* Input arguments */
2561 /* Output event fields */
2563 ctf_integer(int, userid, userid)
2564 ctf_integer(size_t, len, len)
2568 /* The tracepoint instances */
2569 TRACEPOINT_EVENT_INSTANCE(
2570 /* Tracepoint provider name */
2573 /* Tracepoint class name */
2576 /* Tracepoint name */
2579 /* Input arguments */
2585 TRACEPOINT_EVENT_INSTANCE(
2594 TRACEPOINT_EVENT_INSTANCE(
2607 [[assigning-log-levels]]
2608 ===== Assign a log level to a tracepoint definition
2610 You can assign an optional _log level_ to a
2611 <<defining-tracepoints,tracepoint definition>>.
2613 Assigning different levels of severity to tracepoint definitions can
2614 be useful: when you <<enabling-disabling-events,create an event rule>>,
2615 you can target tracepoints having a log level as severe as a specific
2618 The concept of LTTng-UST log levels is similar to the levels found
2619 in typical logging frameworks:
2621 * In a logging framework, the log level is given by the function
2622 or method name you use at the log statement site: `debug()`,
2623 `info()`, `warn()`, `error()`, and so on.
2624 * In LTTng-UST, you statically assign the log level to a tracepoint
2625 definition; any `tracepoint()` macro invocation which refers to
2626 this definition has this log level.
2628 You can assign a log level to a tracepoint definition with the
2629 `TRACEPOINT_LOGLEVEL()` macro. You must use this macro _after_ the
2630 <<defining-tracepoints,`TRACEPOINT_EVENT()`>> or
2631 <<using-tracepoint-classes,`TRACEPOINT_INSTANCE()`>> macro for a given
2634 The syntax of the `TRACEPOINT_LOGLEVEL()` macro is:
2637 .`TRACEPOINT_LOGLEVEL()` macro syntax.
2639 TRACEPOINT_LOGLEVEL(provider_name, tracepoint_name, log_level)
2644 * `provider_name` with the tracepoint provider name.
2645 * `tracepoint_name` with the tracepoint name.
2646 * `log_level` with the log level to assign to the tracepoint
2647 definition named `tracepoint_name` in the `provider_name`
2648 tracepoint provider.
2650 See man:lttng-ust(3) for a list of available log level names.
2652 .Assign the `TRACE_DEBUG_UNIT` log level to a tracepoint definition.
2656 /* Tracepoint definition */
2665 ctf_integer(int, userid, userid)
2666 ctf_integer(size_t, len, len)
2670 /* Log level assignment */
2671 TRACEPOINT_LOGLEVEL(my_app, get_transaction, TRACE_DEBUG_UNIT)
2677 ===== Create a tracepoint provider package source file
2679 A _tracepoint provider package source file_ is a C source file which
2680 includes a <<tpp-header,tracepoint provider header file>> to expand its
2681 macros into event serialization and other functions.
2683 You can always use the following tracepoint provider package source
2687 .Tracepoint provider package source file template.
2689 #define TRACEPOINT_CREATE_PROBES
2694 Replace `tp.h` with the name of your <<tpp-header,tracepoint provider
2695 header file>> name. You may also include more than one tracepoint
2696 provider header file here to create a tracepoint provider package
2697 holding more than one tracepoint providers.
2700 [[probing-the-application-source-code]]
2701 ==== Add tracepoints to an application's source code
2703 Once you <<tpp-header,create a tracepoint provider header file>>, you
2704 can use the `tracepoint()` macro in your application's
2705 source code to insert the tracepoints that this header
2706 <<defining-tracepoints,defines>>.
2708 The `tracepoint()` macro takes at least two parameters: the tracepoint
2709 provider name and the tracepoint name. The corresponding tracepoint
2710 definition defines the other parameters.
2712 .`tracepoint()` usage.
2714 The following <<defining-tracepoints,tracepoint definition>> defines a
2715 tracepoint which takes two input arguments and has two output event
2719 .Tracepoint provider header file.
2721 #include "my-custom-structure.h"
2728 const char*, cmd_name
2731 ctf_string(cmd_name, cmd_name)
2732 ctf_integer(int, number_of_args, argc)
2737 You can refer to this tracepoint definition with the `tracepoint()`
2738 macro in your application's source code like this:
2741 .Application's source file.
2745 int main(int argc, char* argv[])
2747 tracepoint(my_provider, my_tracepoint, argc, argv[0]);
2753 Note how the application's source code includes
2754 the tracepoint provider header file containing the tracepoint
2755 definitions to use, path:{tp.h}.
2758 .`tracepoint()` usage with a complex tracepoint definition.
2760 Consider this complex tracepoint definition, where multiple event
2761 fields refer to the same input arguments in their argument expression
2765 .Tracepoint provider header file.
2767 /* For `struct stat` */
2768 #include <sys/types.h>
2769 #include <sys/stat.h>
2781 ctf_integer(int, my_constant_field, 23 + 17)
2782 ctf_integer(int, my_int_arg_field, my_int_arg)
2783 ctf_integer(int, my_int_arg_field2, my_int_arg * my_int_arg)
2784 ctf_integer(int, sum4_field, my_str_arg[0] + my_str_arg[1] +
2785 my_str_arg[2] + my_str_arg[3])
2786 ctf_string(my_str_arg_field, my_str_arg)
2787 ctf_integer_hex(off_t, size_field, st->st_size)
2788 ctf_float(double, size_dbl_field, (double) st->st_size)
2789 ctf_sequence_text(char, half_my_str_arg_field, my_str_arg,
2790 size_t, strlen(my_str_arg) / 2)
2795 You can refer to this tracepoint definition with the `tracepoint()`
2796 macro in your application's source code like this:
2799 .Application's source file.
2801 #define TRACEPOINT_DEFINE
2808 stat("/etc/fstab", &s);
2809 tracepoint(my_provider, my_tracepoint, 23, "Hello, World!", &s);
2815 If you look at the event record that LTTng writes when tracing this
2816 program, assuming the file size of path:{/etc/fstab} is 301{nbsp}bytes,
2817 it should look like this:
2819 .Event record fields
2821 |Field's name |Field's value
2822 |`my_constant_field` |40
2823 |`my_int_arg_field` |23
2824 |`my_int_arg_field2` |529
2826 |`my_str_arg_field` |`Hello, World!`
2827 |`size_field` |0x12d
2828 |`size_dbl_field` |301.0
2829 |`half_my_str_arg_field` |`Hello,`
2833 Sometimes, the arguments you pass to `tracepoint()` are expensive to
2834 compute--they use the call stack, for example. To avoid this
2835 computation when the tracepoint is disabled, you can use the
2836 `tracepoint_enabled()` and `do_tracepoint()` macros.
2838 The syntax of the `tracepoint_enabled()` and `do_tracepoint()` macros
2842 .`tracepoint_enabled()` and `do_tracepoint()` macros syntax.
2844 tracepoint_enabled(provider_name, tracepoint_name)
2845 do_tracepoint(provider_name, tracepoint_name, ...)
2850 * `provider_name` with the tracepoint provider name.
2851 * `tracepoint_name` with the tracepoint name.
2853 `tracepoint_enabled()` returns a non-zero value if the tracepoint named
2854 `tracepoint_name` from the provider named `provider_name` is enabled
2857 `do_tracepoint()` is like `tracepoint()`, except that it doesn't check
2858 if the tracepoint is enabled. Using `tracepoint()` with
2859 `tracepoint_enabled()` is dangerous since `tracepoint()` also contains
2860 the `tracepoint_enabled()` check, thus a race condition is
2861 possible in this situation:
2864 .Possible race condition when using `tracepoint_enabled()` with `tracepoint()`.
2866 if (tracepoint_enabled(my_provider, my_tracepoint)) {
2867 stuff = prepare_stuff();
2870 tracepoint(my_provider, my_tracepoint, stuff);
2873 If the tracepoint is enabled after the condition, then `stuff` is not
2874 prepared: the emitted event will either contain wrong data, or the whole
2875 application could crash (segmentation fault, for example).
2877 NOTE: Neither `tracepoint_enabled()` nor `do_tracepoint()` have an
2878 `STAP_PROBEV()` call. If you need it, you must emit
2882 [[building-tracepoint-providers-and-user-application]]
2883 ==== Build and link a tracepoint provider package and an application
2885 Once you have one or more <<tpp-header,tracepoint provider header
2886 files>> and a <<tpp-source,tracepoint provider package source file>>,
2887 you can create the tracepoint provider package by compiling its source
2888 file. From here, multiple build and run scenarios are possible. The
2889 following table shows common application and library configurations
2890 along with the required command lines to achieve them.
2892 In the following diagrams, we use the following file names:
2895 Executable application.
2898 Application's object file.
2901 Tracepoint provider package object file.
2904 Tracepoint provider package archive file.
2907 Tracepoint provider package shared object file.
2910 User library object file.
2913 User library shared object file.
2915 We use the following symbols in the diagrams of table below:
2918 .Symbols used in the build scenario diagrams.
2919 image::ust-sit-symbols.png[]
2921 We assume that path:{.} is part of the env:LD_LIBRARY_PATH environment
2922 variable in the following instructions.
2924 [role="growable ust-scenarios",cols="asciidoc,asciidoc"]
2925 .Common tracepoint provider package scenarios.
2927 |Scenario |Instructions
2930 The instrumented application is statically linked with
2931 the tracepoint provider package object.
2933 image::ust-sit+app-linked-with-tp-o+app-instrumented.png[]
2936 include::../common/ust-sit-step-tp-o.txt[]
2938 To build the instrumented application:
2940 . In path:{app.c}, before including path:{tpp.h}, add the following line:
2945 #define TRACEPOINT_DEFINE
2949 . Compile the application source file:
2958 . Build the application:
2963 $ gcc -o app app.o tpp.o -llttng-ust -ldl
2967 To run the instrumented application:
2969 * Start the application:
2979 The instrumented application is statically linked with the
2980 tracepoint provider package archive file.
2982 image::ust-sit+app-linked-with-tp-a+app-instrumented.png[]
2985 To create the tracepoint provider package archive file:
2987 . Compile the <<tpp-source,tracepoint provider package source file>>:
2996 . Create the tracepoint provider package archive file:
3001 $ ar rcs tpp.a tpp.o
3005 To build the instrumented application:
3007 . In path:{app.c}, before including path:{tpp.h}, add the following line:
3012 #define TRACEPOINT_DEFINE
3016 . Compile the application source file:
3025 . Build the application:
3030 $ gcc -o app app.o tpp.a -llttng-ust -ldl
3034 To run the instrumented application:
3036 * Start the application:
3046 The instrumented application is linked with the tracepoint provider
3047 package shared object.
3049 image::ust-sit+app-linked-with-tp-so+app-instrumented.png[]
3052 include::../common/ust-sit-step-tp-so.txt[]
3054 To build the instrumented application:
3056 . In path:{app.c}, before including path:{tpp.h}, add the following line:
3061 #define TRACEPOINT_DEFINE
3065 . Compile the application source file:
3074 . Build the application:
3079 $ gcc -o app app.o -ldl -L. -ltpp
3083 To run the instrumented application:
3085 * Start the application:
3095 The tracepoint provider package shared object is preloaded before the
3096 instrumented application starts.
3098 image::ust-sit+tp-so-preloaded+app-instrumented.png[]
3101 include::../common/ust-sit-step-tp-so.txt[]
3103 To build the instrumented application:
3105 . In path:{app.c}, before including path:{tpp.h}, add the
3111 #define TRACEPOINT_DEFINE
3112 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3116 . Compile the application source file:
3125 . Build the application:
3130 $ gcc -o app app.o -ldl
3134 To run the instrumented application with tracing support:
3136 * Preload the tracepoint provider package shared object and
3137 start the application:
3142 $ LD_PRELOAD=./libtpp.so ./app
3146 To run the instrumented application without tracing support:
3148 * Start the application:
3158 The instrumented application dynamically loads the tracepoint provider
3159 package shared object.
3161 See the <<dlclose-warning,warning about `dlclose()`>>.
3163 image::ust-sit+app-dlopens-tp-so+app-instrumented.png[]
3166 include::../common/ust-sit-step-tp-so.txt[]
3168 To build the instrumented application:
3170 . In path:{app.c}, before including path:{tpp.h}, add the
3176 #define TRACEPOINT_DEFINE
3177 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3181 . Compile the application source file:
3190 . Build the application:
3195 $ gcc -o app app.o -ldl
3199 To run the instrumented application:
3201 * Start the application:
3211 The application is linked with the instrumented user library.
3213 The instrumented user library is statically linked with the tracepoint
3214 provider package object file.
3216 image::ust-sit+app-linked-with-lib+lib-linked-with-tp-o+lib-instrumented.png[]
3219 include::../common/ust-sit-step-tp-o-fpic.txt[]
3221 To build the instrumented user library:
3223 . In path:{emon.c}, before including path:{tpp.h}, add the
3229 #define TRACEPOINT_DEFINE
3233 . Compile the user library source file:
3238 $ gcc -I. -fpic -c emon.c
3242 . Build the user library shared object:
3247 $ gcc -shared -o libemon.so emon.o tpp.o -llttng-ust -ldl
3251 To build the application:
3253 . Compile the application source file:
3262 . Build the application:
3267 $ gcc -o app app.o -L. -lemon
3271 To run the application:
3273 * Start the application:
3283 The application is linked with the instrumented user library.
3285 The instrumented user library is linked with the tracepoint provider
3286 package shared object.
3288 image::ust-sit+app-linked-with-lib+lib-linked-with-tp-so+lib-instrumented.png[]
3291 include::../common/ust-sit-step-tp-so.txt[]
3293 To build the instrumented user library:
3295 . In path:{emon.c}, before including path:{tpp.h}, add the
3301 #define TRACEPOINT_DEFINE
3305 . Compile the user library source file:
3310 $ gcc -I. -fpic -c emon.c
3314 . Build the user library shared object:
3319 $ gcc -shared -o libemon.so emon.o -ldl -L. -ltpp
3323 To build the application:
3325 . Compile the application source file:
3334 . Build the application:
3339 $ gcc -o app app.o -L. -lemon
3343 To run the application:
3345 * Start the application:
3355 The tracepoint provider package shared object is preloaded before the
3358 The application is linked with the instrumented user library.
3360 image::ust-sit+tp-so-preloaded+app-linked-with-lib+lib-instrumented.png[]
3363 include::../common/ust-sit-step-tp-so.txt[]
3365 To build the instrumented user library:
3367 . In path:{emon.c}, before including path:{tpp.h}, add the
3373 #define TRACEPOINT_DEFINE
3374 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3378 . Compile the user library source file:
3383 $ gcc -I. -fpic -c emon.c
3387 . Build the user library shared object:
3392 $ gcc -shared -o libemon.so emon.o -ldl
3396 To build the application:
3398 . Compile the application source file:
3407 . Build the application:
3412 $ gcc -o app app.o -L. -lemon
3416 To run the application with tracing support:
3418 * Preload the tracepoint provider package shared object and
3419 start the application:
3424 $ LD_PRELOAD=./libtpp.so ./app
3428 To run the application without tracing support:
3430 * Start the application:
3440 The application is linked with the instrumented user library.
3442 The instrumented user library dynamically loads the tracepoint provider
3443 package shared object.
