1 The LTTng Documentation
2 =======================
3 Philippe Proulx <pproulx@efficios.com>
4 v2.10, 12 November 2018
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 _KeKriek_. From
78 http://brasseriedunham.com/[Brasserie Dunham], the _**KeKriek**_ is a
79 sour mashed golden wheat ale fermented with local sour cherries from
80 Tougas orchards. Fresh sweet cherry notes with some tartness, lively
81 carbonation with a dry finish.
83 New features and changes in LTTng{nbsp}{revision}:
85 * **Tracing control**:
86 ** You can put more than one wildcard special character (`*`), and not
87 only at the end, when you <<enabling-disabling-events,create an event
88 rule>>, in both the instrumentation point name and the literal
90 link:/man/1/lttng-enable-event/v{revision}/#doc-filter-syntax[filter expressions]:
95 # lttng enable-event --kernel 'x86_*_local_timer_*' \
96 --filter='name == "*a*b*c*d*e" && count >= 23'
103 $ lttng enable-event --userspace '*_my_org:*msg*'
107 ** New trigger and notification API for
108 <<liblttng-ctl-lttng,`liblttng-ctl`>>. This new subsystem allows you
109 to register triggers which emit a notification when a given
110 condition is satisfied. As of LTTng{nbsp}{revision}, only
111 <<channel,channel>> buffer usage conditions are available.
112 Documentation is available in the
113 https://github.com/lttng/lttng-tools/tree/stable-{revision}/include/lttng[`liblttng-ctl`
115 <<notif-trigger-api,Get notified when a channel's buffer usage is too
118 ** You can now embed the whole textual LTTng-tools man pages into the
119 executables at build time with the `--enable-embedded-help`
120 configuration option. Thanks to this option, you don't need the
121 http://www.methods.co.nz/asciidoc/[AsciiDoc] and
122 https://directory.fsf.org/wiki/Xmlto[xmlto] tools at build time, and
123 a manual pager at run time, to get access to this documentation.
125 * **User space tracing**:
126 ** New blocking mode: an LTTng-UST tracepoint can now block until
127 <<channel,sub-buffer>> space is available instead of discarding event
128 records in <<channel-overwrite-mode-vs-discard-mode,discard mode>>.
129 With this feature, you can be sure that no event records are
130 discarded during your application's execution at the expense of
133 For example, the following command lines create a user space tracing
134 channel with an infinite blocking timeout and run an application
135 instrumented with LTTng-UST which is explicitly allowed to block:
141 $ lttng enable-channel --userspace --blocking-timeout=inf blocking-channel
142 $ lttng enable-event --userspace --channel=blocking-channel --all
144 $ LTTNG_UST_ALLOW_BLOCKING=1 my-app
148 See the complete <<blocking-timeout-example,blocking timeout example>>.
150 * **Linux kernel tracing**:
151 ** Linux 4.10, 4.11, and 4.12 support.
152 ** The thread state dump events recorded by LTTng-modules now contain
153 the task's CPU identifier. This improves the precision of the
154 scheduler model for analyses.
155 ** Extended man:socketpair(2) system call tracing data.
161 What is LTTng? As its name suggests, the _Linux Trace Toolkit: next
162 generation_ is a modern toolkit for tracing Linux systems and
163 applications. So your first question might be:
170 As the history of software engineering progressed and led to what
171 we now take for granted--complex, numerous and
172 interdependent software applications running in parallel on
173 sophisticated operating systems like Linux--the authors of such
174 components, software developers, began feeling a natural
175 urge to have tools that would ensure the robustness and good performance
176 of their masterpieces.
178 One major achievement in this field is, inarguably, the
179 https://www.gnu.org/software/gdb/[GNU debugger (GDB)],
180 an essential tool for developers to find and fix bugs. But even the best
181 debugger won't help make your software run faster, and nowadays, faster
182 software means either more work done by the same hardware, or cheaper
183 hardware for the same work.
185 A _profiler_ is often the tool of choice to identify performance
186 bottlenecks. Profiling is suitable to identify _where_ performance is
187 lost in a given software. The profiler outputs a profile, a statistical
188 summary of observed events, which you may use to discover which
189 functions took the most time to execute. However, a profiler won't
190 report _why_ some identified functions are the bottleneck. Bottlenecks
191 might only occur when specific conditions are met, conditions that are
192 sometimes impossible to capture by a statistical profiler, or impossible
193 to reproduce with an application altered by the overhead of an
194 event-based profiler. For a thorough investigation of software
195 performance issues, a history of execution is essential, with the
196 recorded values of variables and context fields you choose, and
197 with as little influence as possible on the instrumented software. This
198 is where tracing comes in handy.
200 _Tracing_ is a technique used to understand what goes on in a running
201 software system. The software used for tracing is called a _tracer_,
202 which is conceptually similar to a tape recorder. When recording,
203 specific instrumentation points placed in the software source code
204 generate events that are saved on a giant tape: a _trace_ file. You
205 can trace user applications and the operating system at the same time,
206 opening the possibility of resolving a wide range of problems that would
207 otherwise be extremely challenging.
209 Tracing is often compared to _logging_. However, tracers and loggers are
210 two different tools, serving two different purposes. Tracers are
211 designed to record much lower-level events that occur much more
212 frequently than log messages, often in the range of thousands per
213 second, with very little execution overhead. Logging is more appropriate
214 for a very high-level analysis of less frequent events: user accesses,
215 exceptional conditions (errors and warnings, for example), database
216 transactions, instant messaging communications, and such. Simply put,
217 logging is one of the many use cases that can be satisfied with tracing.
219 The list of recorded events inside a trace file can be read manually
220 like a log file for the maximum level of detail, but it is generally
221 much more interesting to perform application-specific analyses to
222 produce reduced statistics and graphs that are useful to resolve a
223 given problem. Trace viewers and analyzers are specialized tools
226 In the end, this is what LTTng is: a powerful, open source set of
227 tools to trace the Linux kernel and user applications at the same time.
228 LTTng is composed of several components actively maintained and
229 developed by its link:/community/#where[community].
232 [[lttng-alternatives]]
233 === Alternatives to noch:{LTTng}
235 Excluding proprietary solutions, a few competing software tracers
238 * https://github.com/dtrace4linux/linux[dtrace4linux] is a port of
239 Sun Microsystems's DTrace to Linux. The cmd:dtrace tool interprets
240 user scripts and is responsible for loading code into the
241 Linux kernel for further execution and collecting the outputted data.
242 * https://en.wikipedia.org/wiki/Berkeley_Packet_Filter[eBPF] is a
243 subsystem in the Linux kernel in which a virtual machine can execute
244 programs passed from the user space to the kernel. You can attach
245 such programs to tracepoints and KProbes thanks to a system call, and
246 they can output data to the user space when executed thanks to
247 different mechanisms (pipe, VM register values, and eBPF maps, to name
249 * https://www.kernel.org/doc/Documentation/trace/ftrace.txt[ftrace]
250 is the de facto function tracer of the Linux kernel. Its user
251 interface is a set of special files in sysfs.
252 * https://perf.wiki.kernel.org/[perf] is
253 a performance analyzing tool for Linux which supports hardware
254 performance counters, tracepoints, as well as other counters and
255 types of probes. perf's controlling utility is the cmd:perf command
257 * http://linux.die.net/man/1/strace[strace]
258 is a command-line utility which records system calls made by a
259 user process, as well as signal deliveries and changes of process
260 state. strace makes use of https://en.wikipedia.org/wiki/Ptrace[ptrace]
261 to fulfill its function.
262 * http://www.sysdig.org/[sysdig], like SystemTap, uses scripts to
263 analyze Linux kernel events. You write scripts, or _chisels_ in
264 sysdig's jargon, in Lua and sysdig executes them while the system is
265 being traced or afterwards. sysdig's interface is the cmd:sysdig
266 command-line tool as well as the curses-based cmd:csysdig tool.
267 * https://sourceware.org/systemtap/[SystemTap] is a Linux kernel and
268 user space tracer which uses custom user scripts to produce plain text
269 traces. SystemTap converts the scripts to the C language, and then
270 compiles them as Linux kernel modules which are loaded to produce
271 trace data. SystemTap's primary user interface is the cmd:stap
274 The main distinctive features of LTTng is that it produces correlated
275 kernel and user space traces, as well as doing so with the lowest
276 overhead amongst other solutions. It produces trace files in the
277 http://diamon.org/ctf[CTF] format, a file format optimized
278 for the production and analyses of multi-gigabyte data.
280 LTTng is the result of more than 10 years of active open source
281 development by a community of passionate developers.
282 LTTng{nbsp}{revision} is currently available on major desktop and server
285 The main interface for tracing control is a single command-line tool
286 named cmd:lttng. The latter can create several tracing sessions, enable
287 and disable events on the fly, filter events efficiently with custom
288 user expressions, start and stop tracing, and much more. LTTng can
289 record the traces on the file system or send them over the network, and
290 keep them totally or partially. You can view the traces once tracing
291 becomes inactive or in real-time.
293 <<installing-lttng,Install LTTng now>> and
294 <<getting-started,start tracing>>!
300 **LTTng** is a set of software <<plumbing,components>> which interact to
301 <<instrumenting,instrument>> the Linux kernel and user applications, and
302 to <<controlling-tracing,control tracing>> (start and stop
303 tracing, enable and disable event rules, and the rest). Those
304 components are bundled into the following packages:
306 * **LTTng-tools**: Libraries and command-line interface to
308 * **LTTng-modules**: Linux kernel modules to instrument and
310 * **LTTng-UST**: Libraries and Java/Python packages to instrument and
311 trace user applications.
313 Most distributions mark the LTTng-modules and LTTng-UST packages as
314 optional when installing LTTng-tools (which is always required). In the
315 following sections, we always provide the steps to install all three,
318 * You only need to install LTTng-modules if you intend to trace the
320 * You only need to install LTTng-UST if you intend to trace user
324 .Availability of LTTng{nbsp}{revision} for major Linux distributions as of 12 November 2018.
326 |Distribution |Available in releases |Alternatives
328 |https://www.ubuntu.com/[Ubuntu]
329 |<<ubuntu,Ubuntu{nbsp}18.04 _Bionic Beaver_>>.
331 Ubuntu{nbsp}14.04 _Trusty Tahr_, Ubuntu{nbsp}16.04 _Xenial Xerus_,
332 and Ubuntu{nbsp}18.04 _Bionic Beaver_:
333 <<ubuntu-ppa,use the LTTng Stable{nbsp}{revision} PPA>>.
334 |<<building-from-source,Build LTTng{nbsp}{revision} from source>>.
336 |https://getfedora.org/[Fedora]
337 |xref:fedora[Fedora{nbsp}27, Fedora{nbsp}28, and Fedora{nbsp}29].
338 |<<building-from-source,Build LTTng{nbsp}{revision} from source>>.
340 |https://www.debian.org/[Debian]
341 |<<debian,Debian "buster" (testing)>>.
342 |<<building-from-source,Build LTTng{nbsp}{revision} from source>>.
344 |https://www.archlinux.org/[Arch Linux]
345 |<<arch-linux,Current Arch Linux build>>.
346 |<<building-from-source,Build LTTng{nbsp}{revision} from source>>.
348 |https://alpinelinux.org/[Alpine Linux]
349 |<<alpine-linux,Alpine Linux{nbsp}3.7 and Alpine Linux{nbsp}3.8>>.
350 |<<building-from-source,Build LTTng{nbsp}{revision} from source>>.
352 |https://www.opensuse.org/[openSUSE]
353 |<<opensuse,openSUSE Leap{nbsp}15.0>>.
354 |<<building-from-source,Build LTTng{nbsp}{revision} from source>>.
356 |https://www.redhat.com/[RHEL] and https://www.suse.com/[SLES]
357 |See http://packages.efficios.com/[EfficiOS Enterprise Packages].
360 |https://buildroot.org/[Buildroot]
361 |xref:buildroot[Buildroot{nbsp}2018.02, Buildroot{nbsp}2018.05,
362 Buildroot{nbsp}2018.08, and Buildroot{nbsp}2018.11].
363 |<<building-from-source,Build LTTng{nbsp}{revision} from source>>.
365 |http://www.openembedded.org/wiki/Main_Page[OpenEmbedded] and
366 https://www.yoctoproject.org/[Yocto]
368 |<<building-from-source,Build LTTng{nbsp}{revision} from source>>.
373 === [[ubuntu-official-repositories]]Ubuntu
375 LTTng{nbsp}{revision} is available on Ubuntu{nbsp}18.04 _Bionic Beaver_.
376 For previous supported releases of Ubuntu,
377 <<ubuntu-ppa,use the LTTng Stable{nbsp}{revision} PPA>>.
379 To install LTTng{nbsp}{revision} on Ubuntu{nbsp}18.04 _Bionic Beaver_:
381 . Install the main LTTng{nbsp}{revision} packages:
386 # apt-get install lttng-tools
387 # apt-get install lttng-modules-dkms
388 # apt-get install liblttng-ust-dev
392 . **If you need to instrument and trace
393 <<java-application,Java applications>>**, install the LTTng-UST
399 # apt-get install liblttng-ust-agent-java
403 . **If you need to instrument and trace
404 <<python-application,Python{nbsp}3 applications>>**, install the
405 LTTng-UST Python agent:
410 # apt-get install python3-lttngust
416 ==== noch:{LTTng} Stable {revision} PPA
418 The https://launchpad.net/~lttng/+archive/ubuntu/stable-{revision}[LTTng
419 Stable{nbsp}{revision} PPA] offers the latest stable
420 LTTng{nbsp}{revision} packages for:
422 * Ubuntu{nbsp}14.04 _Trusty Tahr_
423 * Ubuntu{nbsp}16.04 _Xenial Xerus_
424 * Ubuntu{nbsp}18.04 _Bionic Beaver_
426 To install LTTng{nbsp}{revision} from the LTTng Stable{nbsp}{revision} PPA:
428 . Add the LTTng Stable{nbsp}{revision} PPA repository and update the
434 # apt-add-repository ppa:lttng/stable-2.10
439 . Install the main LTTng{nbsp}{revision} packages:
444 # apt-get install lttng-tools
445 # apt-get install lttng-modules-dkms
446 # apt-get install liblttng-ust-dev
450 . **If you need to instrument and trace
451 <<java-application,Java applications>>**, install the LTTng-UST
457 # apt-get install liblttng-ust-agent-java
461 . **If you need to instrument and trace
462 <<python-application,Python{nbsp}3 applications>>**, install the
463 LTTng-UST Python agent:
468 # apt-get install python3-lttngust
476 To install LTTng{nbsp}{revision} on Fedora{nbsp}27, Fedora{nbsp}28,
479 . Install the LTTng-tools{nbsp}{revision} and LTTng-UST{nbsp}{revision}
485 # yum install lttng-tools
486 # yum install lttng-ust
490 . Download, build, and install the latest LTTng-modules{nbsp}{revision}:
496 wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.10.tar.bz2 &&
497 tar -xf lttng-modules-latest-2.10.tar.bz2 &&
498 cd lttng-modules-2.10.* &&
500 sudo make modules_install &&
506 .Java and Python application instrumentation and tracing
508 If you need to instrument and trace <<java-application,Java
509 applications>> on Fedora, you need to build and install
510 LTTng-UST{nbsp}{revision} <<building-from-source,from source>> and pass
511 the `--enable-java-agent-jul`, `--enable-java-agent-log4j`, or
512 `--enable-java-agent-all` options to the `configure` script, depending
513 on which Java logging framework you use.
515 If you need to instrument and trace <<python-application,Python
516 applications>> on Fedora, you need to build and install
517 LTTng-UST{nbsp}{revision} from source and pass the
518 `--enable-python-agent` option to the `configure` script.
525 To install LTTng{nbsp}{revision} on Debian "buster" (testing):
527 . Install the main LTTng{nbsp}{revision} packages:
532 # apt-get install lttng-modules-dkms
533 # apt-get install liblttng-ust-dev
534 # apt-get install lttng-tools
538 . **If you need to instrument and trace <<java-application,Java
539 applications>>**, install the LTTng-UST Java agent:
544 # apt-get install liblttng-ust-agent-java
548 . **If you need to instrument and trace <<python-application,Python
549 applications>>**, install the LTTng-UST Python agent:
554 # apt-get install python3-lttngust
562 LTTng-UST{nbsp}{revision} is available in Arch Linux's _Community_
563 repository, while LTTng-tools{nbsp}{revision} and
564 LTTng-modules{nbsp}{revision} are available in the
565 https://aur.archlinux.org/[AUR].
567 To install LTTng{nbsp}{revision} on Arch Linux, using
568 https://github.com/rmarquis/pacaur[pacaur] for the AUR packages:
570 . Install the main LTTng{nbsp}{revision} packages:
575 # pacman -Sy lttng-ust
576 $ pikaur -Sy lttng-tools
577 $ pikaur -Sy lttng-modules
581 . **If you need to instrument and trace <<python-application,Python
582 applications>>**, install the LTTng-UST Python agent:
587 # pacman -Sy python-lttngust
588 # pacman -Sy python2-lttngust
596 To install LTTng-tools{nbsp}{revision} and LTTng-UST{nbsp}{revision} on
597 Alpine Linux{nbsp}3.7 or Alpine Linux{nbsp}3.8:
599 . Add the LTTng packages:
604 # apk add lttng-tools
605 # apk add lttng-ust-dev
609 . Download, build, and install the latest LTTng-modules{nbsp}{revision}:
615 wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.10.tar.bz2 &&
616 tar -xf lttng-modules-latest-2.10.tar.bz2 &&
617 cd lttng-modules-2.10.* &&
619 sudo make modules_install &&
628 To install LTTng{nbsp}{revision} on openSUSE Leap{nbsp}15.0:
630 * Install the main LTTng{nbsp}{revision} packages:
635 sudo zypper install lttng-tools
636 sudo zypper install lttng-modules
637 sudo zypper install lttng-ust-devel
642 .Java and Python application instrumentation and tracing
644 If you need to instrument and trace <<java-application,Java
645 applications>> on openSUSE, you need to build and install
646 LTTng-UST{nbsp}{revision} <<building-from-source,from source>> and pass
647 the `--enable-java-agent-jul`, `--enable-java-agent-log4j`, or
648 `--enable-java-agent-all` options to the `configure` script, depending
649 on which Java logging framework you use.
651 If you need to instrument and trace <<python-application,Python
652 applications>> on openSUSE, you need to build and install
653 LTTng-UST{nbsp}{revision} from source and pass the
654 `--enable-python-agent` option to the `configure` script.