3445 See the <<dlclose-warning,warning about `dlclose()`>>.
3447 image::ust-sit+app-linked-with-lib+lib-dlopens-tp-so+lib-instrumented.png[]
3450 include::../common/ust-sit-step-tp-so.txt[]
3452 To build the instrumented user library:
3454 . In path:{emon.c}, before including path:{tpp.h}, add the
3460 #define TRACEPOINT_DEFINE
3461 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3465 . Compile the user library source file:
3470 $ gcc -I. -fpic -c emon.c
3474 . Build the user library shared object:
3479 $ gcc -shared -o libemon.so emon.o -ldl
3483 To build the application:
3485 . Compile the application source file:
3494 . Build the application:
3499 $ gcc -o app app.o -L. -lemon
3503 To run the application:
3505 * Start the application:
3515 The application dynamically loads the instrumented user library.
3517 The instrumented user library is linked with the tracepoint provider
3518 package shared object.
3520 See the <<dlclose-warning,warning about `dlclose()`>>.
3522 image::ust-sit+app-dlopens-lib+lib-linked-with-tp-so+lib-instrumented.png[]
3525 include::../common/ust-sit-step-tp-so.txt[]
3527 To build the instrumented user library:
3529 . In path:{emon.c}, before including path:{tpp.h}, add the
3535 #define TRACEPOINT_DEFINE
3539 . Compile the user library source file:
3544 $ gcc -I. -fpic -c emon.c
3548 . Build the user library shared object:
3553 $ gcc -shared -o libemon.so emon.o -ldl -L. -ltpp
3557 To build the application:
3559 . Compile the application source file:
3568 . Build the application:
3573 $ gcc -o app app.o -ldl -L. -lemon
3577 To run the application:
3579 * Start the application:
3589 The application dynamically loads the instrumented user library.
3591 The instrumented user library dynamically loads the tracepoint provider
3592 package shared object.
3594 See the <<dlclose-warning,warning about `dlclose()`>>.
3596 image::ust-sit+app-dlopens-lib+lib-dlopens-tp-so+lib-instrumented.png[]
3599 include::../common/ust-sit-step-tp-so.txt[]
3601 To build the instrumented user library:
3603 . In path:{emon.c}, before including path:{tpp.h}, add the
3609 #define TRACEPOINT_DEFINE
3610 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3614 . Compile the user library source file:
3619 $ gcc -I. -fpic -c emon.c
3623 . Build the user library shared object:
3628 $ gcc -shared -o libemon.so emon.o -ldl
3632 To build the application:
3634 . Compile the application source file:
3643 . Build the application:
3648 $ gcc -o app app.o -ldl -L. -lemon
3652 To run the application:
3654 * Start the application:
3664 The tracepoint provider package shared object is preloaded before the
3667 The application dynamically loads the instrumented user library.
3669 image::ust-sit+tp-so-preloaded+app-dlopens-lib+lib-instrumented.png[]
3672 include::../common/ust-sit-step-tp-so.txt[]
3674 To build the instrumented user library:
3676 . In path:{emon.c}, before including path:{tpp.h}, add the
3682 #define TRACEPOINT_DEFINE
3683 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3687 . Compile the user library source file:
3692 $ gcc -I. -fpic -c emon.c
3696 . Build the user library shared object:
3701 $ gcc -shared -o libemon.so emon.o -ldl
3705 To build the application:
3707 . Compile the application source file:
3716 . Build the application:
3721 $ gcc -o app app.o -L. -lemon
3725 To run the application with tracing support:
3727 * Preload the tracepoint provider package shared object and
3728 start the application:
3733 $ LD_PRELOAD=./libtpp.so ./app
3737 To run the application without tracing support:
3739 * Start the application:
3749 The application is statically linked with the tracepoint provider
3750 package object file.
3752 The application is linked with the instrumented user library.
3754 image::ust-sit+app-linked-with-tp-o+app-linked-with-lib+lib-instrumented.png[]
3757 include::../common/ust-sit-step-tp-o.txt[]
3759 To build the instrumented user library:
3761 . In path:{emon.c}, before including path:{tpp.h}, add the
3767 #define TRACEPOINT_DEFINE
3771 . Compile the user library source file:
3776 $ gcc -I. -fpic -c emon.c
3780 . Build the user library shared object:
3785 $ gcc -shared -o libemon.so emon.o
3789 To build the application:
3791 . Compile the application source file:
3800 . Build the application:
3805 $ gcc -o app app.o tpp.o -llttng-ust -ldl -L. -lemon
3809 To run the instrumented application:
3811 * Start the application:
3821 The application is statically linked with the tracepoint provider
3822 package object file.
3824 The application dynamically loads the instrumented user library.
3826 image::ust-sit+app-linked-with-tp-o+app-dlopens-lib+lib-instrumented.png[]
3829 include::../common/ust-sit-step-tp-o.txt[]
3831 To build the application:
3833 . In path:{app.c}, before including path:{tpp.h}, add the following line:
3838 #define TRACEPOINT_DEFINE
3842 . Compile the application source file:
3851 . Build the application:
3856 $ gcc -Wl,--export-dynamic -o app app.o tpp.o \
3861 The `--export-dynamic` option passed to the linker is necessary for the
3862 dynamically loaded library to ``see'' the tracepoint symbols defined in
3865 To build the instrumented user library:
3867 . Compile the user library source file:
3872 $ gcc -I. -fpic -c emon.c
3876 . Build the user library shared object:
3881 $ gcc -shared -o libemon.so emon.o
3885 To run the application:
3887 * Start the application:
3899 .Do not use man:dlclose(3) on a tracepoint provider package
3901 Never use man:dlclose(3) on any shared object which:
3903 * Is linked with, statically or dynamically, a tracepoint provider
3905 * Calls man:dlopen(3) itself to dynamically open a tracepoint provider
3906 package shared object.
3908 This is currently considered **unsafe** due to a lack of reference
3909 counting from LTTng-UST to the shared object.
3911 A known workaround (available since glibc 2.2) is to use the
3912 `RTLD_NODELETE` flag when calling man:dlopen(3) initially. This has the
3913 effect of not unloading the loaded shared object, even if man:dlclose(3)
3916 You can also preload the tracepoint provider package shared object with
3917 the env:LD_PRELOAD environment variable to overcome this limitation.
3921 [[using-lttng-ust-with-daemons]]
3922 ===== Use noch:{LTTng-UST} with daemons
3924 If your instrumented application calls man:fork(2), man:clone(2),
3925 or BSD's man:rfork(2), without a following man:exec(3)-family
3926 system call, you must preload the path:{liblttng-ust-fork.so} shared
3927 object when you start the application.
3931 $ LD_PRELOAD=liblttng-ust-fork.so ./my-app
3934 If your tracepoint provider package is
3935 a shared library which you also preload, you must put both
3936 shared objects in env:LD_PRELOAD:
3940 $ LD_PRELOAD=liblttng-ust-fork.so:/path/to/tp.so ./my-app
3946 ===== Use noch:{LTTng-UST} with applications which close file descriptors that don't belong to them
3948 If your instrumented application closes one or more file descriptors
3949 which it did not open itself, you must preload the
3950 path:{liblttng-ust-fd.so} shared object when you start the application:
3954 $ LD_PRELOAD=liblttng-ust-fd.so ./my-app
3957 Typical use cases include closing all the file descriptors after
3958 man:fork(2) or man:rfork(2) and buggy applications doing
3962 [[lttng-ust-pkg-config]]
3963 ===== Use noch:{pkg-config}
3965 On some distributions, LTTng-UST ships with a
3966 https://www.freedesktop.org/wiki/Software/pkg-config/[pkg-config]
3967 metadata file. If this is your case, then you can use cmd:pkg-config to
3968 build an application on the command line:
3972 $ gcc -o my-app my-app.o tp.o $(pkg-config --cflags --libs lttng-ust)
3976 [[instrumenting-32-bit-app-on-64-bit-system]]
3977 ===== [[advanced-instrumenting-techniques]]Build a 32-bit instrumented application for a 64-bit target system
3979 In order to trace a 32-bit application running on a 64-bit system,
3980 LTTng must use a dedicated 32-bit
3981 <<lttng-consumerd,consumer daemon>>.
3983 The following steps show how to build and install a 32-bit consumer
3984 daemon, which is _not_ part of the default 64-bit LTTng build, how to
3985 build and install the 32-bit LTTng-UST libraries, and how to build and
3986 link an instrumented 32-bit application in that context.
3988 To build a 32-bit instrumented application for a 64-bit target system,
3989 assuming you have a fresh target system with no installed Userspace RCU
3992 . Download, build, and install a 32-bit version of Userspace RCU:
3997 $ cd $(mktemp -d) &&
3998 wget http://lttng.org/files/urcu/userspace-rcu-latest-0.9.tar.bz2 &&
3999 tar -xf userspace-rcu-latest-0.9.tar.bz2 &&
4000 cd userspace-rcu-0.9.* &&
4001 ./configure --libdir=/usr/local/lib32 CFLAGS=-m32 &&
4003 sudo make install &&
4008 . Using your distribution's package manager, or from source, install
4009 the following 32-bit versions of the following dependencies of
4010 LTTng-tools and LTTng-UST:
4013 * https://sourceforge.net/projects/libuuid/[libuuid]
4014 * http://directory.fsf.org/wiki/Popt[popt]
4015 * http://www.xmlsoft.org/[libxml2]
4018 . Download, build, and install a 32-bit version of the latest
4019 LTTng-UST{nbsp}{revision}:
4024 $ cd $(mktemp -d) &&
4025 wget http://lttng.org/files/lttng-ust/lttng-ust-latest-2.9.tar.bz2 &&
4026 tar -xf lttng-ust-latest-2.9.tar.bz2 &&
4027 cd lttng-ust-2.9.* &&
4028 ./configure --libdir=/usr/local/lib32 \
4029 CFLAGS=-m32 CXXFLAGS=-m32 \
4030 LDFLAGS='-L/usr/local/lib32 -L/usr/lib32' &&
4032 sudo make install &&
4039 Depending on your distribution,
4040 32-bit libraries could be installed at a different location than
4041 `/usr/lib32`. For example, Debian is known to install
4042 some 32-bit libraries in `/usr/lib/i386-linux-gnu`.
4044 In this case, make sure to set `LDFLAGS` to all the
4045 relevant 32-bit library paths, for example:
4049 $ LDFLAGS='-L/usr/lib/i386-linux-gnu -L/usr/lib32'
4053 . Download the latest LTTng-tools{nbsp}{revision}, build, and install
4054 the 32-bit consumer daemon:
4059 $ cd $(mktemp -d) &&
4060 wget http://lttng.org/files/lttng-tools/lttng-tools-latest-2.9.tar.bz2 &&
4061 tar -xf lttng-tools-latest-2.9.tar.bz2 &&
4062 cd lttng-tools-2.9.* &&
4063 ./configure --libdir=/usr/local/lib32 CFLAGS=-m32 CXXFLAGS=-m32 \
4064 LDFLAGS='-L/usr/local/lib32 -L/usr/lib32' \
4065 --disable-bin-lttng --disable-bin-lttng-crash \
4066 --disable-bin-lttng-relayd --disable-bin-lttng-sessiond &&
4068 cd src/bin/lttng-consumerd &&
4069 sudo make install &&
4074 . From your distribution or from source,
4075 <<installing-lttng,install>> the 64-bit versions of
4076 LTTng-UST and Userspace RCU.
4077 . Download, build, and install the 64-bit version of the
4078 latest LTTng-tools{nbsp}{revision}:
4083 $ cd $(mktemp -d) &&
4084 wget http://lttng.org/files/lttng-tools/lttng-tools-latest-2.9.tar.bz2 &&
4085 tar -xf lttng-tools-latest-2.9.tar.bz2 &&
4086 cd lttng-tools-2.9.* &&
4087 ./configure --with-consumerd32-libdir=/usr/local/lib32 \
4088 --with-consumerd32-bin=/usr/local/lib32/lttng/libexec/lttng-consumerd &&
4090 sudo make install &&
4095 . Pass the following options to man:gcc(1), man:g++(1), or man:clang(1)
4096 when linking your 32-bit application:
4099 -m32 -L/usr/lib32 -L/usr/local/lib32 \
4100 -Wl,-rpath,/usr/lib32,-rpath,/usr/local/lib32
4103 For example, let's rebuild the quick start example in
4104 <<tracing-your-own-user-application,Trace a user application>> as an
4105 instrumented 32-bit application:
4110 $ gcc -m32 -c -I. hello-tp.c
4111 $ gcc -m32 -c hello.c
4112 $ gcc -m32 -o hello hello.o hello-tp.o \
4113 -L/usr/lib32 -L/usr/local/lib32 \
4114 -Wl,-rpath,/usr/lib32,-rpath,/usr/local/lib32 \
4119 No special action is required to execute the 32-bit application and
4120 to trace it: use the command-line man:lttng(1) tool as usual.
4127 man:tracef(3) is a small LTTng-UST API designed for quick,
4128 man:printf(3)-like instrumentation without the burden of
4129 <<tracepoint-provider,creating>> and
4130 <<building-tracepoint-providers-and-user-application,building>>
4131 a tracepoint provider package.
4133 To use `tracef()` in your application:
4135 . In the C or C++ source files where you need to use `tracef()`,
4136 include `<lttng/tracef.h>`:
4141 #include <lttng/tracef.h>
4145 . In the application's source code, use `tracef()` like you would use
4153 tracef("my message: %d (%s)", my_integer, my_string);
4159 . Link your application with `liblttng-ust`:
4164 $ gcc -o app app.c -llttng-ust
4168 To trace the events that `tracef()` calls emit:
4170 * <<enabling-disabling-events,Create an event rule>> which matches the
4171 `lttng_ust_tracef:*` event name:
4176 $ lttng enable-event --userspace 'lttng_ust_tracef:*'
4181 .Limitations of `tracef()`
4183 The `tracef()` utility function was developed to make user space tracing
4184 super simple, albeit with notable disadvantages compared to
4185 <<defining-tracepoints,user-defined tracepoints>>:
4187 * All the emitted events have the same tracepoint provider and
4188 tracepoint names, respectively `lttng_ust_tracef` and `event`.
4189 * There is no static type checking.
4190 * The only event record field you actually get, named `msg`, is a string
4191 potentially containing the values you passed to `tracef()`
4192 using your own format string. This also means that you cannot filter
4193 events with a custom expression at run time because there are no
4195 * Since `tracef()` uses the C standard library's man:vasprintf(3)
4196 function behind the scenes to format the strings at run time, its
4197 expected performance is lower than with user-defined tracepoints,
4198 which do not require a conversion to a string.
4200 Taking this into consideration, `tracef()` is useful for some quick
4201 prototyping and debugging, but you should not consider it for any
4202 permanent and serious applicative instrumentation.
4208 ==== Use `tracelog()`
4210 The man:tracelog(3) API is very similar to <<tracef,`tracef()`>>, with
4211 the difference that it accepts an additional log level parameter.
4213 The goal of `tracelog()` is to ease the migration from logging to
4216 To use `tracelog()` in your application:
4218 . In the C or C++ source files where you need to use `tracelog()`,
4219 include `<lttng/tracelog.h>`:
4224 #include <lttng/tracelog.h>
4228 . In the application's source code, use `tracelog()` like you would use
4229 man:printf(3), except for the first parameter which is the log
4237 tracelog(TRACE_WARNING, "my message: %d (%s)",
4238 my_integer, my_string);
4244 See man:lttng-ust(3) for a list of available log level names.
4246 . Link your application with `liblttng-ust`:
4251 $ gcc -o app app.c -llttng-ust
4255 To trace the events that `tracelog()` calls emit with a log level
4256 _as severe as_ a specific log level:
4258 * <<enabling-disabling-events,Create an event rule>> which matches the
4259 `lttng_ust_tracelog:*` event name and a minimum level
4265 $ lttng enable-event --userspace 'lttng_ust_tracelog:*'
4266 --loglevel=TRACE_WARNING
4270 To trace the events that `tracelog()` calls emit with a
4271 _specific log level_:
4273 * Create an event rule which matches the `lttng_ust_tracelog:*`
4274 event name and a specific log level:
4279 $ lttng enable-event --userspace 'lttng_ust_tracelog:*'
4280 --loglevel-only=TRACE_INFO
4285 [[prebuilt-ust-helpers]]
4286 === Prebuilt user space tracing helpers
4288 The LTTng-UST package provides a few helpers in the form or preloadable
4289 shared objects which automatically instrument system functions and
4292 The helper shared objects are normally found in dir:{/usr/lib}. If you
4293 built LTTng-UST <<building-from-source,from source>>, they are probably
4294 located in dir:{/usr/local/lib}.
4296 The installed user space tracing helpers in LTTng-UST{nbsp}{revision}
4299 path:{liblttng-ust-libc-wrapper.so}::
4300 path:{liblttng-ust-pthread-wrapper.so}::
4301 <<liblttng-ust-libc-pthread-wrapper,C{nbsp}standard library
4302 memory and POSIX threads function tracing>>.
4304 path:{liblttng-ust-cyg-profile.so}::
4305 path:{liblttng-ust-cyg-profile-fast.so}::
4306 <<liblttng-ust-cyg-profile,Function entry and exit tracing>>.
4308 path:{liblttng-ust-dl.so}::
4309 <<liblttng-ust-dl,Dynamic linker tracing>>.
4311 To use a user space tracing helper with any user application:
4313 * Preload the helper shared object when you start the application:
4318 $ LD_PRELOAD=liblttng-ust-libc-wrapper.so my-app
4322 You can preload more than one helper:
4327 $ LD_PRELOAD=liblttng-ust-libc-wrapper.so:liblttng-ust-dl.so my-app
4333 [[liblttng-ust-libc-pthread-wrapper]]
4334 ==== Instrument C standard library memory and POSIX threads functions
4336 The path:{liblttng-ust-libc-wrapper.so} and
4337 path:{liblttng-ust-pthread-wrapper.so} helpers
4338 add instrumentation to some C standard library and POSIX
4342 .Functions instrumented by preloading path:{liblttng-ust-libc-wrapper.so}.
4344 |TP provider name |TP name |Instrumented function
4346 .6+|`lttng_ust_libc` |`malloc` |man:malloc(3)
4347 |`calloc` |man:calloc(3)
4348 |`realloc` |man:realloc(3)
4349 |`free` |man:free(3)
4350 |`memalign` |man:memalign(3)
4351 |`posix_memalign` |man:posix_memalign(3)
4355 .Functions instrumented by preloading path:{liblttng-ust-pthread-wrapper.so}.
4357 |TP provider name |TP name |Instrumented function
4359 .4+|`lttng_ust_pthread` |`pthread_mutex_lock_req` |man:pthread_mutex_lock(3p) (request time)
4360 |`pthread_mutex_lock_acq` |man:pthread_mutex_lock(3p) (acquire time)
4361 |`pthread_mutex_trylock` |man:pthread_mutex_trylock(3p)
4362 |`pthread_mutex_unlock` |man:pthread_mutex_unlock(3p)
4365 When you preload the shared object, it replaces the functions listed
4366 in the previous tables by wrappers which contain tracepoints and call
4367 the replaced functions.
4370 [[liblttng-ust-cyg-profile]]
4371 ==== Instrument function entry and exit
4373 The path:{liblttng-ust-cyg-profile*.so} helpers can add instrumentation
4374 to the entry and exit points of functions.
4376 man:gcc(1) and man:clang(1) have an option named
4377 https://gcc.gnu.org/onlinedocs/gcc/Instrumentation-Options.html[`-finstrument-functions`]
4378 which generates instrumentation calls for entry and exit to functions.
4379 The LTTng-UST function tracing helpers,
4380 path:{liblttng-ust-cyg-profile.so} and
4381 path:{liblttng-ust-cyg-profile-fast.so}, take advantage of this feature
4382 to add tracepoints to the two generated functions (which contain
4383 `cyg_profile` in their names, hence the helper's name).
4385 To use the LTTng-UST function tracing helper, the source files to
4386 instrument must be built using the `-finstrument-functions` compiler
4389 There are two versions of the LTTng-UST function tracing helper:
4391 * **path:{liblttng-ust-cyg-profile-fast.so}** is a lightweight variant
4392 that you should only use when it can be _guaranteed_ that the
4393 complete event stream is recorded without any lost event record.
4394 Any kind of duplicate information is left out.
4396 Assuming no event record is lost, having only the function addresses on
4397 entry is enough to create a call graph, since an event record always
4398 contains the ID of the CPU that generated it.
4400 You can use a tool like man:addr2line(1) to convert function addresses
4401 back to source file names and line numbers.
4403 * **path:{liblttng-ust-cyg-profile.so}** is a more robust variant
4404 which also works in use cases where event records might get discarded or
4405 not recorded from application startup.
4406 In these cases, the trace analyzer needs more information to be
4407 able to reconstruct the program flow.
4409 See man:lttng-ust-cyg-profile(3) to learn more about the instrumentation
4410 points of this helper.
4412 All the tracepoints that this helper provides have the
4413 log level `TRACE_DEBUG_FUNCTION` (see man:lttng-ust(3)).
4415 TIP: It's sometimes a good idea to limit the number of source files that
4416 you compile with the `-finstrument-functions` option to prevent LTTng
4417 from writing an excessive amount of trace data at run time. When using
4418 man:gcc(1), you can use the
4419 `-finstrument-functions-exclude-function-list` option to avoid
4420 instrument entries and exits of specific function names.
4425 ==== Instrument the dynamic linker
4427 The path:{liblttng-ust-dl.so} helper adds instrumentation to the
4428 man:dlopen(3) and man:dlclose(3) function calls.
4430 See man:lttng-ust-dl(3) to learn more about the instrumentation points
4435 [[java-application]]
4436 === User space Java agent
4438 You can instrument any Java application which uses one of the following
4441 * The https://docs.oracle.com/javase/7/docs/api/java/util/logging/package-summary.html[**`java.util.logging`**]
4442 (JUL) core logging facilities.
4443 * http://logging.apache.org/log4j/1.2/[**Apache log4j 1.2**], since
4444 LTTng 2.6. Note that Apache Log4j{nbsp}2 is not supported.
4447 .LTTng-UST Java agent imported by a Java application.
4448 image::java-app.png[]
4450 Note that the methods described below are new in LTTng{nbsp}{revision}.
4451 Previous LTTng versions use another technique.
4453 NOTE: We use http://openjdk.java.net/[OpenJDK]{nbsp}8 for development
4454 and https://ci.lttng.org/[continuous integration], thus this version is
4455 directly supported. However, the LTTng-UST Java agent is also tested
4456 with OpenJDK{nbsp}7.
4461 ==== Use the LTTng-UST Java agent for `java.util.logging`
4463 To use the LTTng-UST Java agent in a Java application which uses
4464 `java.util.logging` (JUL):
4466 . In the Java application's source code, import the LTTng-UST
4467 log handler package for `java.util.logging`:
4472 import org.lttng.ust.agent.jul.LttngLogHandler;
4476 . Create an LTTng-UST JUL log handler:
4481 Handler lttngUstLogHandler = new LttngLogHandler();
4485 . Add this handler to the JUL loggers which should emit LTTng events:
4490 Logger myLogger = Logger.getLogger("some-logger");
4492 myLogger.addHandler(lttngUstLogHandler);
4496 . Use `java.util.logging` log statements and configuration as usual.
4497 The loggers with an attached LTTng-UST log handler can emit
4500 . Before exiting the application, remove the LTTng-UST log handler from
4501 the loggers attached to it and call its `close()` method:
4506 myLogger.removeHandler(lttngUstLogHandler);
4507 lttngUstLogHandler.close();
4511 This is not strictly necessary, but it is recommended for a clean
4512 disposal of the handler's resources.
4514 . Include the LTTng-UST Java agent's common and JUL-specific JAR files,
4515 path:{lttng-ust-agent-common.jar} and path:{lttng-ust-agent-jul.jar},
4517 https://docs.oracle.com/javase/tutorial/essential/environment/paths.html[class
4518 path] when you build the Java application.
4520 The JAR files are typically located in dir:{/usr/share/java}.
4522 IMPORTANT: The LTTng-UST Java agent must be
4523 <<installing-lttng,installed>> for the logging framework your
4526 .Use the LTTng-UST Java agent for `java.util.logging`.
4531 import java.io.IOException;
4532 import java.util.logging.Handler;
4533 import java.util.logging.Logger;
4534 import org.lttng.ust.agent.jul.LttngLogHandler;
4538 private static final int answer = 42;
4540 public static void main(String[] argv) throws Exception
4543 Logger logger = Logger.getLogger("jello");
4545 // Create an LTTng-UST log handler
4546 Handler lttngUstLogHandler = new LttngLogHandler();
4548 // Add the LTTng-UST log handler to our logger
4549 logger.addHandler(lttngUstLogHandler);
4552 logger.info("some info");
4553 logger.warning("some warning");
4555 logger.finer("finer information; the answer is " + answer);
4557 logger.severe("error!");
4559 // Not mandatory, but cleaner
4560 logger.removeHandler(lttngUstLogHandler);
4561 lttngUstLogHandler.close();
4570 $ javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar Test.java
4573 <<creating-destroying-tracing-sessions,Create a tracing session>>,
4574 <<enabling-disabling-events,create an event rule>> matching the
4575 `jello` JUL logger, and <<basic-tracing-session-control,start tracing>>:
4580 $ lttng enable-event --jul jello
4584 Run the compiled class:
4588 $ java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar:. Test
4591 <<basic-tracing-session-control,Stop tracing>> and inspect the
4601 In the resulting trace, an <<event,event record>> generated by a Java
4602 application using `java.util.logging` is named `lttng_jul:event` and
4603 has the following fields:
4606 Log record's message.
4612 Name of the class in which the log statement was executed.
4615 Name of the method in which the log statement was executed.
4618 Logging time (timestamp in milliseconds).
4621 Log level integer value.
4624 ID of the thread in which the log statement was executed.
4626 You can use the opt:lttng-enable-event(1):--loglevel or
4627 opt:lttng-enable-event(1):--loglevel-only option of the
4628 man:lttng-enable-event(1) command to target a range of JUL log levels
4629 or a specific JUL log level.
4634 ==== Use the LTTng-UST Java agent for Apache log4j
4636 To use the LTTng-UST Java agent in a Java application which uses
4639 . In the Java application's source code, import the LTTng-UST
4640 log appender package for Apache log4j:
4645 import org.lttng.ust.agent.log4j.LttngLogAppender;
4649 . Create an LTTng-UST log4j log appender:
4654 Appender lttngUstLogAppender = new LttngLogAppender();
4658 . Add this appender to the log4j loggers which should emit LTTng events:
4663 Logger myLogger = Logger.getLogger("some-logger");
4665 myLogger.addAppender(lttngUstLogAppender);
4669 . Use Apache log4j log statements and configuration as usual. The
4670 loggers with an attached LTTng-UST log appender can emit LTTng events.
4672 . Before exiting the application, remove the LTTng-UST log appender from
4673 the loggers attached to it and call its `close()` method:
4678 myLogger.removeAppender(lttngUstLogAppender);
4679 lttngUstLogAppender.close();
4683 This is not strictly necessary, but it is recommended for a clean
4684 disposal of the appender's resources.
4686 . Include the LTTng-UST Java agent's common and log4j-specific JAR
4687 files, path:{lttng-ust-agent-common.jar} and
4688 path:{lttng-ust-agent-log4j.jar}, in the
4689 https://docs.oracle.com/javase/tutorial/essential/environment/paths.html[class
4690 path] when you build the Java application.
4692 The JAR files are typically located in dir:{/usr/share/java}.
4694 IMPORTANT: The LTTng-UST Java agent must be
4695 <<installing-lttng,installed>> for the logging framework your
4698 .Use the LTTng-UST Java agent for Apache log4j.