658 [[enterprise-distributions]]
659 === RHEL, SUSE, and other enterprise distributions
661 To install LTTng on enterprise Linux distributions, such as Red Hat
662 Enterprise Linux (RHEL) and SUSE Linux Enterprise Server (SUSE), please
663 see http://packages.efficios.com/[EfficiOS Enterprise Packages].
669 To install LTTng{nbsp}{revision} on Buildroot{nbsp}2018.02,
670 Buildroot{nbsp}2018.05, Buildroot{nbsp}2018.08, or
671 Buildroot{nbsp}2018.11:
673 . Launch the Buildroot configuration tool:
682 . In **Kernel**, check **Linux kernel**.
683 . In **Toolchain**, check **Enable WCHAR support**.
684 . In **Target packages**{nbsp}→ **Debugging, profiling and benchmark**,
685 check **lttng-modules** and **lttng-tools**.
686 . In **Target packages**{nbsp}→ **Libraries**{nbsp}→
687 **Other**, check **lttng-libust**.
690 [[building-from-source]]
691 === Build from source
693 To build and install LTTng{nbsp}{revision} from source:
695 . Using your distribution's package manager, or from source, install
696 the following dependencies of LTTng-tools and LTTng-UST:
699 * https://sourceforge.net/projects/libuuid/[libuuid]
700 * http://directory.fsf.org/wiki/Popt[popt]
701 * http://liburcu.org/[Userspace RCU]
702 * http://www.xmlsoft.org/[libxml2]
705 . Download, build, and install the latest LTTng-modules{nbsp}{revision}:
711 wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.10.tar.bz2 &&
712 tar -xf lttng-modules-latest-2.10.tar.bz2 &&
713 cd lttng-modules-2.10.* &&
715 sudo make modules_install &&
720 . Download, build, and install the latest LTTng-UST{nbsp}{revision}:
726 wget http://lttng.org/files/lttng-ust/lttng-ust-latest-2.10.tar.bz2 &&
727 tar -xf lttng-ust-latest-2.10.tar.bz2 &&
728 cd lttng-ust-2.10.* &&
738 .Java and Python application tracing
740 If you need to instrument and trace <<java-application,Java
741 applications>>, pass the `--enable-java-agent-jul`,
742 `--enable-java-agent-log4j`, or `--enable-java-agent-all` options to the
743 `configure` script, depending on which Java logging framework you use.
745 If you need to instrument and trace <<python-application,Python
746 applications>>, pass the `--enable-python-agent` option to the
747 `configure` script. You can set the `PYTHON` environment variable to the
748 path to the Python interpreter for which to install the LTTng-UST Python
756 By default, LTTng-UST libraries are installed to
757 dir:{/usr/local/lib}, which is the de facto directory in which to
758 keep self-compiled and third-party libraries.
760 When <<building-tracepoint-providers-and-user-application,linking an
761 instrumented user application with `liblttng-ust`>>:
763 * Append `/usr/local/lib` to the env:LD_LIBRARY_PATH environment
765 * Pass the `-L/usr/local/lib` and `-Wl,-rpath,/usr/local/lib` options to
766 man:gcc(1), man:g++(1), or man:clang(1).
770 . Download, build, and install the latest LTTng-tools{nbsp}{revision}:
776 wget http://lttng.org/files/lttng-tools/lttng-tools-latest-2.10.tar.bz2 &&
777 tar -xf lttng-tools-latest-2.10.tar.bz2 &&
778 cd lttng-tools-2.10.* &&
786 TIP: The https://github.com/eepp/vlttng[vlttng tool] can do all the
787 previous steps automatically for a given version of LTTng and confine
788 the installed files in a specific directory. This can be useful to test
789 LTTng without installing it on your system.
795 This is a short guide to get started quickly with LTTng kernel and user
798 Before you follow this guide, make sure to <<installing-lttng,install>>
801 This tutorial walks you through the steps to:
803 . <<tracing-the-linux-kernel,Trace the Linux kernel>>.
804 . <<tracing-your-own-user-application,Trace a user application>> written
806 . <<viewing-and-analyzing-your-traces,View and analyze the
810 [[tracing-the-linux-kernel]]
811 === Trace the Linux kernel
813 The following command lines start with the `#` prompt because you need
814 root privileges to trace the Linux kernel. You can also trace the kernel
815 as a regular user if your Unix user is a member of the
816 <<tracing-group,tracing group>>.
818 . Create a <<tracing-session,tracing session>> which writes its traces
819 to dir:{/tmp/my-kernel-trace}:
824 # lttng create my-kernel-session --output=/tmp/my-kernel-trace
828 . List the available kernel tracepoints and system calls:
833 # lttng list --kernel
834 # lttng list --kernel --syscall
838 . Create <<event,event rules>> which match the desired instrumentation
839 point names, for example the `sched_switch` and `sched_process_fork`
840 tracepoints, and the man:open(2) and man:close(2) system calls:
845 # lttng enable-event --kernel sched_switch,sched_process_fork
846 # lttng enable-event --kernel --syscall open,close
850 You can also create an event rule which matches _all_ the Linux kernel
851 tracepoints (this will generate a lot of data when tracing):
856 # lttng enable-event --kernel --all
860 . <<basic-tracing-session-control,Start tracing>>:
869 . Do some operation on your system for a few seconds. For example,
870 load a website, or list the files of a directory.
871 . <<basic-tracing-session-control,Stop tracing>> and destroy the
882 The man:lttng-destroy(1) command does not destroy the trace data; it
883 only destroys the state of the tracing session.
885 . For the sake of this example, make the recorded trace accessible to
891 # chown -R $(whoami) /tmp/my-kernel-trace
895 See <<viewing-and-analyzing-your-traces,View and analyze the
896 recorded events>> to view the recorded events.
899 [[tracing-your-own-user-application]]
900 === Trace a user application
902 This section steps you through a simple example to trace a
903 _Hello world_ program written in C.
905 To create the traceable user application:
907 . Create the tracepoint provider header file, which defines the
908 tracepoints and the events they can generate:
914 #undef TRACEPOINT_PROVIDER
915 #define TRACEPOINT_PROVIDER hello_world
917 #undef TRACEPOINT_INCLUDE
918 #define TRACEPOINT_INCLUDE "./hello-tp.h"
920 #if !defined(_HELLO_TP_H) || defined(TRACEPOINT_HEADER_MULTI_READ)
923 #include <lttng/tracepoint.h>
933 ctf_string(my_string_field, my_string_arg)
934 ctf_integer(int, my_integer_field, my_integer_arg)
938 #endif /* _HELLO_TP_H */
940 #include <lttng/tracepoint-event.h>
944 . Create the tracepoint provider package source file:
950 #define TRACEPOINT_CREATE_PROBES
951 #define TRACEPOINT_DEFINE
953 #include "hello-tp.h"
957 . Build the tracepoint provider package:
962 $ gcc -c -I. hello-tp.c
966 . Create the _Hello World_ application source file:
973 #include "hello-tp.h"
975 int main(int argc, char *argv[])
979 puts("Hello, World!\nPress Enter to continue...");
982 * The following getchar() call is only placed here for the purpose
983 * of this demonstration, to pause the application in order for
984 * you to have time to list its tracepoints. It is not
990 * A tracepoint() call.
992 * Arguments, as defined in hello-tp.h:
994 * 1. Tracepoint provider name (required)
995 * 2. Tracepoint name (required)
996 * 3. my_integer_arg (first user-defined argument)
997 * 4. my_string_arg (second user-defined argument)
999 * Notice the tracepoint provider and tracepoint names are
1000 * NOT strings: they are in fact parts of variables that the
1001 * macros in hello-tp.h create.
1003 tracepoint(hello_world, my_first_tracepoint, 23, "hi there!");
1005 for (x = 0; x < argc; ++x) {
1006 tracepoint(hello_world, my_first_tracepoint, x, argv[x]);
1009 puts("Quitting now!");
1010 tracepoint(hello_world, my_first_tracepoint, x * x, "x^2");
1017 . Build the application:
1026 . Link the application with the tracepoint provider package,
1027 `liblttng-ust`, and `libdl`:
1032 $ gcc -o hello hello.o hello-tp.o -llttng-ust -ldl
1036 Here's the whole build process:
1039 .User space tracing tutorial's build steps.
1040 image::ust-flow.png[]
1042 To trace the user application:
1044 . Run the application with a few arguments:
1049 $ ./hello world and beyond
1058 Press Enter to continue...
1062 . Start an LTTng <<lttng-sessiond,session daemon>>:
1067 $ lttng-sessiond --daemonize
1071 Note that a session daemon might already be running, for example as
1072 a service that the distribution's service manager started.
1074 . List the available user space tracepoints:
1079 $ lttng list --userspace
1083 You see the `hello_world:my_first_tracepoint` tracepoint listed
1084 under the `./hello` process.
1086 . Create a <<tracing-session,tracing session>>:
1091 $ lttng create my-user-space-session
1095 . Create an <<event,event rule>> which matches the
1096 `hello_world:my_first_tracepoint` event name:
1101 $ lttng enable-event --userspace hello_world:my_first_tracepoint
1105 . <<basic-tracing-session-control,Start tracing>>:
1114 . Go back to the running `hello` application and press Enter. The
1115 program executes all `tracepoint()` instrumentation points and exits.
1116 . <<basic-tracing-session-control,Stop tracing>> and destroy the
1127 The man:lttng-destroy(1) command does not destroy the trace data; it
1128 only destroys the state of the tracing session.
1130 By default, LTTng saves the traces in
1131 +$LTTNG_HOME/lttng-traces/__name__-__date__-__time__+,
1132 where +__name__+ is the tracing session name. The
1133 env:LTTNG_HOME environment variable defaults to `$HOME` if not set.
1135 See <<viewing-and-analyzing-your-traces,View and analyze the
1136 recorded events>> to view the recorded events.
1139 [[viewing-and-analyzing-your-traces]]
1140 === View and analyze the recorded events
1142 Once you have completed the <<tracing-the-linux-kernel,Trace the Linux
1143 kernel>> and <<tracing-your-own-user-application,Trace a user
1144 application>> tutorials, you can inspect the recorded events.
1146 Many tools are available to read LTTng traces:
1148 * **cmd:babeltrace** is a command-line utility which converts trace
1149 formats; it supports the format that LTTng produces, CTF, as well as a
1150 basic text output which can be ++grep++ed. The cmd:babeltrace command
1151 is part of the http://diamon.org/babeltrace[Babeltrace] project.
1152 * Babeltrace also includes
1153 **https://www.python.org/[Python] bindings** so
1154 that you can easily open and read an LTTng trace with your own script,
1155 benefiting from the power of Python.
1156 * http://tracecompass.org/[**Trace Compass**]
1157 is a graphical user interface for viewing and analyzing any type of
1158 logs or traces, including LTTng's.
1159 * https://github.com/lttng/lttng-analyses[**LTTng analyses**] is a
1160 project which includes many high-level analyses of LTTng kernel
1161 traces, like scheduling statistics, interrupt frequency distribution,
1162 top CPU usage, and more.
1164 NOTE: This section assumes that the traces recorded during the previous
1165 tutorials were saved to their default location, in the
1166 dir:{$LTTNG_HOME/lttng-traces} directory. The env:LTTNG_HOME
1167 environment variable defaults to `$HOME` if not set.
1170 [[viewing-and-analyzing-your-traces-bt]]
1171 ==== Use the cmd:babeltrace command-line tool
1173 The simplest way to list all the recorded events of a trace is to pass
1174 its path to cmd:babeltrace with no options:
1178 $ babeltrace ~/lttng-traces/my-user-space-session*
1181 cmd:babeltrace finds all traces recursively within the given path and
1182 prints all their events, merging them in chronological order.
1184 You can pipe the output of cmd:babeltrace into a tool like man:grep(1) for
1189 $ babeltrace /tmp/my-kernel-trace | grep _switch
1192 You can pipe the output of cmd:babeltrace into a tool like man:wc(1) to
1193 count the recorded events:
1197 $ babeltrace /tmp/my-kernel-trace | grep _open | wc --lines
1201 [[viewing-and-analyzing-your-traces-bt-python]]
1202 ==== Use the Babeltrace Python bindings
1204 The <<viewing-and-analyzing-your-traces-bt,text output of cmd:babeltrace>>
1205 is useful to isolate events by simple matching using man:grep(1) and
1206 similar utilities. However, more elaborate filters, such as keeping only
1207 event records with a field value falling within a specific range, are
1208 not trivial to write using a shell. Moreover, reductions and even the
1209 most basic computations involving multiple event records are virtually
1210 impossible to implement.
1212 Fortunately, Babeltrace ships with Python 3 bindings which makes it easy
1213 to read the event records of an LTTng trace sequentially and compute the
1214 desired information.
1216 The following script accepts an LTTng Linux kernel trace path as its
1217 first argument and prints the short names of the top 5 running processes
1218 on CPU 0 during the whole trace:
1223 from collections import Counter
1229 if len(sys.argv) != 2:
1230 msg = 'Usage: python3 {} TRACEPATH'.format(sys.argv[0])
1231 print(msg, file=sys.stderr)
1234 # A trace collection contains one or more traces
1235 col = babeltrace.TraceCollection()
1237 # Add the trace provided by the user (LTTng traces always have
1239 if col.add_trace(sys.argv[1], 'ctf') is None:
1240 raise RuntimeError('Cannot add trace')
1242 # This counter dict contains execution times:
1244 # task command name -> total execution time (ns)
1245 exec_times = Counter()
1247 # This contains the last `sched_switch` timestamp
1251 for event in col.events:
1252 # Keep only `sched_switch` events
1253 if event.name != 'sched_switch':
1256 # Keep only events which happened on CPU 0
1257 if event['cpu_id'] != 0:
1261 cur_ts = event.timestamp
1267 # Previous task command (short) name
1268 prev_comm = event['prev_comm']
1270 # Initialize entry in our dict if not yet done
1271 if prev_comm not in exec_times:
1272 exec_times[prev_comm] = 0
1274 # Compute previous command execution time
1275 diff = cur_ts - last_ts
1277 # Update execution time of this command
1278 exec_times[prev_comm] += diff
1280 # Update last timestamp
1284 for name, ns in exec_times.most_common(5):
1286 print('{:20}{} s'.format(name, s))
1291 if __name__ == '__main__':
1292 sys.exit(0 if top5proc() else 1)
1299 $ python3 top5proc.py /tmp/my-kernel-trace/kernel
1305 swapper/0 48.607245889 s
1306 chromium 7.192738188 s
1307 pavucontrol 0.709894415 s
1308 Compositor 0.660867933 s
1309 Xorg.bin 0.616753786 s
1312 Note that `swapper/0` is the "idle" process of CPU 0 on Linux; since we
1313 weren't using the CPU that much when tracing, its first position in the
1318 == [[understanding-lttng]]Core concepts
1320 From a user's perspective, the LTTng system is built on a few concepts,
1321 or objects, on which the <<lttng-cli,cmd:lttng command-line tool>>
1322 operates by sending commands to the <<lttng-sessiond,session daemon>>.
1323 Understanding how those objects relate to eachother is key in mastering
1326 The core concepts are:
1328 * <<tracing-session,Tracing session>>
1329 * <<domain,Tracing domain>>
1330 * <<channel,Channel and ring buffer>>
1331 * <<"event","Instrumentation point, event rule, event, and event record">>
1337 A _tracing session_ is a stateful dialogue between you and
1338 a <<lttng-sessiond,session daemon>>. You can
1339 <<creating-destroying-tracing-sessions,create a new tracing
1340 session>> with the `lttng create` command.
1342 Anything that you do when you control LTTng tracers happens within a
1343 tracing session. In particular, a tracing session:
1346 * Has its own set of trace files.
1347 * Has its own state of activity (started or stopped).
1348 * Has its own <<tracing-session-mode,mode>> (local, network streaming,
1350 * Has its own <<channel,channels>> which have their own
1351 <<event,event rules>>.
1354 .A _tracing session_ contains <<channel,channels>> that are members of <<domain,tracing domains>> and contain <<event,event rules>>.
1355 image::concepts.png[]
1357 Those attributes and objects are completely isolated between different
1360 A tracing session is analogous to a cash machine session:
1361 the operations you do on the banking system through the cash machine do
1362 not alter the data of other users of the same system. In the case of
1363 the cash machine, a session lasts as long as your bank card is inside.
1364 In the case of LTTng, a tracing session lasts from the `lttng create`
1365 command to the `lttng destroy` command.
1368 .Each Unix user has its own set of tracing sessions.
1369 image::many-sessions.png[]
1372 [[tracing-session-mode]]
1373 ==== Tracing session mode
1375 LTTng can send the generated trace data to different locations. The
1376 _tracing session mode_ dictates where to send it. The following modes
1377 are available in LTTng{nbsp}{revision}:
1380 LTTng writes the traces to the file system of the machine being traced
1383 Network streaming mode::
1384 LTTng sends the traces over the network to a
1385 <<lttng-relayd,relay daemon>> running on a remote system.
1388 LTTng does not write the traces by default. Instead, you can request
1389 LTTng to <<taking-a-snapshot,take a snapshot>>, that is, a copy of the
1390 current tracing buffers, and to write it to the target's file system
1391 or to send it over the network to a <<lttng-relayd,relay daemon>>
1392 running on a remote system.
1395 This mode is similar to the network streaming mode, but a live
1396 trace viewer can connect to the distant relay daemon to
1397 <<lttng-live,view event records as LTTng generates them>> by
1404 A _tracing domain_ is a namespace for event sources. A tracing domain
1405 has its own properties and features.
1407 There are currently five available tracing domains:
1411 * `java.util.logging` (JUL)
1415 You must specify a tracing domain when using some commands to avoid
1416 ambiguity. For example, since all the domains support named tracepoints
1417 as event sources (instrumentation points that you manually insert in the
1418 source code), you need to specify a tracing domain when
1419 <<enabling-disabling-events,creating an event rule>> because all the
1420 tracing domains could have tracepoints with the same names.
1422 Some features are reserved to specific tracing domains. Dynamic function
1423 entry and return instrumentation points, for example, are currently only
1424 supported in the Linux kernel tracing domain, but support for other
1425 tracing domains could be added in the future.
1427 You can create <<channel,channels>> in the Linux kernel and user space
1428 tracing domains. The other tracing domains have a single default
1433 === Channel and ring buffer
1435 A _channel_ is an object which is responsible for a set of ring buffers.
1436 Each ring buffer is divided into multiple sub-buffers. When an LTTng
1437 tracer emits an event, it can record it to one or more
1438 sub-buffers. The attributes of a channel determine what to do when
1439 there's no space left for a new event record because all sub-buffers
1440 are full, where to send a full sub-buffer, and other behaviours.
1442 A channel is always associated to a <<domain,tracing domain>>. The
1443 `java.util.logging` (JUL), log4j, and Python tracing domains each have
1444 a default channel which you cannot configure.
1446 A channel also owns <<event,event rules>>. When an LTTng tracer emits
1447 an event, it records it to the sub-buffers of all
1448 the enabled channels with a satisfied event rule, as long as those
1449 channels are part of active <<tracing-session,tracing sessions>>.
1452 [[channel-buffering-schemes]]
1453 ==== Per-user vs. per-process buffering schemes
1455 A channel has at least one ring buffer _per CPU_. LTTng always
1456 records an event to the ring buffer associated to the CPU on which it
1459 Two _buffering schemes_ are available when you
1460 <<enabling-disabling-channels,create a channel>> in the
1461 user space <<domain,tracing domain>>:
1463 Per-user buffering::
1464 Allocate one set of ring buffers--one per CPU--shared by all the
1465 instrumented processes of each Unix user.
1469 .Per-user buffering scheme.
1470 image::per-user-buffering.png[]
1473 Per-process buffering::
1474 Allocate one set of ring buffers--one per CPU--for each
1475 instrumented process.
1479 .Per-process buffering scheme.
1480 image::per-process-buffering.png[]
1483 The per-process buffering scheme tends to consume more memory than the
1484 per-user option because systems generally have more instrumented
1485 processes than Unix users running instrumented processes. However, the
1486 per-process buffering scheme ensures that one process having a high
1487 event throughput won't fill all the shared sub-buffers of the same
1490 The Linux kernel tracing domain has only one available buffering scheme
1491 which is to allocate a single set of ring buffers for the whole system.
1492 This scheme is similar to the per-user option, but with a single, global
1493 user "running" the kernel.
1496 [[channel-overwrite-mode-vs-discard-mode]]
1497 ==== Overwrite vs. discard event loss modes
1499 When an event occurs, LTTng records it to a specific sub-buffer (yellow
1500 arc in the following animation) of a specific channel's ring buffer.
1501 When there's no space left in a sub-buffer, the tracer marks it as
1502 consumable (red) and another, empty sub-buffer starts receiving the
1503 following event records. A <<lttng-consumerd,consumer daemon>>
1504 eventually consumes the marked sub-buffer (returns to white).
1507 [role="docsvg-channel-subbuf-anim"]
1512 In an ideal world, sub-buffers are consumed faster than they are filled,
1513 as is the case in the previous animation. In the real world,
1514 however, all sub-buffers can be full at some point, leaving no space to
1515 record the following events.
1517 By default, LTTng-modules and LTTng-UST are _non-blocking_ tracers: when
1518 no empty sub-buffer is available, it is acceptable to lose event records
1519 when the alternative would be to cause substantial delays in the
1520 instrumented application's execution. LTTng privileges performance over
1521 integrity; it aims at perturbing the traced system as little as possible
1522 in order to make tracing of subtle race conditions and rare interrupt
1525 Starting from LTTng{nbsp}2.10, the LTTng user space tracer, LTTng-UST,
1526 supports a _blocking mode_. See the <<blocking-timeout-example,blocking
1527 timeout example>> to learn how to use the blocking mode.
1529 When it comes to losing event records because no empty sub-buffer is
1530 available, or because the <<opt-blocking-timeout,blocking timeout>> is
1531 reached, the channel's _event loss mode_ determines what to do. The
1532 available event loss modes are:
1535 Drop the newest event records until a the tracer releases a
1538 This is the only available mode when you specify a
1539 <<opt-blocking-timeout,blocking timeout>>.
1542 Clear the sub-buffer containing the oldest event records and start
1543 writing the newest event records there.
1545 This mode is sometimes called _flight recorder mode_ because it's
1547 https://en.wikipedia.org/wiki/Flight_recorder[flight recorder]:
1548 always keep a fixed amount of the latest data.
1550 Which mechanism you should choose depends on your context: prioritize
1551 the newest or the oldest event records in the ring buffer?
1553 Beware that, in overwrite mode, the tracer abandons a _whole sub-buffer_
1554 as soon as a there's no space left for a new event record, whereas in
1555 discard mode, the tracer only discards the event record that doesn't
1558 In discard mode, LTTng increments a count of lost event records when an
1559 event record is lost and saves this count to the trace. In overwrite
1560 mode, since LTTng 2.8, LTTng increments a count of lost sub-buffers when
1561 a sub-buffer is lost and saves this count to the trace. In this mode,
1562 the exact number of lost event records in those lost sub-buffers is not
1563 saved to the trace. Trace analyses can use the trace's saved discarded
1564 event record and sub-buffer counts to decide whether or not to perform
1565 the analyses even if trace data is known to be missing.
1567 There are a few ways to decrease your probability of losing event
1569 <<channel-subbuf-size-vs-subbuf-count,Sub-buffer count and size>> shows
1570 how you can fine-une the sub-buffer count and size of a channel to
1571 virtually stop losing event records, though at the cost of greater
1575 [[channel-subbuf-size-vs-subbuf-count]]
1576 ==== Sub-buffer count and size
1578 When you <<enabling-disabling-channels,create a channel>>, you can
1579 set its number of sub-buffers and their size.
1581 Note that there is noticeable CPU overhead introduced when
1582 switching sub-buffers (marking a full one as consumable and switching
1583 to an empty one for the following events to be recorded). Knowing this,
1584 the following list presents a few practical situations along with how
1585 to configure the sub-buffer count and size for them:
1587 * **High event throughput**: In general, prefer bigger sub-buffers to
1588 lower the risk of losing event records.
1590 Having bigger sub-buffers also ensures a lower
1591 <<channel-switch-timer,sub-buffer switching frequency>>.
1593 The number of sub-buffers is only meaningful if you create the channel
1594 in overwrite mode: in this case, if a sub-buffer overwrite happens, the
1595 other sub-buffers are left unaltered.
1597 * **Low event throughput**: In general, prefer smaller sub-buffers
1598 since the risk of losing event records is low.
1600 Because events occur less frequently, the sub-buffer switching frequency
1601 should remain low and thus the tracer's overhead should not be a
1604 * **Low memory system**: If your target system has a low memory
1605 limit, prefer fewer first, then smaller sub-buffers.
1607 Even if the system is limited in memory, you want to keep the
1608 sub-buffers as big as possible to avoid a high sub-buffer switching
1611 Note that LTTng uses http://diamon.org/ctf/[CTF] as its trace format,
1612 which means event data is very compact. For example, the average
1613 LTTng kernel event record weights about 32{nbsp}bytes. Thus, a
1614 sub-buffer size of 1{nbsp}MiB is considered big.
1616 The previous situations highlight the major trade-off between a few big
1617 sub-buffers and more, smaller sub-buffers: sub-buffer switching
1618 frequency vs. how much data is lost in overwrite mode. Assuming a
1619 constant event throughput and using the overwrite mode, the two
1620 following configurations have the same ring buffer total size:
1623 [role="docsvg-channel-subbuf-size-vs-count-anim"]
1628 * **2 sub-buffers of 4{nbsp}MiB each**: Expect a very low sub-buffer
1629 switching frequency, but if a sub-buffer overwrite happens, half of
1630 the event records so far (4{nbsp}MiB) are definitely lost.
1631 * **8 sub-buffers of 1{nbsp}MiB each**: Expect 4{nbsp}times the tracer's
1632 overhead as the previous configuration, but if a sub-buffer
1633 overwrite happens, only the eighth of event records so far are
1636 In discard mode, the sub-buffers count parameter is pointless: use two
1637 sub-buffers and set their size according to the requirements of your
1641 [[channel-switch-timer]]
1642 ==== Switch timer period
1644 The _switch timer period_ is an important configurable attribute of
1645 a channel to ensure periodic sub-buffer flushing.
1647 When the _switch timer_ expires, a sub-buffer switch happens. You can
1648 set the switch timer period attribute when you
1649 <<enabling-disabling-channels,create a channel>> to ensure that event
1650 data is consumed and committed to trace files or to a distant relay
1651 daemon periodically in case of a low event throughput.
1654 [role="docsvg-channel-switch-timer"]
1659 This attribute is also convenient when you use big sub-buffers to cope
1660 with a sporadic high event throughput, even if the throughput is
1664 [[channel-read-timer]]
1665 ==== Read timer period
1667 By default, the LTTng tracers use a notification mechanism to signal a
1668 full sub-buffer so that a consumer daemon can consume it. When such
1669 notifications must be avoided, for example in real-time applications,
1670 you can use the channel's _read timer_ instead. When the read timer
1671 fires, the <<lttng-consumerd,consumer daemon>> checks for full,
1672 consumable sub-buffers.
1675 [[tracefile-rotation]]
1676 ==== Trace file count and size
1678 By default, trace files can grow as large as needed. You can set the
1679 maximum size of each trace file that a channel writes when you
1680 <<enabling-disabling-channels,create a channel>>. When the size of
1681 a trace file reaches the channel's fixed maximum size, LTTng creates
1682 another file to contain the next event records. LTTng appends a file
1683 count to each trace file name in this case.
1685 If you set the trace file size attribute when you create a channel, the
1686 maximum number of trace files that LTTng creates is _unlimited_ by
1687 default. To limit them, you can also set a maximum number of trace
1688 files. When the number of trace files reaches the channel's fixed
1689 maximum count, the oldest trace file is overwritten. This mechanism is
1690 called _trace file rotation_.
1694 === Instrumentation point, event rule, event, and event record
1696 An _event rule_ is a set of conditions which must be **all** satisfied
1697 for LTTng to record an occuring event.
1699 You set the conditions when you <<enabling-disabling-events,create
1702 You always attach an event rule to <<channel,channel>> when you create
1705 When an event passes the conditions of an event rule, LTTng records it
1706 in one of the attached channel's sub-buffers.
1708 The available conditions, as of LTTng{nbsp}{revision}, are:
1710 * The event rule _is enabled_.
1711 * The instrumentation point's type _is{nbsp}T_.
1712 * The instrumentation point's name (sometimes called _event name_)
1713 _matches{nbsp}N_, but _is not{nbsp}E_.
1714 * The instrumentation point's log level _is as severe as{nbsp}L_, or
1715 _is exactly{nbsp}L_.
1716 * The fields of the event's payload _satisfy_ a filter
1717 expression{nbsp}__F__.
1719 As you can see, all the conditions but the dynamic filter are related to
1720 the event rule's status or to the instrumentation point, not to the
1721 occurring events. This is why, without a filter, checking if an event
1722 passes an event rule is not a dynamic task: when you create or modify an
1723 event rule, all the tracers of its tracing domain enable or disable the
1724 instrumentation points themselves once. This is possible because the
1725 attributes of an instrumentation point (type, name, and log level) are
1726 defined statically. In other words, without a dynamic filter, the tracer
1727 _does not evaluate_ the arguments of an instrumentation point unless it
1728 matches an enabled event rule.
1730 Note that, for LTTng to record an event, the <<channel,channel>> to
1731 which a matching event rule is attached must also be enabled, and the
1732 tracing session owning this channel must be active.
1735 .Logical path from an instrumentation point to an event record.
1736 image::event-rule.png[]
1738 .Event, event record, or event rule?
1740 With so many similar terms, it's easy to get confused.
1742 An **event** is the consequence of the execution of an _instrumentation
1743 point_, like a tracepoint that you manually place in some source code,
1744 or a Linux kernel KProbe. An event is said to _occur_ at a specific
1745 time. Different actions can be taken upon the occurrence of an event,
1746 like record the event's payload to a buffer.
1748 An **event record** is the representation of an event in a sub-buffer. A
1749 tracer is responsible for capturing the payload of an event, current
1750 context variables, the event's ID, and the event's timestamp. LTTng
1751 can append this sub-buffer to a trace file.
1753 An **event rule** is a set of conditions which must all be satisfied for
1754 LTTng to record an occuring event. Events still occur without
1755 satisfying event rules, but LTTng does not record them.
1760 == Components of noch:{LTTng}
1762 The second _T_ in _LTTng_ stands for _toolkit_: it would be wrong
1763 to call LTTng a simple _tool_ since it is composed of multiple
1764 interacting components. This section describes those components,
1765 explains their respective roles, and shows how they connect together to
1766 form the LTTng ecosystem.
1768 The following diagram shows how the most important components of LTTng
1769 interact with user applications, the Linux kernel, and you:
1772 .Control and trace data paths between LTTng components.
1773 image::plumbing.png[]
1775 The LTTng project incorporates:
1777 * **LTTng-tools**: Libraries and command-line interface to
1778 control tracing sessions.
1779 ** <<lttng-sessiond,Session daemon>> (man:lttng-sessiond(8)).
1780 ** <<lttng-consumerd,Consumer daemon>> (cmd:lttng-consumerd).
1781 ** <<lttng-relayd,Relay daemon>> (man:lttng-relayd(8)).
1782 ** <<liblttng-ctl-lttng,Tracing control library>> (`liblttng-ctl`).
1783 ** <<lttng-cli,Tracing control command-line tool>> (man:lttng(1)).
1784 * **LTTng-UST**: Libraries and Java/Python packages to trace user
1786 ** <<lttng-ust,User space tracing library>> (`liblttng-ust`) and its
1787 headers to instrument and trace any native user application.
1788 ** <<prebuilt-ust-helpers,Preloadable user space tracing helpers>>:
1789 *** `liblttng-ust-libc-wrapper`
1790 *** `liblttng-ust-pthread-wrapper`
1791 *** `liblttng-ust-cyg-profile`
1792 *** `liblttng-ust-cyg-profile-fast`
1793 *** `liblttng-ust-dl`
1794 ** User space tracepoint provider source files generator command-line
1795 tool (man:lttng-gen-tp(1)).
1796 ** <<lttng-ust-agents,LTTng-UST Java agent>> to instrument and trace
1797 Java applications using `java.util.logging` or
1798 Apache log4j 1.2 logging.
1799 ** <<lttng-ust-agents,LTTng-UST Python agent>> to instrument
1800 Python applications using the standard `logging` package.
1801 * **LTTng-modules**: <<lttng-modules,Linux kernel modules>> to trace
1803 ** LTTng kernel tracer module.
1804 ** Tracing ring buffer kernel modules.
1805 ** Probe kernel modules.
1806 ** LTTng logger kernel module.
1810 === Tracing control command-line interface
1813 .The tracing control command-line interface.
1814 image::plumbing-lttng-cli.png[]
1816 The _man:lttng(1) command-line tool_ is the standard user interface to
1817 control LTTng <<tracing-session,tracing sessions>>. The cmd:lttng tool
1818 is part of LTTng-tools.
1820 The cmd:lttng tool is linked with
1821 <<liblttng-ctl-lttng,`liblttng-ctl`>> to communicate with
1822 one or more <<lttng-sessiond,session daemons>> behind the scenes.
1824 The cmd:lttng tool has a Git-like interface:
1828 $ lttng <GENERAL OPTIONS> <COMMAND> <COMMAND OPTIONS>
1831 The <<controlling-tracing,Tracing control>> section explores the
1832 available features of LTTng using the cmd:lttng tool.
1835 [[liblttng-ctl-lttng]]
1836 === Tracing control library
1839 .The tracing control library.
1840 image::plumbing-liblttng-ctl.png[]
1842 The _LTTng control library_, `liblttng-ctl`, is used to communicate
1843 with a <<lttng-sessiond,session daemon>> using a C API that hides the
1844 underlying protocol's details. `liblttng-ctl` is part of LTTng-tools.
1846 The <<lttng-cli,cmd:lttng command-line tool>>
1847 is linked with `liblttng-ctl`.
1849 You can use `liblttng-ctl` in C or $$C++$$ source code by including its
1854 #include <lttng/lttng.h>
1857 Some objects are referenced by name (C string), such as tracing
1858 sessions, but most of them require to create a handle first using
1859 `lttng_create_handle()`.
1861 The best available developer documentation for `liblttng-ctl` is, as of
1862 LTTng{nbsp}{revision}, its installed header files. Every function and
1863 structure is thoroughly documented.
1867 === User space tracing library
1870 .The user space tracing library.
1871 image::plumbing-liblttng-ust.png[]
1873 The _user space tracing library_, `liblttng-ust` (see man:lttng-ust(3)),
1874 is the LTTng user space tracer. It receives commands from a
1875 <<lttng-sessiond,session daemon>>, for example to
1876 enable and disable specific instrumentation points, and writes event
1877 records to ring buffers shared with a
1878 <<lttng-consumerd,consumer daemon>>.
1879 `liblttng-ust` is part of LTTng-UST.
1881 Public C header files are installed beside `liblttng-ust` to
1882 instrument any <<c-application,C or $$C++$$ application>>.
1884 <<lttng-ust-agents,LTTng-UST agents>>, which are regular Java and Python
1885 packages, use their own library providing tracepoints which is
1886 linked with `liblttng-ust`.
1888 An application or library does not have to initialize `liblttng-ust`
1889 manually: its constructor does the necessary tasks to properly register
1890 to a session daemon. The initialization phase also enables the
1891 instrumentation points matching the <<event,event rules>> that you
1895 [[lttng-ust-agents]]
1896 === User space tracing agents
1899 .The user space tracing agents.
1900 image::plumbing-lttng-ust-agents.png[]
1902 The _LTTng-UST Java and Python agents_ are regular Java and Python
1903 packages which add LTTng tracing capabilities to the
1904 native logging frameworks. The LTTng-UST agents are part of LTTng-UST.
1906 In the case of Java, the
1907 https://docs.oracle.com/javase/7/docs/api/java/util/logging/package-summary.html[`java.util.logging`
1908 core logging facilities] and
1909 https://logging.apache.org/log4j/1.2/[Apache log4j 1.2] are supported.
1910 Note that Apache Log4{nbsp}2 is not supported.
1912 In the case of Python, the standard
1913 https://docs.python.org/3/library/logging.html[`logging`] package
1914 is supported. Both Python 2 and Python 3 modules can import the
1915 LTTng-UST Python agent package.
1917 The applications using the LTTng-UST agents are in the
1918 `java.util.logging` (JUL),
1919 log4j, and Python <<domain,tracing domains>>.
1921 Both agents use the same mechanism to trace the log statements. When an
1922 agent is initialized, it creates a log handler that attaches to the root
1923 logger. The agent also registers to a <<lttng-sessiond,session daemon>>.