4703 import org.apache.log4j.Appender;
4704 import org.apache.log4j.Logger;
4705 import org.lttng.ust.agent.log4j.LttngLogAppender;
4709 private static final int answer = 42;
4711 public static void main(String[] argv) throws Exception
4714 Logger logger = Logger.getLogger("jello");
4716 // Create an LTTng-UST log appender
4717 Appender lttngUstLogAppender = new LttngLogAppender();
4719 // Add the LTTng-UST log appender to our logger
4720 logger.addAppender(lttngUstLogAppender);
4723 logger.info("some info");
4724 logger.warn("some warning");
4726 logger.debug("debug information; the answer is " + answer);
4728 logger.fatal("error!");
4730 // Not mandatory, but cleaner
4731 logger.removeAppender(lttngUstLogAppender);
4732 lttngUstLogAppender.close();
4738 Build this example (`$LOG4JPATH` is the path to the Apache log4j JAR
4743 $ javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-log4j.jar:$LOG4JPATH Test.java
4746 <<creating-destroying-tracing-sessions,Create a tracing session>>,
4747 <<enabling-disabling-events,create an event rule>> matching the
4748 `jello` log4j logger, and <<basic-tracing-session-control,start tracing>>:
4753 $ lttng enable-event --log4j jello
4757 Run the compiled class:
4761 $ java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-log4j.jar:$LOG4JPATH:. Test
4764 <<basic-tracing-session-control,Stop tracing>> and inspect the
4774 In the resulting trace, an <<event,event record>> generated by a Java
4775 application using log4j is named `lttng_log4j:event` and
4776 has the following fields:
4779 Log record's message.
4785 Name of the class in which the log statement was executed.
4788 Name of the method in which the log statement was executed.
4791 Name of the file in which the executed log statement is located.
4794 Line number at which the log statement was executed.
4800 Log level integer value.
4803 Name of the Java thread in which the log statement was executed.
4805 You can use the opt:lttng-enable-event(1):--loglevel or
4806 opt:lttng-enable-event(1):--loglevel-only option of the
4807 man:lttng-enable-event(1) command to target a range of Apache log4j log levels
4808 or a specific log4j log level.
4812 [[java-application-context]]
4813 ==== Provide application-specific context fields in a Java application
4815 A Java application-specific context field is a piece of state provided
4816 by the application which <<adding-context,you can add>>, using the
4817 man:lttng-add-context(1) command, to each <<event,event record>>
4818 produced by the log statements of this application.
4820 For example, a given object might have a current request ID variable.
4821 You can create a context information retriever for this object and
4822 assign a name to this current request ID. You can then, using the
4823 man:lttng-add-context(1) command, add this context field by name to
4824 the JUL or log4j <<channel,channel>>.
4826 To provide application-specific context fields in a Java application:
4828 . In the Java application's source code, import the LTTng-UST
4829 Java agent context classes and interfaces:
4834 import org.lttng.ust.agent.context.ContextInfoManager;
4835 import org.lttng.ust.agent.context.IContextInfoRetriever;
4839 . Create a context information retriever class, that is, a class which
4840 implements the `IContextInfoRetriever` interface:
4845 class MyContextInfoRetriever implements IContextInfoRetriever
4848 public Object retrieveContextInfo(String key)
4850 if (key.equals("intCtx")) {
4852 } else if (key.equals("strContext")) {
4853 return "context value!";
4862 This `retrieveContextInfo()` method is the only member of the
4863 `IContextInfoRetriever` interface. Its role is to return the current
4864 value of a state by name to create a context field. The names of the
4865 context fields and which state variables they return depends on your
4868 All primitive types and objects are supported as context fields.
4869 When `retrieveContextInfo()` returns an object, the context field
4870 serializer calls its `toString()` method to add a string field to
4871 event records. The method can also return `null`, which means that
4872 no context field is available for the required name.
4874 . Register an instance of your context information retriever class to
4875 the context information manager singleton:
4880 IContextInfoRetriever cir = new MyContextInfoRetriever();
4881 ContextInfoManager cim = ContextInfoManager.getInstance();
4882 cim.registerContextInfoRetriever("retrieverName", cir);
4886 . Before exiting the application, remove your context information
4887 retriever from the context information manager singleton:
4892 ContextInfoManager cim = ContextInfoManager.getInstance();
4893 cim.unregisterContextInfoRetriever("retrieverName");
4897 This is not strictly necessary, but it is recommended for a clean
4898 disposal of some manager's resources.
4900 . Build your Java application with LTTng-UST Java agent support as
4901 usual, following the procedure for either the <<jul,JUL>> or
4902 <<log4j,Apache log4j>> framework.
4905 .Provide application-specific context fields in a Java application.
4910 import java.util.logging.Handler;
4911 import java.util.logging.Logger;
4912 import org.lttng.ust.agent.jul.LttngLogHandler;
4913 import org.lttng.ust.agent.context.ContextInfoManager;
4914 import org.lttng.ust.agent.context.IContextInfoRetriever;
4918 // Our context information retriever class
4919 private static class MyContextInfoRetriever
4920 implements IContextInfoRetriever
4923 public Object retrieveContextInfo(String key) {
4924 if (key.equals("intCtx")) {
4926 } else if (key.equals("strContext")) {
4927 return "context value!";
4934 private static final int answer = 42;
4936 public static void main(String args[]) throws Exception
4938 // Get the context information manager instance
4939 ContextInfoManager cim = ContextInfoManager.getInstance();
4941 // Create and register our context information retriever
4942 IContextInfoRetriever cir = new MyContextInfoRetriever();
4943 cim.registerContextInfoRetriever("myRetriever", cir);
4946 Logger logger = Logger.getLogger("jello");
4948 // Create an LTTng-UST log handler
4949 Handler lttngUstLogHandler = new LttngLogHandler();
4951 // Add the LTTng-UST log handler to our logger
4952 logger.addHandler(lttngUstLogHandler);
4955 logger.info("some info");
4956 logger.warning("some warning");
4958 logger.finer("finer information; the answer is " + answer);
4960 logger.severe("error!");
4962 // Not mandatory, but cleaner
4963 logger.removeHandler(lttngUstLogHandler);
4964 lttngUstLogHandler.close();
4965 cim.unregisterContextInfoRetriever("myRetriever");
4974 $ javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar Test.java
4977 <<creating-destroying-tracing-sessions,Create a tracing session>>
4978 and <<enabling-disabling-events,create an event rule>> matching the
4984 $ lttng enable-event --jul jello
4987 <<adding-context,Add the application-specific context fields>> to the
4992 $ lttng add-context --jul --type='$app.myRetriever:intCtx'
4993 $ lttng add-context --jul --type='$app.myRetriever:strContext'
4996 <<basic-tracing-session-control,Start tracing>>:
5003 Run the compiled class:
5007 $ java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar:. Test
5010 <<basic-tracing-session-control,Stop tracing>> and inspect the
5022 [[python-application]]
5023 === User space Python agent
5025 You can instrument a Python 2 or Python 3 application which uses the
5026 standard https://docs.python.org/3/library/logging.html[`logging`]
5029 Each log statement emits an LTTng event once the
5030 application module imports the
5031 <<lttng-ust-agents,LTTng-UST Python agent>> package.
5034 .A Python application importing the LTTng-UST Python agent.
5035 image::python-app.png[]
5037 To use the LTTng-UST Python agent:
5039 . In the Python application's source code, import the LTTng-UST Python
5049 The LTTng-UST Python agent automatically adds its logging handler to the
5050 root logger at import time.
5052 Any log statement that the application executes before this import does
5053 not emit an LTTng event.
5055 IMPORTANT: The LTTng-UST Python agent must be
5056 <<installing-lttng,installed>>.
5058 . Use log statements and logging configuration as usual.
5059 Since the LTTng-UST Python agent adds a handler to the _root_
5060 logger, you can trace any log statement from any logger.
5062 .Use the LTTng-UST Python agent.
5073 logging.basicConfig()
5074 logger = logging.getLogger('my-logger')
5077 logger.debug('debug message')
5078 logger.info('info message')
5079 logger.warn('warn message')
5080 logger.error('error message')
5081 logger.critical('critical message')
5085 if __name__ == '__main__':
5089 NOTE: `logging.basicConfig()`, which adds to the root logger a basic
5090 logging handler which prints to the standard error stream, is not
5091 strictly required for LTTng-UST tracing to work, but in versions of
5092 Python preceding 3.2, you could see a warning message which indicates
5093 that no handler exists for the logger `my-logger`.
5095 <<creating-destroying-tracing-sessions,Create a tracing session>>,
5096 <<enabling-disabling-events,create an event rule>> matching the
5097 `my-logger` Python logger, and <<basic-tracing-session-control,start
5103 $ lttng enable-event --python my-logger
5107 Run the Python script:
5114 <<basic-tracing-session-control,Stop tracing>> and inspect the recorded
5124 In the resulting trace, an <<event,event record>> generated by a Python
5125 application is named `lttng_python:event` and has the following fields:
5128 Logging time (string).
5131 Log record's message.
5137 Name of the function in which the log statement was executed.
5140 Line number at which the log statement was executed.
5143 Log level integer value.
5146 ID of the Python thread in which the log statement was executed.
5149 Name of the Python thread in which the log statement was executed.
5151 You can use the opt:lttng-enable-event(1):--loglevel or
5152 opt:lttng-enable-event(1):--loglevel-only option of the
5153 man:lttng-enable-event(1) command to target a range of Python log levels
5154 or a specific Python log level.
5156 When an application imports the LTTng-UST Python agent, the agent tries
5157 to register to a <<lttng-sessiond,session daemon>>. Note that you must
5158 <<start-sessiond,start the session daemon>> _before_ you run the Python
5159 application. If a session daemon is found, the agent tries to register
5160 to it during 5{nbsp}seconds, after which the application continues
5161 without LTTng tracing support. You can override this timeout value with
5162 the env:LTTNG_UST_PYTHON_REGISTER_TIMEOUT environment variable
5165 If the session daemon stops while a Python application with an imported
5166 LTTng-UST Python agent runs, the agent retries to connect and to
5167 register to a session daemon every 3{nbsp}seconds. You can override this
5168 delay with the env:LTTNG_UST_PYTHON_REGISTER_RETRY_DELAY environment
5173 [[proc-lttng-logger-abi]]
5176 The `lttng-tracer` Linux kernel module, part of
5177 <<lttng-modules,LTTng-modules>>, creates the special LTTng logger file
5178 path:{/proc/lttng-logger} when it's loaded. Any application can write
5179 text data to this file to emit an LTTng event.
5182 .An application writes to the LTTng logger file to emit an LTTng event.
5183 image::lttng-logger.png[]
5185 The LTTng logger is the quickest method--not the most efficient,
5186 however--to add instrumentation to an application. It is designed
5187 mostly to instrument shell scripts:
5191 $ echo "Some message, some $variable" > /proc/lttng-logger
5194 Any event that the LTTng logger emits is named `lttng_logger` and
5195 belongs to the Linux kernel <<domain,tracing domain>>. However, unlike
5196 other instrumentation points in the kernel tracing domain, **any Unix
5197 user** can <<enabling-disabling-events,create an event rule>> which
5198 matches its event name, not only the root user or users in the
5199 <<tracing-group,tracing group>>.
5201 To use the LTTng logger:
5203 * From any application, write text data to the path:{/proc/lttng-logger}
5206 The `msg` field of `lttng_logger` event records contains the
5209 NOTE: The maximum message length of an LTTng logger event is
5210 1024{nbsp}bytes. Writing more than this makes the LTTng logger emit more
5211 than one event to contain the remaining data.
5213 You should not use the LTTng logger to trace a user application which
5214 can be instrumented in a more efficient way, namely:
5216 * <<c-application,C and $$C++$$ applications>>.
5217 * <<java-application,Java applications>>.
5218 * <<python-application,Python applications>>.
5220 .Use the LTTng logger.
5225 echo 'Hello, World!' > /proc/lttng-logger
5227 df --human-readable --print-type / > /proc/lttng-logger
5230 <<creating-destroying-tracing-sessions,Create a tracing session>>,
5231 <<enabling-disabling-events,create an event rule>> matching the
5232 `lttng_logger` Linux kernel tracepoint, and
5233 <<basic-tracing-session-control,start tracing>>:
5238 $ lttng enable-event --kernel lttng_logger
5242 Run the Bash script:
5249 <<basic-tracing-session-control,Stop tracing>> and inspect the recorded
5260 [[instrumenting-linux-kernel]]
5261 === LTTng kernel tracepoints
5263 NOTE: This section shows how to _add_ instrumentation points to the
5264 Linux kernel. The kernel's subsystems are already thoroughly
5265 instrumented at strategic places for LTTng when you
5266 <<installing-lttng,install>> the <<lttng-modules,LTTng-modules>>
5270 There are two methods to instrument the Linux kernel:
5272 . <<linux-add-lttng-layer,Add an LTTng layer>> over an existing ftrace
5273 tracepoint which uses the `TRACE_EVENT()` API.
5275 Choose this if you want to instrumentation a Linux kernel tree with an
5276 instrumentation point compatible with ftrace, perf, and SystemTap.
5278 . Use an <<linux-lttng-tracepoint-event,LTTng-only approach>> to
5279 instrument an out-of-tree kernel module.
5281 Choose this if you don't need ftrace, perf, or SystemTap support.
5285 [[linux-add-lttng-layer]]
5286 ==== [[instrumenting-linux-kernel-itself]][[mainline-trace-event]][[lttng-adaptation-layer]]Add an LTTng layer to an existing ftrace tracepoint
5288 This section shows how to add an LTTng layer to existing ftrace
5289 instrumentation using the `TRACE_EVENT()` API.
5291 This section does not document the `TRACE_EVENT()` macro. You can
5292 read the following articles to learn more about this API:
5294 * http://lwn.net/Articles/379903/[Using the TRACE_EVENT() macro (Part 1)]
5295 * http://lwn.net/Articles/381064/[Using the TRACE_EVENT() macro (Part 2)]
5296 * http://lwn.net/Articles/383362/[Using the TRACE_EVENT() macro (Part 3)]
5298 The following procedure assumes that your ftrace tracepoints are
5299 correctly defined in their own header and that they are created in
5300 one source file using the `CREATE_TRACE_POINTS` definition.
5302 To add an LTTng layer over an existing ftrace tracepoint:
5304 . Make sure the following kernel configuration options are
5310 * `CONFIG_HIGH_RES_TIMERS`
5311 * `CONFIG_TRACEPOINTS`
5314 . Build the Linux source tree with your custom ftrace tracepoints.
5315 . Boot the resulting Linux image on your target system.
5317 Confirm that the tracepoints exist by looking for their names in the
5318 dir:{/sys/kernel/debug/tracing/events/subsys} directory, where `subsys`
5319 is your subsystem's name.
5321 . Get a copy of the latest LTTng-modules{nbsp}{revision}:
5326 $ cd $(mktemp -d) &&
5327 wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.9.tar.bz2 &&
5328 tar -xf lttng-modules-latest-2.9.tar.bz2 &&
5329 cd lttng-modules-2.9.*
5333 . In dir:{instrumentation/events/lttng-module}, relative to the root
5334 of the LTTng-modules source tree, create a header file named
5335 +__subsys__.h+ for your custom subsystem +__subsys__+ and write your
5336 LTTng-modules tracepoint definitions using the LTTng-modules
5339 Start with this template:
5343 .path:{instrumentation/events/lttng-module/my_subsys.h}
5346 #define TRACE_SYSTEM my_subsys
5348 #if !defined(_LTTNG_MY_SUBSYS_H) || defined(TRACE_HEADER_MULTI_READ)
5349 #define _LTTNG_MY_SUBSYS_H
5351 #include "../../../probes/lttng-tracepoint-event.h"
5352 #include <linux/tracepoint.h>
5354 LTTNG_TRACEPOINT_EVENT(
5356 * Format is identical to TRACE_EVENT()'s version for the three
5357 * following macro parameters:
5360 TP_PROTO(int my_int, const char *my_string),
5361 TP_ARGS(my_int, my_string),
5363 /* LTTng-modules specific macros */
5365 ctf_integer(int, my_int_field, my_int)
5366 ctf_string(my_bar_field, my_bar)
5370 #endif /* !defined(_LTTNG_MY_SUBSYS_H) || defined(TRACE_HEADER_MULTI_READ) */
5372 #include "../../../probes/define_trace.h"
5376 The entries in the `TP_FIELDS()` section are the list of fields for the
5377 LTTng tracepoint. This is similar to the `TP_STRUCT__entry()` part of
5378 ftrace's `TRACE_EVENT()` macro.