1924 When the application executes a log statement, it is passed to the
1925 agent's log handler by the root logger. The agent's log handler calls a
1926 native function in a tracepoint provider package shared library linked
1927 with <<lttng-ust,`liblttng-ust`>>, passing the formatted log message and
1928 other fields, like its logger name and its log level. This native
1929 function contains a user space instrumentation point, hence tracing the
1932 The log level condition of an
1933 <<event,event rule>> is considered when tracing
1934 a Java or a Python application, and it's compatible with the standard
1935 JUL, log4j, and Python log levels.
1939 === LTTng kernel modules
1942 .The LTTng kernel modules.
1943 image::plumbing-lttng-modules.png[]
1945 The _LTTng kernel modules_ are a set of Linux kernel modules
1946 which implement the kernel tracer of the LTTng project. The LTTng
1947 kernel modules are part of LTTng-modules.
1949 The LTTng kernel modules include:
1951 * A set of _probe_ modules.
1953 Each module attaches to a specific subsystem
1954 of the Linux kernel using its tracepoint instrument points. There are
1955 also modules to attach to the entry and return points of the Linux
1956 system call functions.
1958 * _Ring buffer_ modules.
1960 A ring buffer implementation is provided as kernel modules. The LTTng
1961 kernel tracer writes to the ring buffer; a
1962 <<lttng-consumerd,consumer daemon>> reads from the ring buffer.
1964 * The _LTTng kernel tracer_ module.
1965 * The _LTTng logger_ module.
1967 The LTTng logger module implements the special path:{/proc/lttng-logger}
1968 file so that any executable can generate LTTng events by opening and
1969 writing to this file.
1971 See <<proc-lttng-logger-abi,LTTng logger>>.
1973 Generally, you do not have to load the LTTng kernel modules manually
1974 (using man:modprobe(8), for example): a root <<lttng-sessiond,session
1975 daemon>> loads the necessary modules when starting. If you have extra
1976 probe modules, you can specify to load them to the session daemon on
1979 The LTTng kernel modules are installed in
1980 +/usr/lib/modules/__release__/extra+ by default, where +__release__+ is
1981 the kernel release (see `uname --kernel-release`).
1988 .The session daemon.
1989 image::plumbing-sessiond.png[]
1991 The _session daemon_, man:lttng-sessiond(8), is a daemon responsible for
1992 managing tracing sessions and for controlling the various components of
1993 LTTng. The session daemon is part of LTTng-tools.
1995 The session daemon sends control requests to and receives control
1998 * The <<lttng-ust,user space tracing library>>.
2000 Any instance of the user space tracing library first registers to
2001 a session daemon. Then, the session daemon can send requests to
2002 this instance, such as:
2005 ** Get the list of tracepoints.
2006 ** Share an <<event,event rule>> so that the user space tracing library
2007 can enable or disable tracepoints. Amongst the possible conditions
2008 of an event rule is a filter expression which `liblttng-ust` evalutes
2009 when an event occurs.
2010 ** Share <<channel,channel>> attributes and ring buffer locations.
2013 The session daemon and the user space tracing library use a Unix
2014 domain socket for their communication.
2016 * The <<lttng-ust-agents,user space tracing agents>>.
2018 Any instance of a user space tracing agent first registers to
2019 a session daemon. Then, the session daemon can send requests to
2020 this instance, such as:
2023 ** Get the list of loggers.
2024 ** Enable or disable a specific logger.
2027 The session daemon and the user space tracing agent use a TCP connection
2028 for their communication.
2030 * The <<lttng-modules,LTTng kernel tracer>>.
2031 * The <<lttng-consumerd,consumer daemon>>.
2033 The session daemon sends requests to the consumer daemon to instruct
2034 it where to send the trace data streams, amongst other information.
2036 * The <<lttng-relayd,relay daemon>>.
2038 The session daemon receives commands from the
2039 <<liblttng-ctl-lttng,tracing control library>>.
2041 The root session daemon loads the appropriate
2042 <<lttng-modules,LTTng kernel modules>> on startup. It also spawns
2043 a <<lttng-consumerd,consumer daemon>> as soon as you create
2044 an <<event,event rule>>.
2046 The session daemon does not send and receive trace data: this is the
2047 role of the <<lttng-consumerd,consumer daemon>> and
2048 <<lttng-relayd,relay daemon>>. It does, however, generate the
2049 http://diamon.org/ctf/[CTF] metadata stream.
2051 Each Unix user can have its own session daemon instance. The
2052 tracing sessions managed by different session daemons are completely
2055 The root user's session daemon is the only one which is
2056 allowed to control the LTTng kernel tracer, and its spawned consumer
2057 daemon is the only one which is allowed to consume trace data from the
2058 LTTng kernel tracer. Note, however, that any Unix user which is a member
2059 of the <<tracing-group,tracing group>> is allowed
2060 to create <<channel,channels>> in the
2061 Linux kernel <<domain,tracing domain>>, and thus to trace the Linux
2064 The <<lttng-cli,cmd:lttng command-line tool>> automatically starts a
2065 session daemon when using its `create` command if none is currently
2066 running. You can also start the session daemon manually.
2073 .The consumer daemon.
2074 image::plumbing-consumerd.png[]
2076 The _consumer daemon_, cmd:lttng-consumerd, is a daemon which shares
2077 ring buffers with user applications or with the LTTng kernel modules to
2078 collect trace data and send it to some location (on disk or to a
2079 <<lttng-relayd,relay daemon>> over the network). The consumer daemon
2080 is part of LTTng-tools.
2082 You do not start a consumer daemon manually: a consumer daemon is always
2083 spawned by a <<lttng-sessiond,session daemon>> as soon as you create an
2084 <<event,event rule>>, that is, before you start tracing. When you kill
2085 its owner session daemon, the consumer daemon also exits because it is
2086 the session daemon's child process. Command-line options of
2087 man:lttng-sessiond(8) target the consumer daemon process.
2089 There are up to two running consumer daemons per Unix user, whereas only
2090 one session daemon can run per user. This is because each process can be
2091 either 32-bit or 64-bit: if the target system runs a mixture of 32-bit
2092 and 64-bit processes, it is more efficient to have separate
2093 corresponding 32-bit and 64-bit consumer daemons. The root user is an
2094 exception: it can have up to _three_ running consumer daemons: 32-bit
2095 and 64-bit instances for its user applications, and one more
2096 reserved for collecting kernel trace data.
2104 image::plumbing-relayd.png[]
2106 The _relay daemon_, man:lttng-relayd(8), is a daemon acting as a bridge
2107 between remote session and consumer daemons, local trace files, and a
2108 remote live trace viewer. The relay daemon is part of LTTng-tools.
2110 The main purpose of the relay daemon is to implement a receiver of
2111 <<sending-trace-data-over-the-network,trace data over the network>>.
2112 This is useful when the target system does not have much file system
2113 space to record trace files locally.
2115 The relay daemon is also a server to which a
2116 <<lttng-live,live trace viewer>> can
2117 connect. The live trace viewer sends requests to the relay daemon to
2118 receive trace data as the target system emits events. The
2119 communication protocol is named _LTTng live_; it is used over TCP
2122 Note that you can start the relay daemon on the target system directly.
2123 This is the setup of choice when the use case is to view events as
2124 the target system emits them without the need of a remote system.
2128 == [[using-lttng]]Instrumentation
2130 There are many examples of tracing and monitoring in our everyday life:
2132 * You have access to real-time and historical weather reports and
2133 forecasts thanks to weather stations installed around the country.
2134 * You know your heart is safe thanks to an electrocardiogram.
2135 * You make sure not to drive your car too fast and to have enough fuel
2136 to reach your destination thanks to gauges visible on your dashboard.
2138 All the previous examples have something in common: they rely on
2139 **instruments**. Without the electrodes attached to the surface of your
2140 body's skin, cardiac monitoring is futile.
2142 LTTng, as a tracer, is no different from those real life examples. If
2143 you're about to trace a software system or, in other words, record its
2144 history of execution, you better have **instrumentation points** in the
2145 subject you're tracing, that is, the actual software.
2147 Various ways were developed to instrument a piece of software for LTTng
2148 tracing. The most straightforward one is to manually place
2149 instrumentation points, called _tracepoints_, in the software's source
2150 code. It is also possible to add instrumentation points dynamically in
2151 the Linux kernel <<domain,tracing domain>>.
2153 If you're only interested in tracing the Linux kernel, your
2154 instrumentation needs are probably already covered by LTTng's built-in
2155 <<lttng-modules,Linux kernel tracepoints>>. You may also wish to trace a
2156 user application which is already instrumented for LTTng tracing.
2157 In such cases, you can skip this whole section and read the topics of
2158 the <<controlling-tracing,Tracing control>> section.
2160 Many methods are available to instrument a piece of software for LTTng
2163 * <<c-application,User space instrumentation for C and $$C++$$
2165 * <<prebuilt-ust-helpers,Prebuilt user space tracing helpers>>.
2166 * <<java-application,User space Java agent>>.
2167 * <<python-application,User space Python agent>>.
2168 * <<proc-lttng-logger-abi,LTTng logger>>.
2169 * <<instrumenting-linux-kernel,LTTng kernel tracepoints>>.
2173 === [[cxx-application]]User space instrumentation for C and $$C++$$ applications
2175 The procedure to instrument a C or $$C++$$ user application with
2176 the <<lttng-ust,LTTng user space tracing library>>, `liblttng-ust`, is:
2178 . <<tracepoint-provider,Create the source files of a tracepoint provider
2180 . <<probing-the-application-source-code,Add tracepoints to
2181 the application's source code>>.
2182 . <<building-tracepoint-providers-and-user-application,Build and link
2183 a tracepoint provider package and the user application>>.
2185 If you need quick, man:printf(3)-like instrumentation, you can skip
2186 those steps and use <<tracef,`tracef()`>> or <<tracelog,`tracelog()`>>
2189 IMPORTANT: You need to <<installing-lttng,install>> LTTng-UST to
2190 instrument a user application with `liblttng-ust`.
2193 [[tracepoint-provider]]
2194 ==== Create the source files of a tracepoint provider package
2196 A _tracepoint provider_ is a set of compiled functions which provide
2197 **tracepoints** to an application, the type of instrumentation point
2198 supported by LTTng-UST. Those functions can emit events with
2199 user-defined fields and serialize those events as event records to one
2200 or more LTTng-UST <<channel,channel>> sub-buffers. The `tracepoint()`
2201 macro, which you <<probing-the-application-source-code,insert in a user
2202 application's source code>>, calls those functions.
2204 A _tracepoint provider package_ is an object file (`.o`) or a shared
2205 library (`.so`) which contains one or more tracepoint providers.
2206 Its source files are:
2208 * One or more <<tpp-header,tracepoint provider header>> (`.h`).
2209 * A <<tpp-source,tracepoint provider package source>> (`.c`).
2211 A tracepoint provider package is dynamically linked with `liblttng-ust`,
2212 the LTTng user space tracer, at run time.
2215 .User application linked with `liblttng-ust` and containing a tracepoint provider.
2216 image::ust-app.png[]
2218 NOTE: If you need quick, man:printf(3)-like instrumentation, you can
2219 skip creating and using a tracepoint provider and use
2220 <<tracef,`tracef()`>> or <<tracelog,`tracelog()`>> instead.
2224 ===== Create a tracepoint provider header file template
2226 A _tracepoint provider header file_ contains the tracepoint
2227 definitions of a tracepoint provider.
2229 To create a tracepoint provider header file:
2231 . Start from this template:
2235 .Tracepoint provider header file template (`.h` file extension).
2237 #undef TRACEPOINT_PROVIDER
2238 #define TRACEPOINT_PROVIDER provider_name
2240 #undef TRACEPOINT_INCLUDE
2241 #define TRACEPOINT_INCLUDE "./tp.h"
2243 #if !defined(_TP_H) || defined(TRACEPOINT_HEADER_MULTI_READ)
2246 #include <lttng/tracepoint.h>
2249 * Use TRACEPOINT_EVENT(), TRACEPOINT_EVENT_CLASS(),
2250 * TRACEPOINT_EVENT_INSTANCE(), and TRACEPOINT_LOGLEVEL() here.
2255 #include <lttng/tracepoint-event.h>
2261 * `provider_name` with the name of your tracepoint provider.
2262 * `"tp.h"` with the name of your tracepoint provider header file.
2264 . Below the `#include <lttng/tracepoint.h>` line, put your
2265 <<defining-tracepoints,tracepoint definitions>>.
2267 Your tracepoint provider name must be unique amongst all the possible
2268 tracepoint provider names used on the same target system. We
2269 suggest to include the name of your project or company in the name,
2270 for example, `org_lttng_my_project_tpp`.
2272 TIP: [[lttng-gen-tp]]You can use the man:lttng-gen-tp(1) tool to create
2273 this boilerplate for you. When using cmd:lttng-gen-tp, all you need to
2274 write are the <<defining-tracepoints,tracepoint definitions>>.
2277 [[defining-tracepoints]]
2278 ===== Create a tracepoint definition
2280 A _tracepoint definition_ defines, for a given tracepoint:
2282 * Its **input arguments**. They are the macro parameters that the
2283 `tracepoint()` macro accepts for this particular tracepoint
2284 in the user application's source code.
2285 * Its **output event fields**. They are the sources of event fields
2286 that form the payload of any event that the execution of the
2287 `tracepoint()` macro emits for this particular tracepoint.
2289 You can create a tracepoint definition by using the
2290 `TRACEPOINT_EVENT()` macro below the `#include <lttng/tracepoint.h>`
2292 <<tpp-header,tracepoint provider header file template>>.
2294 The syntax of the `TRACEPOINT_EVENT()` macro is:
2297 .`TRACEPOINT_EVENT()` macro syntax.
2300 /* Tracepoint provider name */
2303 /* Tracepoint name */
2306 /* Input arguments */
2311 /* Output event fields */
2320 * `provider_name` with your tracepoint provider name.
2321 * `tracepoint_name` with your tracepoint name.
2322 * `arguments` with the <<tpp-def-input-args,input arguments>>.
2323 * `fields` with the <<tpp-def-output-fields,output event field>>
2326 This tracepoint emits events named `provider_name:tracepoint_name`.
2329 .Event name's length limitation
2331 The concatenation of the tracepoint provider name and the
2332 tracepoint name must not exceed **254 characters**. If it does, the
2333 instrumented application compiles and runs, but LTTng throws multiple
2334 warnings and you could experience serious issues.
2337 [[tpp-def-input-args]]The syntax of the `TP_ARGS()` macro is:
2340 .`TP_ARGS()` macro syntax.
2349 * `type` with the C type of the argument.
2350 * `arg_name` with the argument name.
2352 You can repeat `type` and `arg_name` up to 10 times to have
2353 more than one argument.
2355 .`TP_ARGS()` usage with three arguments.
2367 The `TP_ARGS()` and `TP_ARGS(void)` forms are valid to create a
2368 tracepoint definition with no input arguments.
2370 [[tpp-def-output-fields]]The `TP_FIELDS()` macro contains a list of
2371 `ctf_*()` macros. Each `ctf_*()` macro defines one event field. See
2372 man:lttng-ust(3) for a complete description of the available `ctf_*()`
2373 macros. A `ctf_*()` macro specifies the type, size, and byte order of
2376 Each `ctf_*()` macro takes an _argument expression_ parameter. This is a
2377 C expression that the tracer evalutes at the `tracepoint()` macro site
2378 in the application's source code. This expression provides a field's
2379 source of data. The argument expression can include input argument names
2380 listed in the `TP_ARGS()` macro.
2382 Each `ctf_*()` macro also takes a _field name_ parameter. Field names
2383 must be unique within a given tracepoint definition.
2385 Here's a complete tracepoint definition example:
2387 .Tracepoint definition.
2389 The following tracepoint definition defines a tracepoint which takes
2390 three input arguments and has four output event fields.
2394 #include "my-custom-structure.h"
2400 const struct my_custom_structure*, my_custom_structure,
2405 ctf_string(query_field, query)
2406 ctf_float(double, ratio_field, ratio)
2407 ctf_integer(int, recv_size, my_custom_structure->recv_size)
2408 ctf_integer(int, send_size, my_custom_structure->send_size)
2413 You can refer to this tracepoint definition with the `tracepoint()`
2414 macro in your application's source code like this:
2418 tracepoint(my_provider, my_tracepoint,
2419 my_structure, some_ratio, the_query);
2423 NOTE: The LTTng tracer only evaluates tracepoint arguments at run time
2424 if they satisfy an enabled <<event,event rule>>.
2427 [[using-tracepoint-classes]]
2428 ===== Use a tracepoint class
2430 A _tracepoint class_ is a class of tracepoints which share the same
2431 output event field definitions. A _tracepoint instance_ is one
2432 instance of such a defined tracepoint class, with its own tracepoint
2435 The <<defining-tracepoints,`TRACEPOINT_EVENT()` macro>> is actually a
2436 shorthand which defines both a tracepoint class and a tracepoint
2437 instance at the same time.
2439 When you build a tracepoint provider package, the C or $$C++$$ compiler
2440 creates one serialization function for each **tracepoint class**. A
2441 serialization function is responsible for serializing the event fields
2442 of a tracepoint to a sub-buffer when tracing.
2444 For various performance reasons, when your situation requires multiple
2445 tracepoint definitions with different names, but with the same event
2446 fields, we recommend that you manually create a tracepoint class
2447 and instantiate as many tracepoint instances as needed. One positive
2448 effect of such a design, amongst other advantages, is that all
2449 tracepoint instances of the same tracepoint class reuse the same
2450 serialization function, thus reducing
2451 https://en.wikipedia.org/wiki/Cache_pollution[cache pollution].
2453 .Use a tracepoint class and tracepoint instances.
2455 Consider the following three tracepoint definitions:
2467 ctf_integer(int, userid, userid)
2468 ctf_integer(size_t, len, len)
2480 ctf_integer(int, userid, userid)
2481 ctf_integer(size_t, len, len)
2493 ctf_integer(int, userid, userid)
2494 ctf_integer(size_t, len, len)
2499 In this case, we create three tracepoint classes, with one implicit
2500 tracepoint instance for each of them: `get_account`, `get_settings`, and
2501 `get_transaction`. However, they all share the same event field names
2502 and types. Hence three identical, yet independent serialization
2503 functions are created when you build the tracepoint provider package.
2505 A better design choice is to define a single tracepoint class and three
2506 tracepoint instances:
2510 /* The tracepoint class */
2511 TRACEPOINT_EVENT_CLASS(
2512 /* Tracepoint provider name */
2515 /* Tracepoint class name */
2518 /* Input arguments */
2524 /* Output event fields */
2526 ctf_integer(int, userid, userid)
2527 ctf_integer(size_t, len, len)
2531 /* The tracepoint instances */
2532 TRACEPOINT_EVENT_INSTANCE(
2533 /* Tracepoint provider name */
2536 /* Tracepoint class name */
2539 /* Tracepoint name */
2542 /* Input arguments */
2548 TRACEPOINT_EVENT_INSTANCE(
2557 TRACEPOINT_EVENT_INSTANCE(
2570 [[assigning-log-levels]]
2571 ===== Assign a log level to a tracepoint definition
2573 You can assign an optional _log level_ to a
2574 <<defining-tracepoints,tracepoint definition>>.
2576 Assigning different levels of severity to tracepoint definitions can
2577 be useful: when you <<enabling-disabling-events,create an event rule>>,
2578 you can target tracepoints having a log level as severe as a specific
2581 The concept of LTTng-UST log levels is similar to the levels found
2582 in typical logging frameworks:
2584 * In a logging framework, the log level is given by the function
2585 or method name you use at the log statement site: `debug()`,
2586 `info()`, `warn()`, `error()`, and so on.
2587 * In LTTng-UST, you statically assign the log level to a tracepoint
2588 definition; any `tracepoint()` macro invocation which refers to
2589 this definition has this log level.
2591 You can assign a log level to a tracepoint definition with the
2592 `TRACEPOINT_LOGLEVEL()` macro. You must use this macro _after_ the
2593 <<defining-tracepoints,`TRACEPOINT_EVENT()`>> or
2594 <<using-tracepoint-classes,`TRACEPOINT_INSTANCE()`>> macro for a given
2597 The syntax of the `TRACEPOINT_LOGLEVEL()` macro is:
2600 .`TRACEPOINT_LOGLEVEL()` macro syntax.
2602 TRACEPOINT_LOGLEVEL(provider_name, tracepoint_name, log_level)
2607 * `provider_name` with the tracepoint provider name.
2608 * `tracepoint_name` with the tracepoint name.
2609 * `log_level` with the log level to assign to the tracepoint
2610 definition named `tracepoint_name` in the `provider_name`
2611 tracepoint provider.
2613 See man:lttng-ust(3) for a list of available log level names.
2615 .Assign the `TRACE_DEBUG_UNIT` log level to a tracepoint definition.
2619 /* Tracepoint definition */
2628 ctf_integer(int, userid, userid)
2629 ctf_integer(size_t, len, len)
2633 /* Log level assignment */
2634 TRACEPOINT_LOGLEVEL(my_app, get_transaction, TRACE_DEBUG_UNIT)
2640 ===== Create a tracepoint provider package source file
2642 A _tracepoint provider package source file_ is a C source file which
2643 includes a <<tpp-header,tracepoint provider header file>> to expand its
2644 macros into event serialization and other functions.
2646 You can always use the following tracepoint provider package source
2650 .Tracepoint provider package source file template.
2652 #define TRACEPOINT_CREATE_PROBES
2657 Replace `tp.h` with the name of your <<tpp-header,tracepoint provider
2658 header file>> name. You may also include more than one tracepoint
2659 provider header file here to create a tracepoint provider package
2660 holding more than one tracepoint providers.
2663 [[probing-the-application-source-code]]
2664 ==== Add tracepoints to an application's source code
2666 Once you <<tpp-header,create a tracepoint provider header file>>, you
2667 can use the `tracepoint()` macro in your application's
2668 source code to insert the tracepoints that this header
2669 <<defining-tracepoints,defines>>.
2671 The `tracepoint()` macro takes at least two parameters: the tracepoint
2672 provider name and the tracepoint name. The corresponding tracepoint
2673 definition defines the other parameters.
2675 .`tracepoint()` usage.
2677 The following <<defining-tracepoints,tracepoint definition>> defines a
2678 tracepoint which takes two input arguments and has two output event
2682 .Tracepoint provider header file.
2684 #include "my-custom-structure.h"
2691 const char*, cmd_name
2694 ctf_string(cmd_name, cmd_name)
2695 ctf_integer(int, number_of_args, argc)
2700 You can refer to this tracepoint definition with the `tracepoint()`
2701 macro in your application's source code like this:
2704 .Application's source file.
2708 int main(int argc, char* argv[])
2710 tracepoint(my_provider, my_tracepoint, argc, argv[0]);
2716 Note how the application's source code includes
2717 the tracepoint provider header file containing the tracepoint
2718 definitions to use, path:{tp.h}.
2721 .`tracepoint()` usage with a complex tracepoint definition.
2723 Consider this complex tracepoint definition, where multiple event
2724 fields refer to the same input arguments in their argument expression
2728 .Tracepoint provider header file.
2730 /* For `struct stat` */
2731 #include <sys/types.h>
2732 #include <sys/stat.h>
2744 ctf_integer(int, my_constant_field, 23 + 17)
2745 ctf_integer(int, my_int_arg_field, my_int_arg)
2746 ctf_integer(int, my_int_arg_field2, my_int_arg * my_int_arg)
2747 ctf_integer(int, sum4_field, my_str_arg[0] + my_str_arg[1] +
2748 my_str_arg[2] + my_str_arg[3])
2749 ctf_string(my_str_arg_field, my_str_arg)
2750 ctf_integer_hex(off_t, size_field, st->st_size)
2751 ctf_float(double, size_dbl_field, (double) st->st_size)
2752 ctf_sequence_text(char, half_my_str_arg_field, my_str_arg,
2753 size_t, strlen(my_str_arg) / 2)
2758 You can refer to this tracepoint definition with the `tracepoint()`
2759 macro in your application's source code like this:
2762 .Application's source file.
2764 #define TRACEPOINT_DEFINE
2771 stat("/etc/fstab", &s);
2772 tracepoint(my_provider, my_tracepoint, 23, "Hello, World!", &s);
2778 If you look at the event record that LTTng writes when tracing this
2779 program, assuming the file size of path:{/etc/fstab} is 301{nbsp}bytes,
2780 it should look like this:
2782 .Event record fields
2784 |Field's name |Field's value
2785 |`my_constant_field` |40
2786 |`my_int_arg_field` |23
2787 |`my_int_arg_field2` |529
2789 |`my_str_arg_field` |`Hello, World!`
2790 |`size_field` |0x12d
2791 |`size_dbl_field` |301.0
2792 |`half_my_str_arg_field` |`Hello,`
2796 Sometimes, the arguments you pass to `tracepoint()` are expensive to
2797 compute--they use the call stack, for example. To avoid this
2798 computation when the tracepoint is disabled, you can use the
2799 `tracepoint_enabled()` and `do_tracepoint()` macros.
2801 The syntax of the `tracepoint_enabled()` and `do_tracepoint()` macros
2805 .`tracepoint_enabled()` and `do_tracepoint()` macros syntax.
2807 tracepoint_enabled(provider_name, tracepoint_name)
2808 do_tracepoint(provider_name, tracepoint_name, ...)
2813 * `provider_name` with the tracepoint provider name.
2814 * `tracepoint_name` with the tracepoint name.
2816 `tracepoint_enabled()` returns a non-zero value if the tracepoint named
2817 `tracepoint_name` from the provider named `provider_name` is enabled
2820 `do_tracepoint()` is like `tracepoint()`, except that it doesn't check
2821 if the tracepoint is enabled. Using `tracepoint()` with
2822 `tracepoint_enabled()` is dangerous since `tracepoint()` also contains
2823 the `tracepoint_enabled()` check, thus a race condition is
2824 possible in this situation:
2827 .Possible race condition when using `tracepoint_enabled()` with `tracepoint()`.
2829 if (tracepoint_enabled(my_provider, my_tracepoint)) {
2830 stuff = prepare_stuff();
2833 tracepoint(my_provider, my_tracepoint, stuff);
2836 If the tracepoint is enabled after the condition, then `stuff` is not
2837 prepared: the emitted event will either contain wrong data, or the whole
2838 application could crash (segmentation fault, for example).
2840 NOTE: Neither `tracepoint_enabled()` nor `do_tracepoint()` have an
2841 `STAP_PROBEV()` call. If you need it, you must emit
2845 [[building-tracepoint-providers-and-user-application]]
2846 ==== Build and link a tracepoint provider package and an application
2848 Once you have one or more <<tpp-header,tracepoint provider header
2849 files>> and a <<tpp-source,tracepoint provider package source file>>,
2850 you can create the tracepoint provider package by compiling its source
2851 file. From here, multiple build and run scenarios are possible. The
2852 following table shows common application and library configurations
2853 along with the required command lines to achieve them.
2855 In the following diagrams, we use the following file names:
2858 Executable application.
2861 Application's object file.
2864 Tracepoint provider package object file.
2867 Tracepoint provider package archive file.
2870 Tracepoint provider package shared object file.
2873 User library object file.
2876 User library shared object file.
2878 We use the following symbols in the diagrams of table below:
2881 .Symbols used in the build scenario diagrams.
2882 image::ust-sit-symbols.png[]
2884 We assume that path:{.} is part of the env:LD_LIBRARY_PATH environment
2885 variable in the following instructions.
2887 [role="growable ust-scenarios",cols="asciidoc,asciidoc"]
2888 .Common tracepoint provider package scenarios.
2890 |Scenario |Instructions
2893 The instrumented application is statically linked with
2894 the tracepoint provider package object.
2896 image::ust-sit+app-linked-with-tp-o+app-instrumented.png[]
2899 include::../common/ust-sit-step-tp-o.txt[]
2901 To build the instrumented application:
2903 . In path:{app.c}, before including path:{tpp.h}, add the following line:
2908 #define TRACEPOINT_DEFINE
2912 . Compile the application source file:
2921 . Build the application:
2926 $ gcc -o app app.o tpp.o -llttng-ust -ldl
2930 To run the instrumented application:
2932 * Start the application:
2942 The instrumented application is statically linked with the
2943 tracepoint provider package archive file.
2945 image::ust-sit+app-linked-with-tp-a+app-instrumented.png[]
2948 To create the tracepoint provider package archive file:
2950 . Compile the <<tpp-source,tracepoint provider package source file>>:
2959 . Create the tracepoint provider package archive file:
2964 $ ar rcs tpp.a tpp.o
2968 To build the instrumented application:
2970 . In path:{app.c}, before including path:{tpp.h}, add the following line:
2975 #define TRACEPOINT_DEFINE
2979 . Compile the application source file:
2988 . Build the application:
2993 $ gcc -o app app.o tpp.a -llttng-ust -ldl
2997 To run the instrumented application:
2999 * Start the application:
3009 The instrumented application is linked with the tracepoint provider
3010 package shared object.
3012 image::ust-sit+app-linked-with-tp-so+app-instrumented.png[]
3015 include::../common/ust-sit-step-tp-so.txt[]
3017 To build the instrumented application:
3019 . In path:{app.c}, before including path:{tpp.h}, add the following line:
3024 #define TRACEPOINT_DEFINE
3028 . Compile the application source file:
3037 . Build the application:
3042 $ gcc -o app app.o -ldl -L. -ltpp
3046 To run the instrumented application:
3048 * Start the application:
3058 The tracepoint provider package shared object is preloaded before the
3059 instrumented application starts.
3061 image::ust-sit+tp-so-preloaded+app-instrumented.png[]
3064 include::../common/ust-sit-step-tp-so.txt[]
3066 To build the instrumented application:
3068 . In path:{app.c}, before including path:{tpp.h}, add the
3074 #define TRACEPOINT_DEFINE
3075 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3079 . Compile the application source file:
3088 . Build the application:
3093 $ gcc -o app app.o -ldl
3097 To run the instrumented application with tracing support:
3099 * Preload the tracepoint provider package shared object and
3100 start the application:
3105 $ LD_PRELOAD=./libtpp.so ./app
3109 To run the instrumented application without tracing support:
3111 * Start the application:
3121 The instrumented application dynamically loads the tracepoint provider
3122 package shared object.
3124 See the <<dlclose-warning,warning about `dlclose()`>>.
3126 image::ust-sit+app-dlopens-tp-so+app-instrumented.png[]
3129 include::../common/ust-sit-step-tp-so.txt[]
3131 To build the instrumented application:
3133 . In path:{app.c}, before including path:{tpp.h}, add the
3139 #define TRACEPOINT_DEFINE
3140 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3144 . Compile the application source file:
3153 . Build the application:
3158 $ gcc -o app app.o -ldl
3162 To run the instrumented application:
3164 * Start the application:
3174 The application is linked with the instrumented user library.
3176 The instrumented user library is statically linked with the tracepoint
3177 provider package object file.
3179 image::ust-sit+app-linked-with-lib+lib-linked-with-tp-o+lib-instrumented.png[]
3182 include::../common/ust-sit-step-tp-o-fpic.txt[]
3184 To build the instrumented user library:
3186 . In path:{emon.c}, before including path:{tpp.h}, add the
3192 #define TRACEPOINT_DEFINE
3196 . Compile the user library source file:
3201 $ gcc -I. -fpic -c emon.c
3205 . Build the user library shared object:
3210 $ gcc -shared -o libemon.so emon.o tpp.o -llttng-ust -ldl
3214 To build the application:
3216 . Compile the application source file:
3225 . Build the application:
3230 $ gcc -o app app.o -L. -lemon
3234 To run the application:
3236 * Start the application:
3246 The application is linked with the instrumented user library.
3248 The instrumented user library is linked with the tracepoint provider
3249 package shared object.
3251 image::ust-sit+app-linked-with-lib+lib-linked-with-tp-so+lib-instrumented.png[]
3254 include::../common/ust-sit-step-tp-so.txt[]
3256 To build the instrumented user library:
3258 . In path:{emon.c}, before including path:{tpp.h}, add the
3264 #define TRACEPOINT_DEFINE
3268 . Compile the user library source file:
3273 $ gcc -I. -fpic -c emon.c
3277 . Build the user library shared object:
3282 $ gcc -shared -o libemon.so emon.o -ldl -L. -ltpp
3286 To build the application:
3288 . Compile the application source file:
3297 . Build the application:
3302 $ gcc -o app app.o -L. -lemon
3306 To run the application:
3308 * Start the application:
3318 The tracepoint provider package shared object is preloaded before the
3321 The application is linked with the instrumented user library.
3323 image::ust-sit+tp-so-preloaded+app-linked-with-lib+lib-instrumented.png[]
3326 include::../common/ust-sit-step-tp-so.txt[]
3328 To build the instrumented user library:
3330 . In path:{emon.c}, before including path:{tpp.h}, add the
3336 #define TRACEPOINT_DEFINE
3337 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3341 . Compile the user library source file:
3346 $ gcc -I. -fpic -c emon.c
3350 . Build the user library shared object:
3355 $ gcc -shared -o libemon.so emon.o -ldl
3359 To build the application:
3361 . Compile the application source file:
3370 . Build the application:
3375 $ gcc -o app app.o -L. -lemon
3379 To run the application with tracing support:
3381 * Preload the tracepoint provider package shared object and
3382 start the application:
3387 $ LD_PRELOAD=./libtpp.so ./app
3391 To run the application without tracing support:
3393 * Start the application:
3403 The application is linked with the instrumented user library.
3405 The instrumented user library dynamically loads the tracepoint provider
3406 package shared object.
3408 See the <<dlclose-warning,warning about `dlclose()`>>.
3410 image::ust-sit+app-linked-with-lib+lib-dlopens-tp-so+lib-instrumented.png[]
3413 include::../common/ust-sit-step-tp-so.txt[]
3415 To build the instrumented user library:
3417 . In path:{emon.c}, before including path:{tpp.h}, add the
3423 #define TRACEPOINT_DEFINE
3424 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3428 . Compile the user library source file:
3433 $ gcc -I. -fpic -c emon.c
3437 . Build the user library shared object:
3442 $ gcc -shared -o libemon.so emon.o -ldl
3446 To build the application:
3448 . Compile the application source file:
3457 . Build the application:
3462 $ gcc -o app app.o -L. -lemon
3466 To run the application:
3468 * Start the application:
3478 The application dynamically loads the instrumented user library.
3480 The instrumented user library is linked with the tracepoint provider
3481 package shared object.
3483 See the <<dlclose-warning,warning about `dlclose()`>>.
3485 image::ust-sit+app-dlopens-lib+lib-linked-with-tp-so+lib-instrumented.png[]
3488 include::../common/ust-sit-step-tp-so.txt[]
3490 To build the instrumented user library:
3492 . In path:{emon.c}, before including path:{tpp.h}, add the
3498 #define TRACEPOINT_DEFINE
3502 . Compile the user library source file:
3507 $ gcc -I. -fpic -c emon.c
3511 . Build the user library shared object:
3516 $ gcc -shared -o libemon.so emon.o -ldl -L. -ltpp
3520 To build the application:
3522 . Compile the application source file:
3531 . Build the application:
3536 $ gcc -o app app.o -ldl -L. -lemon
3540 To run the application:
3542 * Start the application:
3552 The application dynamically loads the instrumented user library.
3554 The instrumented user library dynamically loads the tracepoint provider
3555 package shared object.
3557 See the <<dlclose-warning,warning about `dlclose()`>>.
3559 image::ust-sit+app-dlopens-lib+lib-dlopens-tp-so+lib-instrumented.png[]
3562 include::../common/ust-sit-step-tp-so.txt[]
3564 To build the instrumented user library:
3566 . In path:{emon.c}, before including path:{tpp.h}, add the
3572 #define TRACEPOINT_DEFINE
3573 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3577 . Compile the user library source file:
3582 $ gcc -I. -fpic -c emon.c
3586 . Build the user library shared object:
3591 $ gcc -shared -o libemon.so emon.o -ldl
3595 To build the application:
3597 . Compile the application source file:
3606 . Build the application:
3611 $ gcc -o app app.o -ldl -L. -lemon
3615 To run the application:
3617 * Start the application:
3627 The tracepoint provider package shared object is preloaded before the
3630 The application dynamically loads the instrumented user library.
3632 image::ust-sit+tp-so-preloaded+app-dlopens-lib+lib-instrumented.png[]
3635 include::../common/ust-sit-step-tp-so.txt[]
3637 To build the instrumented user library:
3639 . In path:{emon.c}, before including path:{tpp.h}, add the
3645 #define TRACEPOINT_DEFINE
3646 #define TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3650 . Compile the user library source file:
3655 $ gcc -I. -fpic -c emon.c
3659 . Build the user library shared object:
3664 $ gcc -shared -o libemon.so emon.o -ldl
3668 To build the application:
3670 . Compile the application source file:
3679 . Build the application:
3684 $ gcc -o app app.o -L. -lemon
3688 To run the application with tracing support:
3690 * Preload the tracepoint provider package shared object and
3691 start the application:
3696 $ LD_PRELOAD=./libtpp.so ./app
3700 To run the application without tracing support:
3702 * Start the application:
3712 The application is statically linked with the tracepoint provider
3713 package object file.