5380 See <<lttng-modules-tp-fields,Tracepoint fields macros>> for a
5381 complete description of the available `ctf_*()` macros.
5383 . Create the LTTng-modules probe's kernel module C source file,
5384 +probes/lttng-probe-__subsys__.c+, where +__subsys__+ is your
5389 .path:{probes/lttng-probe-my-subsys.c}
5391 #include <linux/module.h>
5392 #include "../lttng-tracer.h"
5395 * Build-time verification of mismatch between mainline
5396 * TRACE_EVENT() arguments and the LTTng-modules adaptation
5397 * layer LTTNG_TRACEPOINT_EVENT() arguments.
5399 #include <trace/events/my_subsys.h>
5401 /* Create LTTng tracepoint probes */
5402 #define LTTNG_PACKAGE_BUILD
5403 #define CREATE_TRACE_POINTS
5404 #define TRACE_INCLUDE_PATH ../instrumentation/events/lttng-module
5406 #include "../instrumentation/events/lttng-module/my_subsys.h"
5408 MODULE_LICENSE("GPL and additional rights");
5409 MODULE_AUTHOR("Your name <your-email>");
5410 MODULE_DESCRIPTION("LTTng my_subsys probes");
5411 MODULE_VERSION(__stringify(LTTNG_MODULES_MAJOR_VERSION) "."
5412 __stringify(LTTNG_MODULES_MINOR_VERSION) "."
5413 __stringify(LTTNG_MODULES_PATCHLEVEL_VERSION)
5414 LTTNG_MODULES_EXTRAVERSION);
5418 . Edit path:{probes/KBuild} and add your new kernel module object
5419 next to the existing ones:
5423 .path:{probes/KBuild}
5427 obj-m += lttng-probe-module.o
5428 obj-m += lttng-probe-power.o
5430 obj-m += lttng-probe-my-subsys.o
5436 . Build and install the LTTng kernel modules:
5441 $ make KERNELDIR=/path/to/linux
5442 # make modules_install && depmod -a
5446 Replace `/path/to/linux` with the path to the Linux source tree where
5447 you defined and used tracepoints with ftrace's `TRACE_EVENT()` macro.
5449 Note that you can also use the
5450 <<lttng-tracepoint-event-code,`LTTNG_TRACEPOINT_EVENT_CODE()` macro>>
5451 instead of `LTTNG_TRACEPOINT_EVENT()` to use custom local variables and
5452 C code that need to be executed before the event fields are recorded.
5454 The best way to learn how to use the previous LTTng-modules macros is to
5455 inspect the existing LTTng-modules tracepoint definitions in the
5456 dir:{instrumentation/events/lttng-module} header files. Compare them
5457 with the Linux kernel mainline versions in the
5458 dir:{include/trace/events} directory of the Linux source tree.
5462 [[lttng-tracepoint-event-code]]
5463 ===== Use custom C code to access the data for tracepoint fields
5465 Although we recommended to always use the
5466 <<lttng-adaptation-layer,`LTTNG_TRACEPOINT_EVENT()`>> macro to describe
5467 the arguments and fields of an LTTng-modules tracepoint when possible,
5468 sometimes you need a more complex process to access the data that the
5469 tracer records as event record fields. In other words, you need local
5470 variables and multiple C{nbsp}statements instead of simple
5471 argument-based expressions that you pass to the
5472 <<lttng-modules-tp-fields,`ctf_*()` macros of `TP_FIELDS()`>>.
5474 You can use the `LTTNG_TRACEPOINT_EVENT_CODE()` macro instead of
5475 `LTTNG_TRACEPOINT_EVENT()` to declare custom local variables and define
5476 a block of C{nbsp}code to be executed before LTTng records the fields.
5477 The structure of this macro is:
5480 .`LTTNG_TRACEPOINT_EVENT_CODE()` macro syntax.
5482 LTTNG_TRACEPOINT_EVENT_CODE(
5484 * Format identical to the LTTNG_TRACEPOINT_EVENT()
5485 * version for the following three macro parameters:
5488 TP_PROTO(int my_int, const char *my_string),
5489 TP_ARGS(my_int, my_string),
5491 /* Declarations of custom local variables */
5494 unsigned long b = 0;
5495 const char *name = "(undefined)";
5496 struct my_struct *my_struct;
5500 * Custom code which uses both tracepoint arguments
5501 * (in TP_ARGS()) and local variables (in TP_locvar()).
5503 * Local variables are actually members of a structure pointed
5504 * to by the special variable tp_locvar.
5508 tp_locvar->a = my_int + 17;
5509 tp_locvar->my_struct = get_my_struct_at(tp_locvar->a);
5510 tp_locvar->b = my_struct_compute_b(tp_locvar->my_struct);
5511 tp_locvar->name = my_struct_get_name(tp_locvar->my_struct);
5512 put_my_struct(tp_locvar->my_struct);
5521 * Format identical to the LTTNG_TRACEPOINT_EVENT()
5522 * version for this, except that tp_locvar members can be
5523 * used in the argument expression parameters of
5524 * the ctf_*() macros.
5527 ctf_integer(unsigned long, my_struct_b, tp_locvar->b)
5528 ctf_integer(int, my_struct_a, tp_locvar->a)
5529 ctf_string(my_string_field, my_string)
5530 ctf_string(my_struct_name, tp_locvar->name)
5535 IMPORTANT: The C code defined in `TP_code()` must not have any side
5536 effects when executed. In particular, the code must not allocate
5537 memory or get resources without deallocating this memory or putting
5538 those resources afterwards.
5541 [[instrumenting-linux-kernel-tracing]]
5542 ==== Load and unload a custom probe kernel module
5544 You must load a <<lttng-adaptation-layer,created LTTng-modules probe
5545 kernel module>> in the kernel before it can emit LTTng events.
5547 To load the default probe kernel modules and a custom probe kernel
5550 * Use the opt:lttng-sessiond(8):--extra-kmod-probes option to give extra
5551 probe modules to load when starting a root <<lttng-sessiond,session
5555 .Load the `my_subsys`, `usb`, and the default probe modules.
5559 # lttng-sessiond --extra-kmod-probes=my_subsys,usb
5564 You only need to pass the subsystem name, not the whole kernel module
5567 To load _only_ a given custom probe kernel module:
5569 * Use the opt:lttng-sessiond(8):--kmod-probes option to give the probe
5570 modules to load when starting a root session daemon:
5573 .Load only the `my_subsys` and `usb` probe modules.
5577 # lttng-sessiond --kmod-probes=my_subsys,usb
5582 To confirm that a probe module is loaded:
5589 $ lsmod | grep lttng_probe_usb
5593 To unload the loaded probe modules:
5595 * Kill the session daemon with `SIGTERM`:
5600 # pkill lttng-sessiond
5604 You can also use man:modprobe(8)'s `--remove` option if the session
5605 daemon terminates abnormally.
5608 [[controlling-tracing]]
5611 Once an application or a Linux kernel is
5612 <<instrumenting,instrumented>> for LTTng tracing,
5615 This section is divided in topics on how to use the various
5616 <<plumbing,components of LTTng>>, in particular the <<lttng-cli,cmd:lttng
5617 command-line tool>>, to _control_ the LTTng daemons and tracers.
5619 NOTE: In the following subsections, we refer to an man:lttng(1) command
5620 using its man page name. For example, instead of _Run the `create`
5621 command to..._, we use _Run the man:lttng-create(1) command to..._.
5625 === Start a session daemon
5627 In some situations, you need to run a <<lttng-sessiond,session daemon>>
5628 (man:lttng-sessiond(8)) _before_ you can use the man:lttng(1)
5631 You will see the following error when you run a command while no session
5635 Error: No session daemon is available
5638 The only command that automatically runs a session daemon is
5639 man:lttng-create(1), which you use to
5640 <<creating-destroying-tracing-sessions,create a tracing session>>. While
5641 this is most of the time the first operation that you do, sometimes it's
5642 not. Some examples are:
5644 * <<list-instrumentation-points,List the available instrumentation points>>.
5645 * <<saving-loading-tracing-session,Load a tracing session configuration>>.
5647 [[tracing-group]] Each Unix user must have its own running session
5648 daemon to trace user applications. The session daemon that the root user
5649 starts is the only one allowed to control the LTTng kernel tracer. Users
5650 that are part of the _tracing group_ can control the root session
5651 daemon. The default tracing group name is `tracing`; you can set it to
5652 something else with the opt:lttng-sessiond(8):--group option when you
5653 start the root session daemon.
5655 To start a user session daemon:
5657 * Run man:lttng-sessiond(8):
5662 $ lttng-sessiond --daemonize
5666 To start the root session daemon:
5668 * Run man:lttng-sessiond(8) as the root user:
5673 # lttng-sessiond --daemonize
5677 In both cases, remove the opt:lttng-sessiond(8):--daemonize option to
5678 start the session daemon in foreground.
5680 To stop a session daemon, use man:kill(1) on its process ID (standard
5683 Note that some Linux distributions could manage the LTTng session daemon
5684 as a service. In this case, you should use the service manager to
5685 start, restart, and stop session daemons.
5688 [[creating-destroying-tracing-sessions]]
5689 === Create and destroy a tracing session
5691 Almost all the LTTng control operations happen in the scope of
5692 a <<tracing-session,tracing session>>, which is the dialogue between the
5693 <<lttng-sessiond,session daemon>> and you.
5695 To create a tracing session with a generated name:
5697 * Use the man:lttng-create(1) command:
5706 The created tracing session's name is `auto` followed by the
5709 To create a tracing session with a specific name:
5711 * Use the optional argument of the man:lttng-create(1) command:
5716 $ lttng create my-session
5720 Replace `my-session` with the specific tracing session name.
5722 LTTng appends the creation date to the created tracing session's name.
5724 LTTng writes the traces of a tracing session in
5725 +$LTTNG_HOME/lttng-trace/__name__+ by default, where +__name__+ is the
5726 name of the tracing session. Note that the env:LTTNG_HOME environment
5727 variable defaults to `$HOME` if not set.
5729 To output LTTng traces to a non-default location:
5731 * Use the opt:lttng-create(1):--output option of the man:lttng-create(1) command:
5736 $ lttng create my-session --output=/tmp/some-directory
5740 You may create as many tracing sessions as you wish.
5742 To list all the existing tracing sessions for your Unix user:
5744 * Use the man:lttng-list(1) command:
5753 When you create a tracing session, it is set as the _current tracing
5754 session_. The following man:lttng(1) commands operate on the current
5755 tracing session when you don't specify one:
5757 [role="list-3-cols"]
5774 To change the current tracing session:
5776 * Use the man:lttng-set-session(1) command:
5781 $ lttng set-session new-session
5785 Replace `new-session` by the name of the new current tracing session.
5787 When you are done tracing in a given tracing session, you can destroy
5788 it. This operation frees the resources taken by the tracing session
5789 to destroy; it does not destroy the trace data that LTTng wrote for
5790 this tracing session.
5792 To destroy the current tracing session:
5794 * Use the man:lttng-destroy(1) command:
5804 [[list-instrumentation-points]]
5805 === List the available instrumentation points
5807 The <<lttng-sessiond,session daemon>> can query the running instrumented
5808 user applications and the Linux kernel to get a list of available
5809 instrumentation points. For the Linux kernel <<domain,tracing domain>>,
5810 they are tracepoints and system calls. For the user space tracing
5811 domain, they are tracepoints. For the other tracing domains, they are
5814 To list the available instrumentation points:
5816 * Use the man:lttng-list(1) command with the requested tracing domain's
5820 * opt:lttng-list(1):--kernel: Linux kernel tracepoints (your Unix user
5821 must be a root user, or it must be a member of the
5822 <<tracing-group,tracing group>>).
5823 * opt:lttng-list(1):--kernel with opt:lttng-list(1):--syscall: Linux
5824 kernel system calls (your Unix user must be a root user, or it must be
5825 a member of the tracing group).
5826 * opt:lttng-list(1):--userspace: user space tracepoints.
5827 * opt:lttng-list(1):--jul: `java.util.logging` loggers.
5828 * opt:lttng-list(1):--log4j: Apache log4j loggers.
5829 * opt:lttng-list(1):--python: Python loggers.
5832 .List the available user space tracepoints.
5836 $ lttng list --userspace
5840 .List the available Linux kernel system call tracepoints.
5844 $ lttng list --kernel --syscall
5849 [[enabling-disabling-events]]
5850 === Create and enable an event rule
5852 Once you <<creating-destroying-tracing-sessions,create a tracing
5853 session>>, you can create <<event,event rules>> with the
5854 man:lttng-enable-event(1) command.
5856 You specify each condition with a command-line option. The available
5857 condition options are shown in the following table.
5859 [role="growable",cols="asciidoc,asciidoc,default"]
5860 .Condition command-line options for the man:lttng-enable-event(1) command.
5862 |Option |Description |Applicable tracing domains
5868 . +--probe=__ADDR__+
5869 . +--function=__ADDR__+
5872 Instead of using the default _tracepoint_ instrumentation type, use:
5874 . A Linux system call.
5875 . A Linux https://lwn.net/Articles/132196/[KProbe] (symbol or address).
5876 . The entry and return points of a Linux function (symbol or address).
5880 |First positional argument.
5883 Tracepoint or system call name. In the case of a Linux KProbe or
5884 function, this is a custom name given to the event rule. With the
5885 JUL, log4j, and Python domains, this is a logger name.
5887 With a tracepoint, logger, or system call name, the last character
5888 can be `*` to match anything that remains.
5895 . +--loglevel=__LEVEL__+
5896 . +--loglevel-only=__LEVEL__+
5899 . Match only tracepoints or log statements with a logging level at
5900 least as severe as +__LEVEL__+.
5901 . Match only tracepoints or log statements with a logging level
5902 equal to +__LEVEL__+.
5904 See man:lttng-enable-event(1) for the list of available logging level
5907 |User space, JUL, log4j, and Python.