3715 The application is linked with the instrumented user library.
3717 image::ust-sit+app-linked-with-tp-o+app-linked-with-lib+lib-instrumented.png[]
3720 include::../common/ust-sit-step-tp-o.txt[]
3722 To build the instrumented user library:
3724 . In path:{emon.c}, before including path:{tpp.h}, add the
3730 #define TRACEPOINT_DEFINE
3734 . Compile the user library source file:
3739 $ gcc -I. -fpic -c emon.c
3743 . Build the user library shared object:
3748 $ gcc -shared -o libemon.so emon.o
3752 To build the application:
3754 . Compile the application source file:
3763 . Build the application:
3768 $ gcc -o app app.o tpp.o -llttng-ust -ldl -L. -lemon
3772 To run the instrumented application:
3774 * Start the application:
3784 The application is statically linked with the tracepoint provider
3785 package object file.
3787 The application dynamically loads the instrumented user library.
3789 image::ust-sit+app-linked-with-tp-o+app-dlopens-lib+lib-instrumented.png[]
3792 include::../common/ust-sit-step-tp-o.txt[]
3794 To build the application:
3796 . In path:{app.c}, before including path:{tpp.h}, add the following line:
3801 #define TRACEPOINT_DEFINE
3805 . Compile the application source file:
3814 . Build the application:
3819 $ gcc -Wl,--export-dynamic -o app app.o tpp.o \
3824 The `--export-dynamic` option passed to the linker is necessary for the
3825 dynamically loaded library to ``see'' the tracepoint symbols defined in
3828 To build the instrumented user library:
3830 . Compile the user library source file:
3835 $ gcc -I. -fpic -c emon.c
3839 . Build the user library shared object:
3844 $ gcc -shared -o libemon.so emon.o
3848 To run the application:
3850 * Start the application:
3862 .Do not use man:dlclose(3) on a tracepoint provider package
3864 Never use man:dlclose(3) on any shared object which:
3866 * Is linked with, statically or dynamically, a tracepoint provider
3868 * Calls man:dlopen(3) itself to dynamically open a tracepoint provider
3869 package shared object.
3871 This is currently considered **unsafe** due to a lack of reference
3872 counting from LTTng-UST to the shared object.
3874 A known workaround (available since glibc 2.2) is to use the
3875 `RTLD_NODELETE` flag when calling man:dlopen(3) initially. This has the
3876 effect of not unloading the loaded shared object, even if man:dlclose(3)
3879 You can also preload the tracepoint provider package shared object with
3880 the env:LD_PRELOAD environment variable to overcome this limitation.
3884 [[using-lttng-ust-with-daemons]]
3885 ===== Use noch:{LTTng-UST} with daemons
3887 If your instrumented application calls man:fork(2), man:clone(2),
3888 or BSD's man:rfork(2), without a following man:exec(3)-family
3889 system call, you must preload the path:{liblttng-ust-fork.so} shared
3890 object when you start the application.
3894 $ LD_PRELOAD=liblttng-ust-fork.so ./my-app
3897 If your tracepoint provider package is
3898 a shared library which you also preload, you must put both
3899 shared objects in env:LD_PRELOAD:
3903 $ LD_PRELOAD=liblttng-ust-fork.so:/path/to/tp.so ./my-app
3909 ===== Use noch:{LTTng-UST} with applications which close file descriptors that don't belong to them
3911 If your instrumented application closes one or more file descriptors
3912 which it did not open itself, you must preload the
3913 path:{liblttng-ust-fd.so} shared object when you start the application:
3917 $ LD_PRELOAD=liblttng-ust-fd.so ./my-app
3920 Typical use cases include closing all the file descriptors after
3921 man:fork(2) or man:rfork(2) and buggy applications doing
3925 [[lttng-ust-pkg-config]]
3926 ===== Use noch:{pkg-config}
3928 On some distributions, LTTng-UST ships with a
3929 https://www.freedesktop.org/wiki/Software/pkg-config/[pkg-config]
3930 metadata file. If this is your case, then you can use cmd:pkg-config to
3931 build an application on the command line:
3935 $ gcc -o my-app my-app.o tp.o $(pkg-config --cflags --libs lttng-ust)
3939 [[instrumenting-32-bit-app-on-64-bit-system]]
3940 ===== [[advanced-instrumenting-techniques]]Build a 32-bit instrumented application for a 64-bit target system
3942 In order to trace a 32-bit application running on a 64-bit system,
3943 LTTng must use a dedicated 32-bit
3944 <<lttng-consumerd,consumer daemon>>.
3946 The following steps show how to build and install a 32-bit consumer
3947 daemon, which is _not_ part of the default 64-bit LTTng build, how to
3948 build and install the 32-bit LTTng-UST libraries, and how to build and
3949 link an instrumented 32-bit application in that context.
3951 To build a 32-bit instrumented application for a 64-bit target system,
3952 assuming you have a fresh target system with no installed Userspace RCU
3955 . Download, build, and install a 32-bit version of Userspace RCU:
3960 $ cd $(mktemp -d) &&
3961 wget http://lttng.org/files/urcu/userspace-rcu-latest-0.9.tar.bz2 &&
3962 tar -xf userspace-rcu-latest-0.9.tar.bz2 &&
3963 cd userspace-rcu-0.9.* &&
3964 ./configure --libdir=/usr/local/lib32 CFLAGS=-m32 &&
3966 sudo make install &&
3971 . Using your distribution's package manager, or from source, install
3972 the following 32-bit versions of the following dependencies of
3973 LTTng-tools and LTTng-UST:
3976 * https://sourceforge.net/projects/libuuid/[libuuid]
3977 * http://directory.fsf.org/wiki/Popt[popt]
3978 * http://www.xmlsoft.org/[libxml2]
3981 . Download, build, and install a 32-bit version of the latest
3982 LTTng-UST{nbsp}{revision}:
3987 $ cd $(mktemp -d) &&
3988 wget http://lttng.org/files/lttng-ust/lttng-ust-latest-2.10.tar.bz2 &&
3989 tar -xf lttng-ust-latest-2.10.tar.bz2 &&
3990 cd lttng-ust-2.10.* &&
3991 ./configure --libdir=/usr/local/lib32 \
3992 CFLAGS=-m32 CXXFLAGS=-m32 \
3993 LDFLAGS='-L/usr/local/lib32 -L/usr/lib32' &&
3995 sudo make install &&
4002 Depending on your distribution,
4003 32-bit libraries could be installed at a different location than
4004 `/usr/lib32`. For example, Debian is known to install
4005 some 32-bit libraries in `/usr/lib/i386-linux-gnu`.
4007 In this case, make sure to set `LDFLAGS` to all the
4008 relevant 32-bit library paths, for example:
4012 $ LDFLAGS='-L/usr/lib/i386-linux-gnu -L/usr/lib32'
4016 . Download the latest LTTng-tools{nbsp}{revision}, build, and install
4017 the 32-bit consumer daemon:
4022 $ cd $(mktemp -d) &&
4023 wget http://lttng.org/files/lttng-tools/lttng-tools-latest-2.10.tar.bz2 &&
4024 tar -xf lttng-tools-latest-2.10.tar.bz2 &&
4025 cd lttng-tools-2.10.* &&
4026 ./configure --libdir=/usr/local/lib32 CFLAGS=-m32 CXXFLAGS=-m32 \
4027 LDFLAGS='-L/usr/local/lib32 -L/usr/lib32' \
4028 --disable-bin-lttng --disable-bin-lttng-crash \
4029 --disable-bin-lttng-relayd --disable-bin-lttng-sessiond &&
4031 cd src/bin/lttng-consumerd &&
4032 sudo make install &&
4037 . From your distribution or from source,
4038 <<installing-lttng,install>> the 64-bit versions of
4039 LTTng-UST and Userspace RCU.
4040 . Download, build, and install the 64-bit version of the
4041 latest LTTng-tools{nbsp}{revision}:
4046 $ cd $(mktemp -d) &&
4047 wget http://lttng.org/files/lttng-tools/lttng-tools-latest-2.10.tar.bz2 &&
4048 tar -xf lttng-tools-latest-2.10.tar.bz2 &&
4049 cd lttng-tools-2.10.* &&
4050 ./configure --with-consumerd32-libdir=/usr/local/lib32 \
4051 --with-consumerd32-bin=/usr/local/lib32/lttng/libexec/lttng-consumerd &&
4053 sudo make install &&
4058 . Pass the following options to man:gcc(1), man:g++(1), or man:clang(1)
4059 when linking your 32-bit application:
4062 -m32 -L/usr/lib32 -L/usr/local/lib32 \
4063 -Wl,-rpath,/usr/lib32,-rpath,/usr/local/lib32
4066 For example, let's rebuild the quick start example in
4067 <<tracing-your-own-user-application,Trace a user application>> as an
4068 instrumented 32-bit application:
4073 $ gcc -m32 -c -I. hello-tp.c
4074 $ gcc -m32 -c hello.c
4075 $ gcc -m32 -o hello hello.o hello-tp.o \
4076 -L/usr/lib32 -L/usr/local/lib32 \
4077 -Wl,-rpath,/usr/lib32,-rpath,/usr/local/lib32 \
4082 No special action is required to execute the 32-bit application and
4083 to trace it: use the command-line man:lttng(1) tool as usual.
4090 man:tracef(3) is a small LTTng-UST API designed for quick,
4091 man:printf(3)-like instrumentation without the burden of
4092 <<tracepoint-provider,creating>> and
4093 <<building-tracepoint-providers-and-user-application,building>>
4094 a tracepoint provider package.
4096 To use `tracef()` in your application:
4098 . In the C or C++ source files where you need to use `tracef()`,
4099 include `<lttng/tracef.h>`:
4104 #include <lttng/tracef.h>
4108 . In the application's source code, use `tracef()` like you would use
4116 tracef("my message: %d (%s)", my_integer, my_string);
4122 . Link your application with `liblttng-ust`:
4127 $ gcc -o app app.c -llttng-ust
4131 To trace the events that `tracef()` calls emit:
4133 * <<enabling-disabling-events,Create an event rule>> which matches the
4134 `lttng_ust_tracef:*` event name:
4139 $ lttng enable-event --userspace 'lttng_ust_tracef:*'
4144 .Limitations of `tracef()`
4146 The `tracef()` utility function was developed to make user space tracing
4147 super simple, albeit with notable disadvantages compared to
4148 <<defining-tracepoints,user-defined tracepoints>>:
4150 * All the emitted events have the same tracepoint provider and
4151 tracepoint names, respectively `lttng_ust_tracef` and `event`.
4152 * There is no static type checking.
4153 * The only event record field you actually get, named `msg`, is a string
4154 potentially containing the values you passed to `tracef()`
4155 using your own format string. This also means that you cannot filter
4156 events with a custom expression at run time because there are no
4158 * Since `tracef()` uses the C standard library's man:vasprintf(3)
4159 function behind the scenes to format the strings at run time, its
4160 expected performance is lower than with user-defined tracepoints,
4161 which do not require a conversion to a string.
4163 Taking this into consideration, `tracef()` is useful for some quick
4164 prototyping and debugging, but you should not consider it for any
4165 permanent and serious applicative instrumentation.
4171 ==== Use `tracelog()`
4173 The man:tracelog(3) API is very similar to <<tracef,`tracef()`>>, with
4174 the difference that it accepts an additional log level parameter.
4176 The goal of `tracelog()` is to ease the migration from logging to
4179 To use `tracelog()` in your application:
4181 . In the C or C++ source files where you need to use `tracelog()`,
4182 include `<lttng/tracelog.h>`:
4187 #include <lttng/tracelog.h>
4191 . In the application's source code, use `tracelog()` like you would use
4192 man:printf(3), except for the first parameter which is the log
4200 tracelog(TRACE_WARNING, "my message: %d (%s)",
4201 my_integer, my_string);
4207 See man:lttng-ust(3) for a list of available log level names.
4209 . Link your application with `liblttng-ust`:
4214 $ gcc -o app app.c -llttng-ust
4218 To trace the events that `tracelog()` calls emit with a log level
4219 _as severe as_ a specific log level:
4221 * <<enabling-disabling-events,Create an event rule>> which matches the
4222 `lttng_ust_tracelog:*` event name and a minimum level
4228 $ lttng enable-event --userspace 'lttng_ust_tracelog:*'
4229 --loglevel=TRACE_WARNING
4233 To trace the events that `tracelog()` calls emit with a
4234 _specific log level_:
4236 * Create an event rule which matches the `lttng_ust_tracelog:*`
4237 event name and a specific log level:
4242 $ lttng enable-event --userspace 'lttng_ust_tracelog:*'
4243 --loglevel-only=TRACE_INFO
4248 [[prebuilt-ust-helpers]]
4249 === Prebuilt user space tracing helpers
4251 The LTTng-UST package provides a few helpers in the form or preloadable
4252 shared objects which automatically instrument system functions and
4255 The helper shared objects are normally found in dir:{/usr/lib}. If you
4256 built LTTng-UST <<building-from-source,from source>>, they are probably
4257 located in dir:{/usr/local/lib}.
4259 The installed user space tracing helpers in LTTng-UST{nbsp}{revision}
4262 path:{liblttng-ust-libc-wrapper.so}::
4263 path:{liblttng-ust-pthread-wrapper.so}::
4264 <<liblttng-ust-libc-pthread-wrapper,C{nbsp}standard library
4265 memory and POSIX threads function tracing>>.
4267 path:{liblttng-ust-cyg-profile.so}::
4268 path:{liblttng-ust-cyg-profile-fast.so}::
4269 <<liblttng-ust-cyg-profile,Function entry and exit tracing>>.
4271 path:{liblttng-ust-dl.so}::
4272 <<liblttng-ust-dl,Dynamic linker tracing>>.
4274 To use a user space tracing helper with any user application:
4276 * Preload the helper shared object when you start the application:
4281 $ LD_PRELOAD=liblttng-ust-libc-wrapper.so my-app
4285 You can preload more than one helper:
4290 $ LD_PRELOAD=liblttng-ust-libc-wrapper.so:liblttng-ust-dl.so my-app
4296 [[liblttng-ust-libc-pthread-wrapper]]
4297 ==== Instrument C standard library memory and POSIX threads functions
4299 The path:{liblttng-ust-libc-wrapper.so} and
4300 path:{liblttng-ust-pthread-wrapper.so} helpers
4301 add instrumentation to some C standard library and POSIX
4305 .Functions instrumented by preloading path:{liblttng-ust-libc-wrapper.so}.
4307 |TP provider name |TP name |Instrumented function
4309 .6+|`lttng_ust_libc` |`malloc` |man:malloc(3)
4310 |`calloc` |man:calloc(3)
4311 |`realloc` |man:realloc(3)
4312 |`free` |man:free(3)
4313 |`memalign` |man:memalign(3)
4314 |`posix_memalign` |man:posix_memalign(3)
4318 .Functions instrumented by preloading path:{liblttng-ust-pthread-wrapper.so}.
4320 |TP provider name |TP name |Instrumented function
4322 .4+|`lttng_ust_pthread` |`pthread_mutex_lock_req` |man:pthread_mutex_lock(3p) (request time)
4323 |`pthread_mutex_lock_acq` |man:pthread_mutex_lock(3p) (acquire time)
4324 |`pthread_mutex_trylock` |man:pthread_mutex_trylock(3p)
4325 |`pthread_mutex_unlock` |man:pthread_mutex_unlock(3p)
4328 When you preload the shared object, it replaces the functions listed
4329 in the previous tables by wrappers which contain tracepoints and call
4330 the replaced functions.
4333 [[liblttng-ust-cyg-profile]]
4334 ==== Instrument function entry and exit
4336 The path:{liblttng-ust-cyg-profile*.so} helpers can add instrumentation
4337 to the entry and exit points of functions.
4339 man:gcc(1) and man:clang(1) have an option named
4340 https://gcc.gnu.org/onlinedocs/gcc/Instrumentation-Options.html[`-finstrument-functions`]
4341 which generates instrumentation calls for entry and exit to functions.
4342 The LTTng-UST function tracing helpers,
4343 path:{liblttng-ust-cyg-profile.so} and
4344 path:{liblttng-ust-cyg-profile-fast.so}, take advantage of this feature
4345 to add tracepoints to the two generated functions (which contain
4346 `cyg_profile` in their names, hence the helper's name).
4348 To use the LTTng-UST function tracing helper, the source files to
4349 instrument must be built using the `-finstrument-functions` compiler
4352 There are two versions of the LTTng-UST function tracing helper:
4354 * **path:{liblttng-ust-cyg-profile-fast.so}** is a lightweight variant
4355 that you should only use when it can be _guaranteed_ that the
4356 complete event stream is recorded without any lost event record.
4357 Any kind of duplicate information is left out.
4359 Assuming no event record is lost, having only the function addresses on
4360 entry is enough to create a call graph, since an event record always
4361 contains the ID of the CPU that generated it.
4363 You can use a tool like man:addr2line(1) to convert function addresses
4364 back to source file names and line numbers.
4366 * **path:{liblttng-ust-cyg-profile.so}** is a more robust variant
4367 which also works in use cases where event records might get discarded or
4368 not recorded from application startup.
4369 In these cases, the trace analyzer needs more information to be
4370 able to reconstruct the program flow.
4372 See man:lttng-ust-cyg-profile(3) to learn more about the instrumentation
4373 points of this helper.
4375 All the tracepoints that this helper provides have the
4376 log level `TRACE_DEBUG_FUNCTION` (see man:lttng-ust(3)).
4378 TIP: It's sometimes a good idea to limit the number of source files that
4379 you compile with the `-finstrument-functions` option to prevent LTTng
4380 from writing an excessive amount of trace data at run time. When using
4381 man:gcc(1), you can use the
4382 `-finstrument-functions-exclude-function-list` option to avoid
4383 instrument entries and exits of specific function names.
4388 ==== Instrument the dynamic linker
4390 The path:{liblttng-ust-dl.so} helper adds instrumentation to the
4391 man:dlopen(3) and man:dlclose(3) function calls.
4393 See man:lttng-ust-dl(3) to learn more about the instrumentation points
4398 [[java-application]]
4399 === User space Java agent
4401 You can instrument any Java application which uses one of the following
4404 * The https://docs.oracle.com/javase/7/docs/api/java/util/logging/package-summary.html[**`java.util.logging`**]
4405 (JUL) core logging facilities.
4406 * http://logging.apache.org/log4j/1.2/[**Apache log4j 1.2**], since
4407 LTTng 2.6. Note that Apache Log4j{nbsp}2 is not supported.
4410 .LTTng-UST Java agent imported by a Java application.
4411 image::java-app.png[]
4413 Note that the methods described below are new in LTTng{nbsp}{revision}.
4414 Previous LTTng versions use another technique.
4416 NOTE: We use http://openjdk.java.net/[OpenJDK]{nbsp}8 for development
4417 and https://ci.lttng.org/[continuous integration], thus this version is
4418 directly supported. However, the LTTng-UST Java agent is also tested
4419 with OpenJDK{nbsp}7.
4424 ==== Use the LTTng-UST Java agent for `java.util.logging`
4426 To use the LTTng-UST Java agent in a Java application which uses
4427 `java.util.logging` (JUL):
4429 . In the Java application's source code, import the LTTng-UST
4430 log handler package for `java.util.logging`:
4435 import org.lttng.ust.agent.jul.LttngLogHandler;
4439 . Create an LTTng-UST JUL log handler:
4444 Handler lttngUstLogHandler = new LttngLogHandler();
4448 . Add this handler to the JUL loggers which should emit LTTng events:
4453 Logger myLogger = Logger.getLogger("some-logger");
4455 myLogger.addHandler(lttngUstLogHandler);
4459 . Use `java.util.logging` log statements and configuration as usual.
4460 The loggers with an attached LTTng-UST log handler can emit
4463 . Before exiting the application, remove the LTTng-UST log handler from
4464 the loggers attached to it and call its `close()` method:
4469 myLogger.removeHandler(lttngUstLogHandler);
4470 lttngUstLogHandler.close();
4474 This is not strictly necessary, but it is recommended for a clean
4475 disposal of the handler's resources.
4477 . Include the LTTng-UST Java agent's common and JUL-specific JAR files,
4478 path:{lttng-ust-agent-common.jar} and path:{lttng-ust-agent-jul.jar},
4480 https://docs.oracle.com/javase/tutorial/essential/environment/paths.html[class
4481 path] when you build the Java application.
4483 The JAR files are typically located in dir:{/usr/share/java}.
4485 IMPORTANT: The LTTng-UST Java agent must be
4486 <<installing-lttng,installed>> for the logging framework your
4489 .Use the LTTng-UST Java agent for `java.util.logging`.
4494 import java.io.IOException;
4495 import java.util.logging.Handler;
4496 import java.util.logging.Logger;
4497 import org.lttng.ust.agent.jul.LttngLogHandler;
4501 private static final int answer = 42;
4503 public static void main(String[] argv) throws Exception
4506 Logger logger = Logger.getLogger("jello");
4508 // Create an LTTng-UST log handler
4509 Handler lttngUstLogHandler = new LttngLogHandler();
4511 // Add the LTTng-UST log handler to our logger
4512 logger.addHandler(lttngUstLogHandler);
4515 logger.info("some info");
4516 logger.warning("some warning");
4518 logger.finer("finer information; the answer is " + answer);
4520 logger.severe("error!");
4522 // Not mandatory, but cleaner
4523 logger.removeHandler(lttngUstLogHandler);
4524 lttngUstLogHandler.close();
4533 $ javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar Test.java
4536 <<creating-destroying-tracing-sessions,Create a tracing session>>,
4537 <<enabling-disabling-events,create an event rule>> matching the
4538 `jello` JUL logger, and <<basic-tracing-session-control,start tracing>>:
4543 $ lttng enable-event --jul jello
4547 Run the compiled class:
4551 $ java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar:. Test
4554 <<basic-tracing-session-control,Stop tracing>> and inspect the
4564 In the resulting trace, an <<event,event record>> generated by a Java
4565 application using `java.util.logging` is named `lttng_jul:event` and
4566 has the following fields:
4569 Log record's message.
4575 Name of the class in which the log statement was executed.
4578 Name of the method in which the log statement was executed.
4581 Logging time (timestamp in milliseconds).
4584 Log level integer value.
4587 ID of the thread in which the log statement was executed.
4589 You can use the opt:lttng-enable-event(1):--loglevel or
4590 opt:lttng-enable-event(1):--loglevel-only option of the
4591 man:lttng-enable-event(1) command to target a range of JUL log levels
4592 or a specific JUL log level.
4597 ==== Use the LTTng-UST Java agent for Apache log4j
4599 To use the LTTng-UST Java agent in a Java application which uses
4602 . In the Java application's source code, import the LTTng-UST
4603 log appender package for Apache log4j:
4608 import org.lttng.ust.agent.log4j.LttngLogAppender;
4612 . Create an LTTng-UST log4j log appender:
4617 Appender lttngUstLogAppender = new LttngLogAppender();
4621 . Add this appender to the log4j loggers which should emit LTTng events:
4626 Logger myLogger = Logger.getLogger("some-logger");
4628 myLogger.addAppender(lttngUstLogAppender);
4632 . Use Apache log4j log statements and configuration as usual. The
4633 loggers with an attached LTTng-UST log appender can emit LTTng events.
4635 . Before exiting the application, remove the LTTng-UST log appender from
4636 the loggers attached to it and call its `close()` method:
4641 myLogger.removeAppender(lttngUstLogAppender);
4642 lttngUstLogAppender.close();
4646 This is not strictly necessary, but it is recommended for a clean
4647 disposal of the appender's resources.
4649 . Include the LTTng-UST Java agent's common and log4j-specific JAR
4650 files, path:{lttng-ust-agent-common.jar} and
4651 path:{lttng-ust-agent-log4j.jar}, in the
4652 https://docs.oracle.com/javase/tutorial/essential/environment/paths.html[class
4653 path] when you build the Java application.
4655 The JAR files are typically located in dir:{/usr/share/java}.
4657 IMPORTANT: The LTTng-UST Java agent must be
4658 <<installing-lttng,installed>> for the logging framework your
4661 .Use the LTTng-UST Java agent for Apache log4j.
4666 import org.apache.log4j.Appender;
4667 import org.apache.log4j.Logger;
4668 import org.lttng.ust.agent.log4j.LttngLogAppender;
4672 private static final int answer = 42;
4674 public static void main(String[] argv) throws Exception
4677 Logger logger = Logger.getLogger("jello");
4679 // Create an LTTng-UST log appender
4680 Appender lttngUstLogAppender = new LttngLogAppender();
4682 // Add the LTTng-UST log appender to our logger
4683 logger.addAppender(lttngUstLogAppender);
4686 logger.info("some info");
4687 logger.warn("some warning");
4689 logger.debug("debug information; the answer is " + answer);
4691 logger.fatal("error!");
4693 // Not mandatory, but cleaner
4694 logger.removeAppender(lttngUstLogAppender);
4695 lttngUstLogAppender.close();
4701 Build this example (`$LOG4JPATH` is the path to the Apache log4j JAR
4706 $ javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-log4j.jar:$LOG4JPATH Test.java
4709 <<creating-destroying-tracing-sessions,Create a tracing session>>,
4710 <<enabling-disabling-events,create an event rule>> matching the
4711 `jello` log4j logger, and <<basic-tracing-session-control,start tracing>>:
4716 $ lttng enable-event --log4j jello
4720 Run the compiled class:
4724 $ java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-log4j.jar:$LOG4JPATH:. Test
4727 <<basic-tracing-session-control,Stop tracing>> and inspect the
4737 In the resulting trace, an <<event,event record>> generated by a Java
4738 application using log4j is named `lttng_log4j:event` and
4739 has the following fields:
4742 Log record's message.
4748 Name of the class in which the log statement was executed.
4751 Name of the method in which the log statement was executed.
4754 Name of the file in which the executed log statement is located.
4757 Line number at which the log statement was executed.
4763 Log level integer value.
4766 Name of the Java thread in which the log statement was executed.
4768 You can use the opt:lttng-enable-event(1):--loglevel or
4769 opt:lttng-enable-event(1):--loglevel-only option of the
4770 man:lttng-enable-event(1) command to target a range of Apache log4j log levels
4771 or a specific log4j log level.
4775 [[java-application-context]]
4776 ==== Provide application-specific context fields in a Java application
4778 A Java application-specific context field is a piece of state provided
4779 by the application which <<adding-context,you can add>>, using the
4780 man:lttng-add-context(1) command, to each <<event,event record>>
4781 produced by the log statements of this application.
4783 For example, a given object might have a current request ID variable.
4784 You can create a context information retriever for this object and
4785 assign a name to this current request ID. You can then, using the
4786 man:lttng-add-context(1) command, add this context field by name to
4787 the JUL or log4j <<channel,channel>>.
4789 To provide application-specific context fields in a Java application:
4791 . In the Java application's source code, import the LTTng-UST
4792 Java agent context classes and interfaces:
4797 import org.lttng.ust.agent.context.ContextInfoManager;
4798 import org.lttng.ust.agent.context.IContextInfoRetriever;
4802 . Create a context information retriever class, that is, a class which
4803 implements the `IContextInfoRetriever` interface:
4808 class MyContextInfoRetriever implements IContextInfoRetriever
4811 public Object retrieveContextInfo(String key)
4813 if (key.equals("intCtx")) {
4815 } else if (key.equals("strContext")) {
4816 return "context value!";
4825 This `retrieveContextInfo()` method is the only member of the
4826 `IContextInfoRetriever` interface. Its role is to return the current
4827 value of a state by name to create a context field. The names of the
4828 context fields and which state variables they return depends on your
4831 All primitive types and objects are supported as context fields.
4832 When `retrieveContextInfo()` returns an object, the context field
4833 serializer calls its `toString()` method to add a string field to
4834 event records. The method can also return `null`, which means that
4835 no context field is available for the required name.
4837 . Register an instance of your context information retriever class to
4838 the context information manager singleton:
4843 IContextInfoRetriever cir = new MyContextInfoRetriever();
4844 ContextInfoManager cim = ContextInfoManager.getInstance();
4845 cim.registerContextInfoRetriever("retrieverName", cir);
4849 . Before exiting the application, remove your context information
4850 retriever from the context information manager singleton:
4855 ContextInfoManager cim = ContextInfoManager.getInstance();
4856 cim.unregisterContextInfoRetriever("retrieverName");
4860 This is not strictly necessary, but it is recommended for a clean
4861 disposal of some manager's resources.
4863 . Build your Java application with LTTng-UST Java agent support as
4864 usual, following the procedure for either the <<jul,JUL>> or
4865 <<log4j,Apache log4j>> framework.
4868 .Provide application-specific context fields in a Java application.
4873 import java.util.logging.Handler;
4874 import java.util.logging.Logger;
4875 import org.lttng.ust.agent.jul.LttngLogHandler;
4876 import org.lttng.ust.agent.context.ContextInfoManager;
4877 import org.lttng.ust.agent.context.IContextInfoRetriever;
4881 // Our context information retriever class
4882 private static class MyContextInfoRetriever
4883 implements IContextInfoRetriever
4886 public Object retrieveContextInfo(String key) {
4887 if (key.equals("intCtx")) {
4889 } else if (key.equals("strContext")) {
4890 return "context value!";
4897 private static final int answer = 42;
4899 public static void main(String args[]) throws Exception
4901 // Get the context information manager instance
4902 ContextInfoManager cim = ContextInfoManager.getInstance();
4904 // Create and register our context information retriever
4905 IContextInfoRetriever cir = new MyContextInfoRetriever();
4906 cim.registerContextInfoRetriever("myRetriever", cir);
4909 Logger logger = Logger.getLogger("jello");
4911 // Create an LTTng-UST log handler
4912 Handler lttngUstLogHandler = new LttngLogHandler();
4914 // Add the LTTng-UST log handler to our logger
4915 logger.addHandler(lttngUstLogHandler);
4918 logger.info("some info");
4919 logger.warning("some warning");
4921 logger.finer("finer information; the answer is " + answer);
4923 logger.severe("error!");
4925 // Not mandatory, but cleaner
4926 logger.removeHandler(lttngUstLogHandler);
4927 lttngUstLogHandler.close();
4928 cim.unregisterContextInfoRetriever("myRetriever");
4937 $ javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar Test.java
4940 <<creating-destroying-tracing-sessions,Create a tracing session>>
4941 and <<enabling-disabling-events,create an event rule>> matching the
4947 $ lttng enable-event --jul jello
4950 <<adding-context,Add the application-specific context fields>> to the
4955 $ lttng add-context --jul --type='$app.myRetriever:intCtx'
4956 $ lttng add-context --jul --type='$app.myRetriever:strContext'
4959 <<basic-tracing-session-control,Start tracing>>:
4966 Run the compiled class:
4970 $ java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar:. Test
4973 <<basic-tracing-session-control,Stop tracing>> and inspect the
4985 [[python-application]]
4986 === User space Python agent
4988 You can instrument a Python 2 or Python 3 application which uses the
4989 standard https://docs.python.org/3/library/logging.html[`logging`]
4992 Each log statement emits an LTTng event once the
4993 application module imports the
4994 <<lttng-ust-agents,LTTng-UST Python agent>> package.
4997 .A Python application importing the LTTng-UST Python agent.
4998 image::python-app.png[]
5000 To use the LTTng-UST Python agent:
5002 . In the Python application's source code, import the LTTng-UST Python
5012 The LTTng-UST Python agent automatically adds its logging handler to the
5013 root logger at import time.
5015 Any log statement that the application executes before this import does
5016 not emit an LTTng event.
5018 IMPORTANT: The LTTng-UST Python agent must be
5019 <<installing-lttng,installed>>.
5021 . Use log statements and logging configuration as usual.
5022 Since the LTTng-UST Python agent adds a handler to the _root_
5023 logger, you can trace any log statement from any logger.
5025 .Use the LTTng-UST Python agent.
5036 logging.basicConfig()
5037 logger = logging.getLogger('my-logger')
5040 logger.debug('debug message')
5041 logger.info('info message')
5042 logger.warn('warn message')
5043 logger.error('error message')
5044 logger.critical('critical message')
5048 if __name__ == '__main__':
5052 NOTE: `logging.basicConfig()`, which adds to the root logger a basic
5053 logging handler which prints to the standard error stream, is not
5054 strictly required for LTTng-UST tracing to work, but in versions of
5055 Python preceding 3.2, you could see a warning message which indicates
5056 that no handler exists for the logger `my-logger`.
5058 <<creating-destroying-tracing-sessions,Create a tracing session>>,
5059 <<enabling-disabling-events,create an event rule>> matching the
5060 `my-logger` Python logger, and <<basic-tracing-session-control,start
5066 $ lttng enable-event --python my-logger
5070 Run the Python script:
5077 <<basic-tracing-session-control,Stop tracing>> and inspect the recorded
5087 In the resulting trace, an <<event,event record>> generated by a Python
5088 application is named `lttng_python:event` and has the following fields:
5091 Logging time (string).
5094 Log record's message.
5100 Name of the function in which the log statement was executed.
5103 Line number at which the log statement was executed.
5106 Log level integer value.
5109 ID of the Python thread in which the log statement was executed.
5112 Name of the Python thread in which the log statement was executed.
5114 You can use the opt:lttng-enable-event(1):--loglevel or
5115 opt:lttng-enable-event(1):--loglevel-only option of the
5116 man:lttng-enable-event(1) command to target a range of Python log levels
5117 or a specific Python log level.
5119 When an application imports the LTTng-UST Python agent, the agent tries
5120 to register to a <<lttng-sessiond,session daemon>>. Note that you must
5121 <<start-sessiond,start the session daemon>> _before_ you run the Python
5122 application. If a session daemon is found, the agent tries to register
5123 to it during 5{nbsp}seconds, after which the application continues
5124 without LTTng tracing support. You can override this timeout value with
5125 the env:LTTNG_UST_PYTHON_REGISTER_TIMEOUT environment variable
5128 If the session daemon stops while a Python application with an imported
5129 LTTng-UST Python agent runs, the agent retries to connect and to
5130 register to a session daemon every 3{nbsp}seconds. You can override this
5131 delay with the env:LTTNG_UST_PYTHON_REGISTER_RETRY_DELAY environment
5136 [[proc-lttng-logger-abi]]
5139 The `lttng-tracer` Linux kernel module, part of
5140 <<lttng-modules,LTTng-modules>>, creates the special LTTng logger file
5141 path:{/proc/lttng-logger} when it's loaded. Any application can write
5142 text data to this file to emit an LTTng event.
5145 .An application writes to the LTTng logger file to emit an LTTng event.
5146 image::lttng-logger.png[]
5148 The LTTng logger is the quickest method--not the most efficient,
5149 however--to add instrumentation to an application. It is designed
5150 mostly to instrument shell scripts:
5154 $ echo "Some message, some $variable" > /proc/lttng-logger
5157 Any event that the LTTng logger emits is named `lttng_logger` and
5158 belongs to the Linux kernel <<domain,tracing domain>>. However, unlike
5159 other instrumentation points in the kernel tracing domain, **any Unix
5160 user** can <<enabling-disabling-events,create an event rule>> which
5161 matches its event name, not only the root user or users in the
5162 <<tracing-group,tracing group>>.
5164 To use the LTTng logger:
5166 * From any application, write text data to the path:{/proc/lttng-logger}
5169 The `msg` field of `lttng_logger` event records contains the
5172 NOTE: The maximum message length of an LTTng logger event is
5173 1024{nbsp}bytes. Writing more than this makes the LTTng logger emit more
5174 than one event to contain the remaining data.
5176 You should not use the LTTng logger to trace a user application which
5177 can be instrumented in a more efficient way, namely:
5179 * <<c-application,C and $$C++$$ applications>>.
5180 * <<java-application,Java applications>>.
5181 * <<python-application,Python applications>>.
5183 .Use the LTTng logger.
5188 echo 'Hello, World!' > /proc/lttng-logger
5190 df --human-readable --print-type / > /proc/lttng-logger
5193 <<creating-destroying-tracing-sessions,Create a tracing session>>,
5194 <<enabling-disabling-events,create an event rule>> matching the
5195 `lttng_logger` Linux kernel tracepoint, and
5196 <<basic-tracing-session-control,start tracing>>:
5201 $ lttng enable-event --kernel lttng_logger
5205 Run the Bash script:
5212 <<basic-tracing-session-control,Stop tracing>> and inspect the recorded
5223 [[instrumenting-linux-kernel]]
5224 === LTTng kernel tracepoints
5226 NOTE: This section shows how to _add_ instrumentation points to the
5227 Linux kernel. The kernel's subsystems are already thoroughly
5228 instrumented at strategic places for LTTng when you
5229 <<installing-lttng,install>> the <<lttng-modules,LTTng-modules>>
5233 There are two methods to instrument the Linux kernel:
5235 . <<linux-add-lttng-layer,Add an LTTng layer>> over an existing ftrace
5236 tracepoint which uses the `TRACE_EVENT()` API.