5909 |+--exclude=__EXCLUSIONS__+
5912 When you use a `*` character at the end of the tracepoint or logger
5913 name (first positional argument), exclude the specific names in the
5914 comma-delimited list +__EXCLUSIONS__+.
5917 User space, JUL, log4j, and Python.
5919 |+--filter=__EXPR__+
5922 Match only events which satisfy the expression +__EXPR__+.
5924 See man:lttng-enable-event(1) to learn more about the syntax of a
5931 You attach an event rule to a <<channel,channel>> on creation. If you do
5932 not specify the channel with the opt:lttng-enable-event(1):--channel
5933 option, and if the event rule to create is the first in its
5934 <<domain,tracing domain>> for a given tracing session, then LTTng
5935 creates a _default channel_ for you. This default channel is reused in
5936 subsequent invocations of the man:lttng-enable-event(1) command for the
5937 same tracing domain.
5939 An event rule is always enabled at creation time.
5941 The following examples show how you can combine the previous
5942 command-line options to create simple to more complex event rules.
5944 .Create an event rule targetting a Linux kernel tracepoint (default channel).
5948 $ lttng enable-event --kernel sched_switch
5952 .Create an event rule matching four Linux kernel system calls (default channel).
5956 $ lttng enable-event --kernel --syscall open,write,read,close
5960 .Create event rules matching tracepoints with filter expressions (default channel).
5964 $ lttng enable-event --kernel sched_switch --filter='prev_comm == "bash"'
5969 $ lttng enable-event --kernel --all \
5970 --filter='$ctx.tid == 1988 || $ctx.tid == 1534'
5975 $ lttng enable-event --jul my_logger \
5976 --filter='$app.retriever:cur_msg_id > 3'
5979 IMPORTANT: Make sure to always quote the filter string when you
5980 use man:lttng(1) from a shell.
5983 .Create an event rule matching any user space tracepoint of a given tracepoint provider with a log level range (default channel).
5987 $ lttng enable-event --userspace my_app:'*' --loglevel=TRACE_INFO
5990 IMPORTANT: Make sure to always quote the wildcard character when you
5991 use man:lttng(1) from a shell.
5994 .Create an event rule matching multiple Python loggers with a wildcard and with exclusions (default channel).
5998 $ lttng enable-event --python my-app.'*' \
5999 --exclude='my-app.module,my-app.hello'
6003 .Create an event rule matching any Apache log4j logger with a specific log level (default channel).
6007 $ lttng enable-event --log4j --all --loglevel-only=LOG4J_WARN
6011 .Create an event rule attached to a specific channel matching a specific user space tracepoint provider and tracepoint.
6015 $ lttng enable-event --userspace my_app:my_tracepoint --channel=my-channel
6019 The event rules of a given channel form a whitelist: as soon as an
6020 emitted event passes one of them, LTTng can record the event. For
6021 example, an event named `my_app:my_tracepoint` emitted from a user space
6022 tracepoint with a `TRACE_ERROR` log level passes both of the following
6027 $ lttng enable-event --userspace my_app:my_tracepoint
6028 $ lttng enable-event --userspace my_app:my_tracepoint \
6029 --loglevel=TRACE_INFO
6032 The second event rule is redundant: the first one includes
6036 [[disable-event-rule]]
6037 === Disable an event rule
6039 To disable an event rule that you <<enabling-disabling-events,created>>
6040 previously, use the man:lttng-disable-event(1) command. This command
6041 disables _all_ the event rules (of a given tracing domain and channel)
6042 which match an instrumentation point. The other conditions are not
6043 supported as of LTTng{nbsp}{revision}.
6045 The LTTng tracer does not record an emitted event which passes
6046 a _disabled_ event rule.
6048 .Disable an event rule matching a Python logger (default channel).
6052 $ lttng disable-event --python my-logger
6056 .Disable an event rule matching all `java.util.logging` loggers (default channel).
6060 $ lttng disable-event --jul '*'
6064 .Disable _all_ the event rules of the default channel.
6066 The opt:lttng-disable-event(1):--all-events option is not, like the
6067 opt:lttng-enable-event(1):--all option of man:lttng-enable-event(1), the
6068 equivalent of the event name `*` (wildcard): it disables _all_ the event
6069 rules of a given channel.
6073 $ lttng disable-event --jul --all-events
6077 NOTE: You cannot delete an event rule once you create it.
6081 === Get the status of a tracing session
6083 To get the status of the current tracing session, that is, its
6084 parameters, its channels, event rules, and their attributes:
6086 * Use the man:lttng-status(1) command:
6096 To get the status of any tracing session:
6098 * Use the man:lttng-list(1) command with the tracing session's name:
6103 $ lttng list my-session
6107 Replace `my-session` with the desired tracing session's name.
6110 [[basic-tracing-session-control]]
6111 === Start and stop a tracing session
6113 Once you <<creating-destroying-tracing-sessions,create a tracing
6115 <<enabling-disabling-events,create one or more event rules>>,
6116 you can start and stop the tracers for this tracing session.
6118 To start tracing in the current tracing session:
6120 * Use the man:lttng-start(1) command:
6129 LTTng is very flexible: you can launch user applications before
6130 or after the you start the tracers. The tracers only record the events
6131 if they pass enabled event rules and if they occur while the tracers are
6134 To stop tracing in the current tracing session:
6136 * Use the man:lttng-stop(1) command:
6145 If there were <<channel-overwrite-mode-vs-discard-mode,lost event
6146 records>> or lost sub-buffers since the last time you ran
6147 man:lttng-start(1), warnings are printed when you run the
6148 man:lttng-stop(1) command.
6151 [[enabling-disabling-channels]]
6152 === Create a channel
6154 Once you create a tracing session, you can create a <<channel,channel>>
6155 with the man:lttng-enable-channel(1) command.
6157 Note that LTTng automatically creates a default channel when, for a
6158 given <<domain,tracing domain>>, no channels exist and you
6159 <<enabling-disabling-events,create>> the first event rule. This default
6160 channel is named `channel0` and its attributes are set to reasonable
6161 values. Therefore, you only need to create a channel when you need
6162 non-default attributes.
6164 You specify each non-default channel attribute with a command-line
6165 option when you use the man:lttng-enable-channel(1) command. The
6166 available command-line options are:
6168 [role="growable",cols="asciidoc,asciidoc"]
6169 .Command-line options for the man:lttng-enable-channel(1) command.
6171 |Option |Description
6177 <<channel-overwrite-mode-vs-discard-mode,event loss mode>> instead of
6178 the default _discard_ mode.
6180 |`--buffers-pid` (user space tracing domain only)
6183 Use the per-process <<channel-buffering-schemes,buffering scheme>>
6184 instead of the default per-user buffering scheme.
6186 |+--subbuf-size=__SIZE__+
6189 Allocate sub-buffers of +__SIZE__+ bytes (power of two), for each CPU,
6190 either for each Unix user (default), or for each instrumented process.
6192 See <<channel-subbuf-size-vs-subbuf-count,Sub-buffer count and size>>.
6194 |+--num-subbuf=__COUNT__+
6197 Allocate +__COUNT__+ sub-buffers (power of two), for each CPU, either
6198 for each Unix user (default), or for each instrumented process.
6200 See <<channel-subbuf-size-vs-subbuf-count,Sub-buffer count and size>>.
6202 |+--tracefile-size=__SIZE__+
6205 Set the maximum size of each trace file that this channel writes within
6206 a stream to +__SIZE__+ bytes instead of no maximum.
6208 See <<tracefile-rotation,Trace file count and size>>.
6210 |+--tracefile-count=__COUNT__+
6213 Limit the number of trace files that this channel creates to
6214 +__COUNT__+ channels instead of no limit.
6216 See <<tracefile-rotation,Trace file count and size>>.
6218 |+--switch-timer=__PERIODUS__+
6221 Set the <<channel-switch-timer,switch timer period>>
6222 to +__PERIODUS__+{nbsp}µs.
6224 |+--read-timer=__PERIODUS__+
6227 Set the <<channel-read-timer,read timer period>>
6228 to +__PERIODUS__+{nbsp}µs.
6230 |+--output=__TYPE__+ (Linux kernel tracing domain only)
6233 Set the channel's output type to +__TYPE__+, either `mmap` or `splice`.
6237 You can only create a channel in the Linux kernel and user space
6238 <<domain,tracing domains>>: other tracing domains have their own channel
6239 created on the fly when <<enabling-disabling-events,creating event
6244 Because of a current LTTng limitation, you must create all channels
6245 _before_ you <<basic-tracing-session-control,start tracing>> in a given
6246 tracing session, that is, before the first time you run
6249 Since LTTng automatically creates a default channel when you use the
6250 man:lttng-enable-event(1) command with a specific tracing domain, you
6251 cannot, for example, create a Linux kernel event rule, start tracing,
6252 and then create a user space event rule, because no user space channel
6253 exists yet and it's too late to create one.
6255 For this reason, make sure to configure your channels properly
6256 before starting the tracers for the first time!
6259 The following examples show how you can combine the previous
6260 command-line options to create simple to more complex channels.
6262 .Create a Linux kernel channel with default attributes.
6266 $ lttng enable-channel --kernel my-channel
6270 .Create a user space channel with 4 sub-buffers or 1{nbsp}MiB each, per CPU, per instrumented process.
6274 $ lttng enable-channel --userspace --num-subbuf=4 --subbuf-size=1M \
6275 --buffers-pid my-channel
6279 .Create a Linux kernel channel which rotates 8 trace files of 4{nbsp}MiB each for each stream
6283 $ lttng enable-channel --kernel --tracefile-count=8 \
6284 --tracefile-size=4194304 my-channel
6288 .Create a user space channel in overwrite (or _flight recorder_) mode.
6292 $ lttng enable-channel --userspace --overwrite my-channel
6296 You can <<enabling-disabling-events,create>> the same event rule in
6297 two different channels:
6301 $ lttng enable-event --userspace --channel=my-channel app:tp
6302 $ lttng enable-event --userspace --channel=other-channel app:tp
6305 If both channels are enabled, when a tracepoint named `app:tp` is
6306 reached, LTTng records two events, one for each channel.
6310 === Disable a channel
6312 To disable a specific channel that you <<enabling-disabling-channels,created>>
6313 previously, use the man:lttng-disable-channel(1) command.
6315 .Disable a specific Linux kernel channel.
6319 $ lttng disable-channel --kernel my-channel
6323 The state of a channel precedes the individual states of event rules
6324 attached to it: event rules which belong to a disabled channel, even if
6325 they are enabled, are also considered disabled.
6329 === Add context fields to a channel
6331 Event record fields in trace files provide important information about
6332 events that occured previously, but sometimes some external context may
6333 help you solve a problem faster. Examples of context fields are:
6335 * The **process ID**, **thread ID**, **process name**, and
6336 **process priority** of the thread in which the event occurs.
6337 * The **hostname** of the system on which the event occurs.
6338 * The current values of many possible **performance counters** using
6340 ** CPU cycles, stalled cycles, idle cycles, and the other cycle types.
6342 ** Branch instructions, misses, and loads.
6344 * Any context defined at the application level (supported for the
6345 JUL and log4j <<domain,tracing domains>>).
6347 To get the full list of available context fields, see
6348 `lttng add-context --list`. Some context fields are reserved for a
6349 specific <<domain,tracing domain>> (Linux kernel or user space).
6351 You add context fields to <<channel,channels>>. All the events
6352 that a channel with added context fields records contain those fields.
6354 To add context fields to one or all the channels of a given tracing
6357 * Use the man:lttng-add-context(1) command.
6359 .Add context fields to all the channels of the current tracing session.
6361 The following command line adds the virtual process identifier and
6362 the per-thread CPU cycles count fields to all the user space channels
6363 of the current tracing session.
6367 $ lttng add-context --userspace --type=vpid --type=perf:thread:cpu-cycles
6371 .Add performance counter context fields by raw ID
6373 See man:lttng-add-context(1) for the exact format of the context field
6374 type, which is partly compatible with the format used in
6379 $ lttng add-context --userspace --type=perf:thread:raw:r0110:test
6380 $ lttng add-context --kernel --type=perf:cpu:raw:r0013c:x86unhalted
6384 .Add a context field to a specific channel.
6386 The following command line adds the thread identifier context field
6387 to the Linux kernel channel named `my-channel` in the current
6392 $ lttng add-context --kernel --channel=my-channel --type=tid
6396 .Add an application-specific context field to a specific channel.
6398 The following command line adds the `cur_msg_id` context field of the
6399 `retriever` context retriever for all the instrumented
6400 <<java-application,Java applications>> recording <<event,event records>>
6401 in the channel named `my-channel`:
6405 $ lttng add-context --kernel --channel=my-channel \
6406 --type='$app:retriever:cur_msg_id'
6409 IMPORTANT: Make sure to always quote the `$` character when you
6410 use man:lttng-add-context(1) from a shell.
6413 NOTE: You cannot remove context fields from a channel once you add it.
6418 === Track process IDs
6420 It's often useful to allow only specific process IDs (PIDs) to emit
6421 events. For example, you may wish to record all the system calls made by
6422 a given process (à la http://linux.die.net/man/1/strace[strace]).
6424 The man:lttng-track(1) and man:lttng-untrack(1) commands serve this
6425 purpose. Both commands operate on a whitelist of process IDs. You _add_
6426 entries to this whitelist with the man:lttng-track(1) command and remove
6427 entries with the man:lttng-untrack(1) command. Any process which has one
6428 of the PIDs in the whitelist is allowed to emit LTTng events which pass
6429 an enabled <<event,event rule>>.
6431 NOTE: The PID tracker tracks the _numeric process IDs_. Should a
6432 process with a given tracked ID exit and another process be given this
6433 ID, then the latter would also be allowed to emit events.
6435 .Track and untrack process IDs.
6437 For the sake of the following example, assume the target system has 16
6441 <<creating-destroying-tracing-sessions,create a tracing session>>,
6442 the whitelist contains all the possible PIDs:
6445 .All PIDs are tracked.
6446 image::track-all.png[]
6448 When the whitelist is full and you use the man:lttng-track(1) command to
6449 specify some PIDs to track, LTTng first clears the whitelist, then it
6450 tracks the specific PIDs. After:
6454 $ lttng track --pid=3,4,7,10,13
6460 .PIDs 3, 4, 7, 10, and 13 are tracked.
6461 image::track-3-4-7-10-13.png[]
6463 You can add more PIDs to the whitelist afterwards:
6467 $ lttng track --pid=1,15,16
6473 .PIDs 1, 15, and 16 are added to the whitelist.
6474 image::track-1-3-4-7-10-13-15-16.png[]
6476 The man:lttng-untrack(1) command removes entries from the PID tracker's
6477 whitelist. Given the previous example, the following command:
6481 $ lttng untrack --pid=3,7,10,13
6484 leads to this whitelist:
6487 .PIDs 3, 7, 10, and 13 are removed from the whitelist.
6488 image::track-1-4-15-16.png[]
6490 LTTng can track all possible PIDs again using the opt:track(1):--all
6495 $ lttng track --pid --all
6498 The result is, again:
6501 .All PIDs are tracked.
6502 image::track-all.png[]
6505 .Track only specific PIDs
6507 A very typical use case with PID tracking is to start with an empty
6508 whitelist, then <<basic-tracing-session-control,start the tracers>>, and
6509 then add PIDs manually while tracers are active. You can accomplish this
6510 by using the opt:lttng-untrack(1):--all option of the
6511 man:lttng-untrack(1) command to clear the whitelist after you
6512 <<creating-destroying-tracing-sessions,create a tracing session>>:
6516 $ lttng untrack --pid --all
6522 .No PIDs are tracked.
6523 image::untrack-all.png[]
6525 If you trace with this whitelist configuration, the tracer records no
6526 events for this <<domain,tracing domain>> because no processes are
6527 tracked. You can use the man:lttng-track(1) command as usual to track
6528 specific PIDs, for example:
6532 $ lttng track --pid=6,11
6538 .PIDs 6 and 11 are tracked.
6539 image::track-6-11.png[]
6544 [[saving-loading-tracing-session]]
6545 === Save and load tracing session configurations
6547 Configuring a <<tracing-session,tracing session>> can be long. Some of
6548 the tasks involved are:
6550 * <<enabling-disabling-channels,Create channels>> with
6551 specific attributes.
6552 * <<adding-context,Add context fields>> to specific channels.
6553 * <<enabling-disabling-events,Create event rules>> with specific log
6554 level and filter conditions.
6556 If you use LTTng to solve real world problems, chances are you have to
6557 record events using the same tracing session setup over and over,
6558 modifying a few variables each time in your instrumented program
6559 or environment. To avoid constant tracing session reconfiguration,
6560 the man:lttng(1) command-line tool can save and load tracing session
6561 configurations to/from XML files.
6563 To save a given tracing session configuration:
6565 * Use the man:lttng-save(1) command:
6570 $ lttng save my-session
6574 Replace `my-session` with the name of the tracing session to save.
6576 LTTng saves tracing session configurations to
6577 dir:{$LTTNG_HOME/.lttng/sessions} by default. Note that the
6578 env:LTTNG_HOME environment variable defaults to `$HOME` if not set. Use
6579 the opt:lttng-save(1):--output-path option to change this destination
6582 LTTng saves all configuration parameters, for example:
6584 * The tracing session name.
6585 * The trace data output path.
6586 * The channels with their state and all their attributes.
6587 * The context fields you added to channels.
6588 * The event rules with their state, log level and filter conditions.
6590 To load a tracing session:
6592 * Use the man:lttng-load(1) command:
6597 $ lttng load my-session
6601 Replace `my-session` with the name of the tracing session to load.
6603 When LTTng loads a configuration, it restores your saved tracing session
6604 as if you just configured it manually.
6606 See man:lttng(1) for the complete list of command-line options. You
6607 can also save and load all many sessions at a time, and decide in which
6608 directory to output the XML files.
6611 [[sending-trace-data-over-the-network]]
6612 === Send trace data over the network
6614 LTTng can send the recorded trace data to a remote system over the
6615 network instead of writing it to the local file system.
6617 To send the trace data over the network:
6619 . On the _remote_ system (which can also be the target system),
6620 start an LTTng <<lttng-relayd,relay daemon>> (man:lttng-relayd(8)):
6629 . On the _target_ system, create a tracing session configured to
6630 send trace data over the network:
6635 $ lttng create my-session --set-url=net://remote-system
6639 Replace `remote-system` by the host name or IP address of the
6640 remote system. See man:lttng-create(1) for the exact URL format.
6642 . On the target system, use the man:lttng(1) command-line tool as usual.
6643 When tracing is active, the target's consumer daemon sends sub-buffers
6644 to the relay daemon running on the remote system instead of flushing
6645 them to the local file system. The relay daemon writes the received
6646 packets to the local file system.
6648 The relay daemon writes trace files to
6649 +$LTTNG_HOME/lttng-traces/__hostname__/__session__+ by default, where
6650 +__hostname__+ is the host name of the target system and +__session__+
6651 is the tracing session name. Note that the env:LTTNG_HOME environment
6652 variable defaults to `$HOME` if not set. Use the
6653 opt:lttng-relayd(8):--output option of man:lttng-relayd(8) to write
6654 trace files to another base directory.
6659 === View events as LTTng emits them (noch:{LTTng} live)
6661 LTTng live is a network protocol implemented by the <<lttng-relayd,relay
6662 daemon>> (man:lttng-relayd(8)) to allow compatible trace viewers to
6663 display events as LTTng emits them on the target system while tracing is
6666 The relay daemon creates a _tee_: it forwards the trace data to both
6667 the local file system and to connected live viewers:
6670 .The relay daemon creates a _tee_, forwarding the trace data to both trace files and a connected live viewer.
6675 . On the _target system_, create a <<tracing-session,tracing session>>
6681 $ lttng create my-session --live
6685 This spawns a local relay daemon.
6687 . Start the live viewer and configure it to connect to the relay
6688 daemon. For example, with http://diamon.org/babeltrace[Babeltrace]:
6693 $ babeltrace --input-format=lttng-live \
6694 net://localhost/host/hostname/my-session
6701 * `hostname` with the host name of the target system.
6702 * `my-session` with the name of the tracing session to view.
6705 . Configure the tracing session as usual with the man:lttng(1)
6706 command-line tool, and <<basic-tracing-session-control,start tracing>>.
6708 You can list the available live tracing sessions with Babeltrace:
6712 $ babeltrace --input-format=lttng-live net://localhost
6715 You can start the relay daemon on another system. In this case, you need
6716 to specify the relay daemon's URL when you create the tracing session
6717 with the opt:lttng-create(1):--set-url option. You also need to replace
6718 `localhost` in the procedure above with the host name of the system on
6719 which the relay daemon is running.
6721 See man:lttng-create(1) and man:lttng-relayd(8) for the complete list of
6722 command-line options.
6726 [[taking-a-snapshot]]
6727 === Take a snapshot of the current sub-buffers of a tracing session
6729 The normal behavior of LTTng is to append full sub-buffers to growing
6730 trace data files. This is ideal to keep a full history of the events
6731 that occurred on the target system, but it can
6732 represent too much data in some situations. For example, you may wish
6733 to trace your application continuously until some critical situation
6734 happens, in which case you only need the latest few recorded
6735 events to perform the desired analysis, not multi-gigabyte trace files.
6737 With the man:lttng-snapshot(1) command, you can take a snapshot of the
6738 current sub-buffers of a given <<tracing-session,tracing session>>.
6739 LTTng can write the snapshot to the local file system or send it over
6744 . Create a tracing session in _snapshot mode_:
6749 $ lttng create my-session --snapshot
6753 The <<channel-overwrite-mode-vs-discard-mode,event loss mode>> of
6754 <<channel,channels>> created in this mode is automatically set to
6755 _overwrite_ (flight recorder mode).
6757 . Configure the tracing session as usual with the man:lttng(1)
6758 command-line tool, and <<basic-tracing-session-control,start tracing>>.
6760 . **Optional**: When you need to take a snapshot,
6761 <<basic-tracing-session-control,stop tracing>>.
6763 You can take a snapshot when the tracers are active, but if you stop
6764 them first, you are sure that the data in the sub-buffers does not
6765 change before you actually take the snapshot.
6772 $ lttng snapshot record --name=my-first-snapshot
6776 LTTng writes the current sub-buffers of all the current tracing
6777 session's channels to trace files on the local file system. Those trace
6778 files have `my-first-snapshot` in their name.
6780 There is no difference between the format of a normal trace file and the
6781 format of a snapshot: viewers of LTTng traces also support LTTng
6784 By default, LTTng writes snapshot files to the path shown by
6785 `lttng snapshot list-output`. You can change this path or decide to send
6786 snapshots over the network using either:
6788 . An output path or URL that you specify when you create the
6790 . An snapshot output path or URL that you add using
6791 `lttng snapshot add-output`
6792 . An output path or URL that you provide directly to the
6793 `lttng snapshot record` command.
6795 Method 3 overrides method 2, which overrides method 1. When you
6796 specify a URL, a relay daemon must listen on a remote system (see
6797 <<sending-trace-data-over-the-network,Send trace data over the network>>).
6802 === Use the machine interface
6804 With any command of the man:lttng(1) command-line tool, you can set the
6805 opt:lttng(1):--mi option to `xml` (before the command name) to get an
6806 XML machine interface output, for example:
6810 $ lttng --mi=xml enable-event --kernel --syscall open
6813 A schema definition (XSD) is
6814 https://github.com/lttng/lttng-tools/blob/stable-2.9/src/common/mi-lttng-3.0.xsd[available]
6815 to ease the integration with external tools as much as possible.
6819 [[metadata-regenerate]]
6820 === Regenerate the metadata of an LTTng trace
6822 An LTTng trace, which is a http://diamon.org/ctf[CTF] trace, has both
6823 data stream files and a metadata file. This metadata file contains,
6824 amongst other things, information about the offset of the clock sources
6825 used to timestamp <<event,event records>> when tracing.
6827 If, once a <<tracing-session,tracing session>> is
6828 <<basic-tracing-session-control,started>>, a major
6829 https://en.wikipedia.org/wiki/Network_Time_Protocol[NTP] correction
6830 happens, the trace's clock offset also needs to be updated. You
6831 can use the `metadata` item of the man:lttng-regenerate(1) command
6834 The main use case of this command is to allow a system to boot with
6835 an incorrect wall time and trace it with LTTng before its wall time
6836 is corrected. Once the system is known to be in a state where its
6837 wall time is correct, it can run `lttng regenerate metadata`.
6839 To regenerate the metadata of an LTTng trace:
6841 * Use the `metadata` item of the man:lttng-regenerate(1) command:
6846 $ lttng regenerate metadata
6852 `lttng regenerate metadata` has the following limitations:
6854 * Tracing session <<creating-destroying-tracing-sessions,created>>
6856 * User space <<channel,channels>>, if any, are using
6857 <<channel-buffering-schemes,per-user buffering>>.
6862 [[regenerate-statedump]]
6863 === Regenerate the state dump of a tracing session
6865 The LTTng kernel and user space tracers generate state dump
6866 <<event,event records>> when the application starts or when you
6867 <<basic-tracing-session-control,start a tracing session>>. An analysis
6868 can use the state dump event records to set an initial state before it
6869 builds the rest of the state from the following event records.
6870 http://tracecompass.org/[Trace Compass] is a notable example of an
6871 application which uses the state dump of an LTTng trace.
6873 When you <<taking-a-snapshot,take a snapshot>>, it's possible that the
6874 state dump event records are not included in the snapshot because they
6875 were recorded to a sub-buffer that has been consumed or overwritten
6878 You can use the `lttng regenerate statedump` command to emit the state
6879 dump event records again.
6881 To regenerate the state dump of the current tracing session, provided
6882 create it in snapshot mode, before you take a snapshot:
6884 . Use the `statedump` item of the man:lttng-regenerate(1) command:
6889 $ lttng regenerate statedump
6893 . <<basic-tracing-session-control,Stop the tracing session>>:
6902 . <<taking-a-snapshot,Take a snapshot>>:
6907 $ lttng snapshot record --name=my-snapshot
6911 Depending on the event throughput, you should run steps 1 and 2
6912 as closely as possible.
6914 NOTE: To record the state dump events, you need to
6915 <<enabling-disabling-events,create event rules>> which enable them.
6916 LTTng-UST state dump tracepoints start with `lttng_ust_statedump:`.
6917 LTTng-modules state dump tracepoints start with `lttng_statedump_`.
6921 [[persistent-memory-file-systems]]
6922 === Record trace data on persistent memory file systems
6924 https://en.wikipedia.org/wiki/Non-volatile_random-access_memory[Non-volatile random-access memory]
6925 (NVRAM) is random-access memory that retains its information when power
6926 is turned off (non-volatile). Systems with such memory can store data
6927 structures in RAM and retrieve them after a reboot, without flushing
6928 to typical _storage_.
6930 Linux supports NVRAM file systems thanks to either
6931 http://pramfs.sourceforge.net/[PRAMFS] or
6932 https://www.kernel.org/doc/Documentation/filesystems/dax.txt[DAX]{nbsp}+{nbsp}http://lkml.iu.edu/hypermail/linux/kernel/1504.1/03463.html[pmem]
6933 (requires Linux 4.1+).
6935 This section does not describe how to operate such file systems;
6936 we assume that you have a working persistent memory file system.
6938 When you create a <<tracing-session,tracing session>>, you can specify
6939 the path of the shared memory holding the sub-buffers. If you specify a
6940 location on an NVRAM file system, then you can retrieve the latest
6941 recorded trace data when the system reboots after a crash.
6943 To record trace data on a persistent memory file system and retrieve the
6944 trace data after a system crash:
6946 . Create a tracing session with a sub-buffer shared memory path located
6947 on an NVRAM file system:
6952 $ lttng create my-session --shm-path=/path/to/shm
6956 . Configure the tracing session as usual with the man:lttng(1)
6957 command-line tool, and <<basic-tracing-session-control,start tracing>>.
6959 . After a system crash, use the man:lttng-crash(1) command-line tool to
6960 view the trace data recorded on the NVRAM file system:
6965 $ lttng-crash /path/to/shm
6969 The binary layout of the ring buffer files is not exactly the same as
6970 the trace files layout. This is why you need to use man:lttng-crash(1)
6971 instead of your preferred trace viewer directly.
6973 To convert the ring buffer files to LTTng trace files:
6975 * Use the opt:lttng-crash(1):--extract option of man:lttng-crash(1):
6980 $ lttng-crash --extract=/path/to/trace /path/to/shm
6988 [[lttng-modules-ref]]
6989 === noch:{LTTng-modules}
6993 [[lttng-tracepoint-enum]]
6994 ==== `LTTNG_TRACEPOINT_ENUM()` usage
6996 Use the `LTTNG_TRACEPOINT_ENUM()` macro to define an enumeration:
7000 LTTNG_TRACEPOINT_ENUM(name, TP_ENUM_VALUES(entries))
7005 * `name` with the name of the enumeration (C identifier, unique
7006 amongst all the defined enumerations).
7007 * `entries` with a list of enumeration entries.
7009 The available enumeration entry macros are:
7011 +ctf_enum_value(__name__, __value__)+::
7012 Entry named +__name__+ mapped to the integral value +__value__+.
7014 +ctf_enum_range(__name__, __begin__, __end__)+::
7015 Entry named +__name__+ mapped to the range of integral values between
7016 +__begin__+ (included) and +__end__+ (included).
7018 +ctf_enum_auto(__name__)+::
7019 Entry named +__name__+ mapped to the integral value following the
7020 last mapping's value.
7022 The last value of a `ctf_enum_value()` entry is its +__value__+
7025 The last value of a `ctf_enum_range()` entry is its +__end__+ parameter.
7027 If `ctf_enum_auto()` is the first entry in the list, its integral
7030 Use the `ctf_enum()` <<lttng-modules-tp-fields,field definition macro>>
7031 to use a defined enumeration as a tracepoint field.
7033 .Define an enumeration with `LTTNG_TRACEPOINT_ENUM()`.
7037 LTTNG_TRACEPOINT_ENUM(
7040 ctf_enum_auto("AUTO: EXPECT 0")
7041 ctf_enum_value("VALUE: 23", 23)
7042 ctf_enum_value("VALUE: 27", 27)
7043 ctf_enum_auto("AUTO: EXPECT 28")
7044 ctf_enum_range("RANGE: 101 TO 303", 101, 303)
7045 ctf_enum_auto("AUTO: EXPECT 304")
7053 [[lttng-modules-tp-fields]]
7054 ==== Tracepoint fields macros (for `TP_FIELDS()`)
7056 [[tp-fast-assign]][[tp-struct-entry]]The available macros to define
7057 tracepoint fields, which must be listed within `TP_FIELDS()` in
7058 `LTTNG_TRACEPOINT_EVENT()`, are:
7060 [role="func-desc growable",cols="asciidoc,asciidoc"]
7061 .Available macros to define LTTng-modules tracepoint fields
7063 |Macro |Description and parameters
7066 +ctf_integer(__t__, __n__, __e__)+
7068 +ctf_integer_nowrite(__t__, __n__, __e__)+
7070 +ctf_user_integer(__t__, __n__, __e__)+
7072 +ctf_user_integer_nowrite(__t__, __n__, __e__)+
7074 Standard integer, displayed in base 10.
7077 Integer C type (`int`, `long`, `size_t`, ...).
7083 Argument expression.
7086 +ctf_integer_hex(__t__, __n__, __e__)+
7088 +ctf_user_integer_hex(__t__, __n__, __e__)+
7090 Standard integer, displayed in base 16.
7099 Argument expression.
7101 |+ctf_integer_oct(__t__, __n__, __e__)+
7103 Standard integer, displayed in base 8.
7112 Argument expression.
7115 +ctf_integer_network(__t__, __n__, __e__)+
7117 +ctf_user_integer_network(__t__, __n__, __e__)+
7119 Integer in network byte order (big-endian), displayed in base 10.
7128 Argument expression.
7131 +ctf_integer_network_hex(__t__, __n__, __e__)+
7133 +ctf_user_integer_network_hex(__t__, __n__, __e__)+
7135 Integer in network byte order, displayed in base 16.
7144 Argument expression.
7147 +ctf_enum(__N__, __t__, __n__, __e__)+
7149 +ctf_enum_nowrite(__N__, __t__, __n__, __e__)+
7151 +ctf_user_enum(__N__, __t__, __n__, __e__)+
7153 +ctf_user_enum_nowrite(__N__, __t__, __n__, __e__)+
7158 Name of a <<lttng-tracepoint-enum,previously defined enumeration>>.
7161 Integer C type (`int`, `long`, `size_t`, ...).
7167 Argument expression.
7170 +ctf_string(__n__, __e__)+
7172 +ctf_string_nowrite(__n__, __e__)+
7174 +ctf_user_string(__n__, __e__)+
7176 +ctf_user_string_nowrite(__n__, __e__)+
7178 Null-terminated string; undefined behavior if +__e__+ is `NULL`.
7184 Argument expression.
7187 +ctf_array(__t__, __n__, __e__, __s__)+
7189 +ctf_array_nowrite(__t__, __n__, __e__, __s__)+
7191 +ctf_user_array(__t__, __n__, __e__, __s__)+
7193 +ctf_user_array_nowrite(__t__, __n__, __e__, __s__)+
7195 Statically-sized array of integers.
7198 Array element C type.
7204 Argument expression.
7210 +ctf_array_bitfield(__t__, __n__, __e__, __s__)+
7212 +ctf_array_bitfield_nowrite(__t__, __n__, __e__, __s__)+
7214 +ctf_user_array_bitfield(__t__, __n__, __e__, __s__)+
7216 +ctf_user_array_bitfield_nowrite(__t__, __n__, __e__, __s__)+
7218 Statically-sized array of bits.
7220 The type of +__e__+ must be an integer type. +__s__+ is the number
7221 of elements of such type in +__e__+, not the number of bits.
7224 Array element C type.
7230 Argument expression.
7236 +ctf_array_text(__t__, __n__, __e__, __s__)+
7238 +ctf_array_text_nowrite(__t__, __n__, __e__, __s__)+
7240 +ctf_user_array_text(__t__, __n__, __e__, __s__)+
7242 +ctf_user_array_text_nowrite(__t__, __n__, __e__, __s__)+
7244 Statically-sized array, printed as text.
7246 The string does not need to be null-terminated.
7249 Array element C type (always `char`).
7255 Argument expression.
7261 +ctf_sequence(__t__, __n__, __e__, __T__, __E__)+
7263 +ctf_sequence_nowrite(__t__, __n__, __e__, __T__, __E__)+
7265 +ctf_user_sequence(__t__, __n__, __e__, __T__, __E__)+
7267 +ctf_user_sequence_nowrite(__t__, __n__, __e__, __T__, __E__)+
7269 Dynamically-sized array of integers.
7271 The type of +__E__+ must be unsigned.
7274 Array element C type.
7280 Argument expression.
7283 Length expression C type.
7289 +ctf_sequence_hex(__t__, __n__, __e__, __T__, __E__)+
7291 +ctf_user_sequence_hex(__t__, __n__, __e__, __T__, __E__)+
7293 Dynamically-sized array of integers, displayed in base 16.
7295 The type of +__E__+ must be unsigned.
7298 Array element C type.
7304 Argument expression.
7307 Length expression C type.
7312 |+ctf_sequence_network(__t__, __n__, __e__, __T__, __E__)+
7314 Dynamically-sized array of integers in network byte order (big-endian),
7315 displayed in base 10.
7317 The type of +__E__+ must be unsigned.
7320 Array element C type.
7326 Argument expression.
7329 Length expression C type.
7335 +ctf_sequence_bitfield(__t__, __n__, __e__, __T__, __E__)+
7337 +ctf_sequence_bitfield_nowrite(__t__, __n__, __e__, __T__, __E__)+
7339 +ctf_user_sequence_bitfield(__t__, __n__, __e__, __T__, __E__)+
7341 +ctf_user_sequence_bitfield_nowrite(__t__, __n__, __e__, __T__, __E__)+
7343 Dynamically-sized array of bits.
7345 The type of +__e__+ must be an integer type. +__s__+ is the number
7346 of elements of such type in +__e__+, not the number of bits.
7348 The type of +__E__+ must be unsigned.
7351 Array element C type.
7357 Argument expression.
7360 Length expression C type.
7366 +ctf_sequence_text(__t__, __n__, __e__, __T__, __E__)+
7368 +ctf_sequence_text_nowrite(__t__, __n__, __e__, __T__, __E__)+
7370 +ctf_user_sequence_text(__t__, __n__, __e__, __T__, __E__)+
7372 +ctf_user_sequence_text_nowrite(__t__, __n__, __e__, __T__, __E__)+
7374 Dynamically-sized array, displayed as text.
7376 The string does not need to be null-terminated.
7378 The type of +__E__+ must be unsigned.
7380 The behaviour is undefined if +__e__+ is `NULL`.
7383 Sequence element C type (always `char`).
7389 Argument expression.
7392 Length expression C type.
7398 Use the `_user` versions when the argument expression, `e`, is
7399 a user space address. In the cases of `ctf_user_integer*()` and
7400 `ctf_user_float*()`, `&e` must be a user space address, thus `e` must
7403 The `_nowrite` versions omit themselves from the session trace, but are
7404 otherwise identical. This means the `_nowrite` fields won't be written
7405 in the recorded trace. Their primary purpose is to make some
7406 of the event context available to the
7407 <<enabling-disabling-events,event filters>> without having to
7408 commit the data to sub-buffers.
7414 Terms related to LTTng and to tracing in general:
7417 The http://diamon.org/babeltrace[Babeltrace] project, which includes
7418 the cmd:babeltrace command, some libraries, and Python bindings.
7420 <<channel-buffering-schemes,buffering scheme>>::
7421 A layout of sub-buffers applied to a given channel.
7423 <<channel,channel>>::
7424 An entity which is responsible for a set of ring buffers.
7426 <<event,Event rules>> are always attached to a specific channel.
7429 A reference of time for a tracer.
7431 <<lttng-consumerd,consumer daemon>>::
7432 A process which is responsible for consuming the full sub-buffers
7433 and write them to a file system or send them over the network.
7435 <<channel-overwrite-mode-vs-discard-mode,discard mode>>:: The event loss
7436 mode in which the tracer _discards_ new event records when there's no
7437 sub-buffer space left to store them.
7440 The consequence of the execution of an instrumentation
7441 point, like a tracepoint that you manually place in some source code,
7442 or a Linux kernel KProbe.
7444 An event is said to _occur_ at a specific time. Different actions can
7445 be taken upon the occurrence of an event, like record the event's payload
7448 <<channel-overwrite-mode-vs-discard-mode,event loss mode>>::
7449 The mechanism by which event records of a given channel are lost
7450 (not recorded) when there is no sub-buffer space left to store them.
7452 [[def-event-name]]event name::
7453 The name of an event, which is also the name of the event record.
7454 This is also called the _instrumentation point name_.
7457 A record, in a trace, of the payload of an event which occured.
7459 <<event,event rule>>::
7460 Set of conditions which must be satisfied for one or more occuring
7461 events to be recorded.
7463 `java.util.logging`::
7465 https://docs.oracle.com/javase/7/docs/api/java/util/logging/package-summary.html[core logging facilities].
7467 <<instrumenting,instrumentation>>::
7468 The use of LTTng probes to make a piece of software traceable.
7470 instrumentation point::
7471 A point in the execution path of a piece of software that, when
7472 reached by this execution, can emit an event.
7474 instrumentation point name::
7475 See _<<def-event-name,event name>>_.
7478 A http://logging.apache.org/log4j/1.2/[logging library] for Java
7479 developed by the Apache Software Foundation.
7482 Level of severity of a log statement or user space
7483 instrumentation point.
7486 The _Linux Trace Toolkit: next generation_ project.
7488 <<lttng-cli,cmd:lttng>>::
7489 A command-line tool provided by the LTTng-tools project which you
7490 can use to send and receive control messages to and from a
7494 The https://github.com/lttng/lttng-analyses[LTTng analyses] project,
7495 which is a set of analyzing programs that are used to obtain a
7496 higher level view of an LTTng trace.
7498 cmd:lttng-consumerd::
7499 The name of the consumer daemon program.
7502 A utility provided by the LTTng-tools project which can convert
7503 ring buffer files (usually
7504 <<persistent-memory-file-systems,saved on a persistent memory file system>>)
7507 LTTng Documentation::
7510 <<lttng-live,LTTng live>>::
7511 A communication protocol between the relay daemon and live viewers
7512 which makes it possible to see events "live", as they are received by
7515 <<lttng-modules,LTTng-modules>>::
7516 The https://github.com/lttng/lttng-modules[LTTng-modules] project,
7517 which contains the Linux kernel modules to make the Linux kernel
7518 instrumentation points available for LTTng tracing.
7521 The name of the relay daemon program.
7523 cmd:lttng-sessiond::
7524 The name of the session daemon program.
7527 The https://github.com/lttng/lttng-tools[LTTng-tools] project, which
7528 contains the various programs and libraries used to
7529 <<controlling-tracing,control tracing>>.
7531 <<lttng-ust,LTTng-UST>>::
7532 The https://github.com/lttng/lttng-ust[LTTng-UST] project, which
7533 contains libraries to instrument user applications.
7535 <<lttng-ust-agents,LTTng-UST Java agent>>::
7536 A Java package provided by the LTTng-UST project to allow the
7537 LTTng instrumentation of `java.util.logging` and Apache log4j 1.2
7540 <<lttng-ust-agents,LTTng-UST Python agent>>::
7541 A Python package provided by the LTTng-UST project to allow the
7542 LTTng instrumentation of Python logging statements.
7544 <<channel-overwrite-mode-vs-discard-mode,overwrite mode>>::
7545 The event loss mode in which new event records overwrite older
7546 event records when there's no sub-buffer space left to store them.
7548 <<channel-buffering-schemes,per-process buffering>>::
7549 A buffering scheme in which each instrumented process has its own
7550 sub-buffers for a given user space channel.
7552 <<channel-buffering-schemes,per-user buffering>>::
7553 A buffering scheme in which all the processes of a Unix user share the
7554 same sub-buffer for a given user space channel.
7556 <<lttng-relayd,relay daemon>>::
7557 A process which is responsible for receiving the trace data sent by
7558 a distant consumer daemon.
7561 A set of sub-buffers.
7563 <<lttng-sessiond,session daemon>>::
7564 A process which receives control commands from you and orchestrates
7565 the tracers and various LTTng daemons.
7567 <<taking-a-snapshot,snapshot>>::
7568 A copy of the current data of all the sub-buffers of a given tracing
7569 session, saved as trace files.
7572 One part of an LTTng ring buffer which contains event records.
7575 The time information attached to an event when it is emitted.
7578 A set of files which are the concatenations of one or more
7579 flushed sub-buffers.
7582 The action of recording the events emitted by an application
7583 or by a system, or to initiate such recording by controlling
7587 The http://tracecompass.org[Trace Compass] project and application.
7590 An instrumentation point using the tracepoint mechanism of the Linux
7591 kernel or of LTTng-UST.
7593 tracepoint definition::
7594 The definition of a single tracepoint.
7597 The name of a tracepoint.
7599 tracepoint provider::
7600 A set of functions providing tracepoints to an instrumented user
7603 Not to be confused with a _tracepoint provider package_: many tracepoint
7604 providers can exist within a tracepoint provider package.
7606 tracepoint provider package::
7607 One or more tracepoint providers compiled as an object file or as
7611 A software which records emitted events.
7613 <<domain,tracing domain>>::
7614 A namespace for event sources.
7616 <<tracing-group,tracing group>>::
7617 The Unix group in which a Unix user can be to be allowed to trace the
7620 <<tracing-session,tracing session>>::
7621 A stateful dialogue between you and a <<lttng-sessiond,session
7625 An application running in user space, as opposed to a Linux kernel
7626 module, for example.