5238 Choose this if you want to instrumentation a Linux kernel tree with an
5239 instrumentation point compatible with ftrace, perf, and SystemTap.
5241 . Use an <<linux-lttng-tracepoint-event,LTTng-only approach>> to
5242 instrument an out-of-tree kernel module.
5244 Choose this if you don't need ftrace, perf, or SystemTap support.
5248 [[linux-add-lttng-layer]]
5249 ==== [[instrumenting-linux-kernel-itself]][[mainline-trace-event]][[lttng-adaptation-layer]]Add an LTTng layer to an existing ftrace tracepoint
5251 This section shows how to add an LTTng layer to existing ftrace
5252 instrumentation using the `TRACE_EVENT()` API.
5254 This section does not document the `TRACE_EVENT()` macro. You can
5255 read the following articles to learn more about this API:
5257 * http://lwn.net/Articles/379903/[Using the TRACE_EVENT() macro (Part 1)]
5258 * http://lwn.net/Articles/381064/[Using the TRACE_EVENT() macro (Part 2)]
5259 * http://lwn.net/Articles/383362/[Using the TRACE_EVENT() macro (Part 3)]
5261 The following procedure assumes that your ftrace tracepoints are
5262 correctly defined in their own header and that they are created in
5263 one source file using the `CREATE_TRACE_POINTS` definition.
5265 To add an LTTng layer over an existing ftrace tracepoint:
5267 . Make sure the following kernel configuration options are
5273 * `CONFIG_HIGH_RES_TIMERS`
5274 * `CONFIG_TRACEPOINTS`
5277 . Build the Linux source tree with your custom ftrace tracepoints.
5278 . Boot the resulting Linux image on your target system.
5280 Confirm that the tracepoints exist by looking for their names in the
5281 dir:{/sys/kernel/debug/tracing/events/subsys} directory, where `subsys`
5282 is your subsystem's name.
5284 . Get a copy of the latest LTTng-modules{nbsp}{revision}:
5289 $ cd $(mktemp -d) &&
5290 wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.10.tar.bz2 &&
5291 tar -xf lttng-modules-latest-2.10.tar.bz2 &&
5292 cd lttng-modules-2.10.*
5296 . In dir:{instrumentation/events/lttng-module}, relative to the root
5297 of the LTTng-modules source tree, create a header file named
5298 +__subsys__.h+ for your custom subsystem +__subsys__+ and write your
5299 LTTng-modules tracepoint definitions using the LTTng-modules
5302 Start with this template:
5306 .path:{instrumentation/events/lttng-module/my_subsys.h}
5309 #define TRACE_SYSTEM my_subsys
5311 #if !defined(_LTTNG_MY_SUBSYS_H) || defined(TRACE_HEADER_MULTI_READ)
5312 #define _LTTNG_MY_SUBSYS_H
5314 #include "../../../probes/lttng-tracepoint-event.h"
5315 #include <linux/tracepoint.h>
5317 LTTNG_TRACEPOINT_EVENT(
5319 * Format is identical to TRACE_EVENT()'s version for the three
5320 * following macro parameters:
5323 TP_PROTO(int my_int, const char *my_string),
5324 TP_ARGS(my_int, my_string),
5326 /* LTTng-modules specific macros */
5328 ctf_integer(int, my_int_field, my_int)
5329 ctf_string(my_bar_field, my_bar)
5333 #endif /* !defined(_LTTNG_MY_SUBSYS_H) || defined(TRACE_HEADER_MULTI_READ) */
5335 #include "../../../probes/define_trace.h"
5339 The entries in the `TP_FIELDS()` section are the list of fields for the
5340 LTTng tracepoint. This is similar to the `TP_STRUCT__entry()` part of
5341 ftrace's `TRACE_EVENT()` macro.
5343 See <<lttng-modules-tp-fields,Tracepoint fields macros>> for a
5344 complete description of the available `ctf_*()` macros.
5346 . Create the LTTng-modules probe's kernel module C source file,
5347 +probes/lttng-probe-__subsys__.c+, where +__subsys__+ is your
5352 .path:{probes/lttng-probe-my-subsys.c}
5354 #include <linux/module.h>
5355 #include "../lttng-tracer.h"
5358 * Build-time verification of mismatch between mainline
5359 * TRACE_EVENT() arguments and the LTTng-modules adaptation
5360 * layer LTTNG_TRACEPOINT_EVENT() arguments.
5362 #include <trace/events/my_subsys.h>
5364 /* Create LTTng tracepoint probes */
5365 #define LTTNG_PACKAGE_BUILD
5366 #define CREATE_TRACE_POINTS
5367 #define TRACE_INCLUDE_PATH ../instrumentation/events/lttng-module
5369 #include "../instrumentation/events/lttng-module/my_subsys.h"
5371 MODULE_LICENSE("GPL and additional rights");
5372 MODULE_AUTHOR("Your name <your-email>");
5373 MODULE_DESCRIPTION("LTTng my_subsys probes");
5374 MODULE_VERSION(__stringify(LTTNG_MODULES_MAJOR_VERSION) "."
5375 __stringify(LTTNG_MODULES_MINOR_VERSION) "."
5376 __stringify(LTTNG_MODULES_PATCHLEVEL_VERSION)
5377 LTTNG_MODULES_EXTRAVERSION);
5381 . Edit path:{probes/KBuild} and add your new kernel module object
5382 next to the existing ones:
5386 .path:{probes/KBuild}
5390 obj-m += lttng-probe-module.o
5391 obj-m += lttng-probe-power.o
5393 obj-m += lttng-probe-my-subsys.o
5399 . Build and install the LTTng kernel modules:
5404 $ make KERNELDIR=/path/to/linux
5405 # make modules_install && depmod -a
5409 Replace `/path/to/linux` with the path to the Linux source tree where
5410 you defined and used tracepoints with ftrace's `TRACE_EVENT()` macro.
5412 Note that you can also use the
5413 <<lttng-tracepoint-event-code,`LTTNG_TRACEPOINT_EVENT_CODE()` macro>>
5414 instead of `LTTNG_TRACEPOINT_EVENT()` to use custom local variables and
5415 C code that need to be executed before the event fields are recorded.
5417 The best way to learn how to use the previous LTTng-modules macros is to
5418 inspect the existing LTTng-modules tracepoint definitions in the
5419 dir:{instrumentation/events/lttng-module} header files. Compare them
5420 with the Linux kernel mainline versions in the
5421 dir:{include/trace/events} directory of the Linux source tree.
5425 [[lttng-tracepoint-event-code]]
5426 ===== Use custom C code to access the data for tracepoint fields
5428 Although we recommended to always use the
5429 <<lttng-adaptation-layer,`LTTNG_TRACEPOINT_EVENT()`>> macro to describe
5430 the arguments and fields of an LTTng-modules tracepoint when possible,
5431 sometimes you need a more complex process to access the data that the
5432 tracer records as event record fields. In other words, you need local
5433 variables and multiple C{nbsp}statements instead of simple
5434 argument-based expressions that you pass to the
5435 <<lttng-modules-tp-fields,`ctf_*()` macros of `TP_FIELDS()`>>.
5437 You can use the `LTTNG_TRACEPOINT_EVENT_CODE()` macro instead of
5438 `LTTNG_TRACEPOINT_EVENT()` to declare custom local variables and define
5439 a block of C{nbsp}code to be executed before LTTng records the fields.
5440 The structure of this macro is:
5443 .`LTTNG_TRACEPOINT_EVENT_CODE()` macro syntax.
5445 LTTNG_TRACEPOINT_EVENT_CODE(
5447 * Format identical to the LTTNG_TRACEPOINT_EVENT()
5448 * version for the following three macro parameters:
5451 TP_PROTO(int my_int, const char *my_string),
5452 TP_ARGS(my_int, my_string),
5454 /* Declarations of custom local variables */
5457 unsigned long b = 0;
5458 const char *name = "(undefined)";
5459 struct my_struct *my_struct;
5463 * Custom code which uses both tracepoint arguments
5464 * (in TP_ARGS()) and local variables (in TP_locvar()).
5466 * Local variables are actually members of a structure pointed
5467 * to by the special variable tp_locvar.
5471 tp_locvar->a = my_int + 17;
5472 tp_locvar->my_struct = get_my_struct_at(tp_locvar->a);
5473 tp_locvar->b = my_struct_compute_b(tp_locvar->my_struct);
5474 tp_locvar->name = my_struct_get_name(tp_locvar->my_struct);
5475 put_my_struct(tp_locvar->my_struct);
5484 * Format identical to the LTTNG_TRACEPOINT_EVENT()
5485 * version for this, except that tp_locvar members can be
5486 * used in the argument expression parameters of
5487 * the ctf_*() macros.
5490 ctf_integer(unsigned long, my_struct_b, tp_locvar->b)
5491 ctf_integer(int, my_struct_a, tp_locvar->a)
5492 ctf_string(my_string_field, my_string)
5493 ctf_string(my_struct_name, tp_locvar->name)
5498 IMPORTANT: The C code defined in `TP_code()` must not have any side
5499 effects when executed. In particular, the code must not allocate
5500 memory or get resources without deallocating this memory or putting
5501 those resources afterwards.
5504 [[instrumenting-linux-kernel-tracing]]
5505 ==== Load and unload a custom probe kernel module
5507 You must load a <<lttng-adaptation-layer,created LTTng-modules probe
5508 kernel module>> in the kernel before it can emit LTTng events.
5510 To load the default probe kernel modules and a custom probe kernel
5513 * Use the opt:lttng-sessiond(8):--extra-kmod-probes option to give extra
5514 probe modules to load when starting a root <<lttng-sessiond,session
5518 .Load the `my_subsys`, `usb`, and the default probe modules.
5522 # lttng-sessiond --extra-kmod-probes=my_subsys,usb
5527 You only need to pass the subsystem name, not the whole kernel module
5530 To load _only_ a given custom probe kernel module:
5532 * Use the opt:lttng-sessiond(8):--kmod-probes option to give the probe
5533 modules to load when starting a root session daemon:
5536 .Load only the `my_subsys` and `usb` probe modules.
5540 # lttng-sessiond --kmod-probes=my_subsys,usb
5545 To confirm that a probe module is loaded:
5552 $ lsmod | grep lttng_probe_usb
5556 To unload the loaded probe modules:
5558 * Kill the session daemon with `SIGTERM`:
5563 # pkill lttng-sessiond
5567 You can also use man:modprobe(8)'s `--remove` option if the session
5568 daemon terminates abnormally.
5571 [[controlling-tracing]]
5574 Once an application or a Linux kernel is
5575 <<instrumenting,instrumented>> for LTTng tracing,
5578 This section is divided in topics on how to use the various
5579 <<plumbing,components of LTTng>>, in particular the <<lttng-cli,cmd:lttng
5580 command-line tool>>, to _control_ the LTTng daemons and tracers.
5582 NOTE: In the following subsections, we refer to an man:lttng(1) command
5583 using its man page name. For example, instead of _Run the `create`
5584 command to..._, we use _Run the man:lttng-create(1) command to..._.
5588 === Start a session daemon
5590 In some situations, you need to run a <<lttng-sessiond,session daemon>>
5591 (man:lttng-sessiond(8)) _before_ you can use the man:lttng(1)
5594 You will see the following error when you run a command while no session
5598 Error: No session daemon is available
5601 The only command that automatically runs a session daemon is
5602 man:lttng-create(1), which you use to
5603 <<creating-destroying-tracing-sessions,create a tracing session>>. While
5604 this is most of the time the first operation that you do, sometimes it's
5605 not. Some examples are:
5607 * <<list-instrumentation-points,List the available instrumentation points>>.
5608 * <<saving-loading-tracing-session,Load a tracing session configuration>>.
5610 [[tracing-group]] Each Unix user must have its own running session
5611 daemon to trace user applications. The session daemon that the root user
5612 starts is the only one allowed to control the LTTng kernel tracer. Users
5613 that are part of the _tracing group_ can control the root session
5614 daemon. The default tracing group name is `tracing`; you can set it to
5615 something else with the opt:lttng-sessiond(8):--group option when you
5616 start the root session daemon.
5618 To start a user session daemon:
5620 * Run man:lttng-sessiond(8):
5625 $ lttng-sessiond --daemonize
5629 To start the root session daemon:
5631 * Run man:lttng-sessiond(8) as the root user:
5636 # lttng-sessiond --daemonize
5640 In both cases, remove the opt:lttng-sessiond(8):--daemonize option to
5641 start the session daemon in foreground.
5643 To stop a session daemon, use man:kill(1) on its process ID (standard
5646 Note that some Linux distributions could manage the LTTng session daemon
5647 as a service. In this case, you should use the service manager to
5648 start, restart, and stop session daemons.
5651 [[creating-destroying-tracing-sessions]]
5652 === Create and destroy a tracing session
5654 Almost all the LTTng control operations happen in the scope of
5655 a <<tracing-session,tracing session>>, which is the dialogue between the
5656 <<lttng-sessiond,session daemon>> and you.
5658 To create a tracing session with a generated name:
5660 * Use the man:lttng-create(1) command:
5669 The created tracing session's name is `auto` followed by the
5672 To create a tracing session with a specific name:
5674 * Use the optional argument of the man:lttng-create(1) command:
5679 $ lttng create my-session
5683 Replace `my-session` with the specific tracing session name.
5685 LTTng appends the creation date to the created tracing session's name.
5687 LTTng writes the traces of a tracing session in
5688 +$LTTNG_HOME/lttng-trace/__name__+ by default, where +__name__+ is the
5689 name of the tracing session. Note that the env:LTTNG_HOME environment
5690 variable defaults to `$HOME` if not set.
5692 To output LTTng traces to a non-default location:
5694 * Use the opt:lttng-create(1):--output option of the man:lttng-create(1) command:
5699 $ lttng create my-session --output=/tmp/some-directory
5703 You may create as many tracing sessions as you wish.
5705 To list all the existing tracing sessions for your Unix user:
5707 * Use the man:lttng-list(1) command:
5716 When you create a tracing session, it is set as the _current tracing
5717 session_. The following man:lttng(1) commands operate on the current
5718 tracing session when you don't specify one:
5720 [role="list-3-cols"]
5737 To change the current tracing session:
5739 * Use the man:lttng-set-session(1) command:
5744 $ lttng set-session new-session
5748 Replace `new-session` by the name of the new current tracing session.
5750 When you are done tracing in a given tracing session, you can destroy
5751 it. This operation frees the resources taken by the tracing session
5752 to destroy; it does not destroy the trace data that LTTng wrote for
5753 this tracing session.
5755 To destroy the current tracing session:
5757 * Use the man:lttng-destroy(1) command:
5767 [[list-instrumentation-points]]
5768 === List the available instrumentation points
5770 The <<lttng-sessiond,session daemon>> can query the running instrumented
5771 user applications and the Linux kernel to get a list of available
5772 instrumentation points. For the Linux kernel <<domain,tracing domain>>,
5773 they are tracepoints and system calls. For the user space tracing
5774 domain, they are tracepoints. For the other tracing domains, they are
5777 To list the available instrumentation points:
5779 * Use the man:lttng-list(1) command with the requested tracing domain's
5783 * opt:lttng-list(1):--kernel: Linux kernel tracepoints (your Unix user
5784 must be a root user, or it must be a member of the
5785 <<tracing-group,tracing group>>).
5786 * opt:lttng-list(1):--kernel with opt:lttng-list(1):--syscall: Linux
5787 kernel system calls (your Unix user must be a root user, or it must be
5788 a member of the tracing group).
5789 * opt:lttng-list(1):--userspace: user space tracepoints.
5790 * opt:lttng-list(1):--jul: `java.util.logging` loggers.
5791 * opt:lttng-list(1):--log4j: Apache log4j loggers.
5792 * opt:lttng-list(1):--python: Python loggers.
5795 .List the available user space tracepoints.
5799 $ lttng list --userspace
5803 .List the available Linux kernel system call tracepoints.
5807 $ lttng list --kernel --syscall
5812 [[enabling-disabling-events]]
5813 === Create and enable an event rule
5815 Once you <<creating-destroying-tracing-sessions,create a tracing
5816 session>>, you can create <<event,event rules>> with the
5817 man:lttng-enable-event(1) command.
5819 You specify each condition with a command-line option. The available
5820 condition options are shown in the following table.
5822 [role="growable",cols="asciidoc,asciidoc,default"]
5823 .Condition command-line options for the man:lttng-enable-event(1) command.
5825 |Option |Description |Applicable tracing domains
5831 . +--probe=__ADDR__+
5832 . +--function=__ADDR__+
5835 Instead of using the default _tracepoint_ instrumentation type, use:
5837 . A Linux system call.
5838 . A Linux https://lwn.net/Articles/132196/[KProbe] (symbol or address).
5839 . The entry and return points of a Linux function (symbol or address).
5843 |First positional argument.
5846 Tracepoint or system call name. In the case of a Linux KProbe or
5847 function, this is a custom name given to the event rule. With the
5848 JUL, log4j, and Python domains, this is a logger name.
5850 With a tracepoint, logger, or system call name, you can use the special
5851 `*` globbing character to match anything (for example, `sched_*`,
5859 . +--loglevel=__LEVEL__+
5860 . +--loglevel-only=__LEVEL__+
5863 . Match only tracepoints or log statements with a logging level at
5864 least as severe as +__LEVEL__+.
5865 . Match only tracepoints or log statements with a logging level
5866 equal to +__LEVEL__+.
5868 See man:lttng-enable-event(1) for the list of available logging level
5871 |User space, JUL, log4j, and Python.
5873 |+--exclude=__EXCLUSIONS__+
5876 When you use a `*` character at the end of the tracepoint or logger
5877 name (first positional argument), exclude the specific names in the
5878 comma-delimited list +__EXCLUSIONS__+.
5881 User space, JUL, log4j, and Python.
5883 |+--filter=__EXPR__+
5886 Match only events which satisfy the expression +__EXPR__+.
5888 See man:lttng-enable-event(1) to learn more about the syntax of a
5895 You attach an event rule to a <<channel,channel>> on creation. If you do
5896 not specify the channel with the opt:lttng-enable-event(1):--channel
5897 option, and if the event rule to create is the first in its
5898 <<domain,tracing domain>> for a given tracing session, then LTTng
5899 creates a _default channel_ for you. This default channel is reused in
5900 subsequent invocations of the man:lttng-enable-event(1) command for the
5901 same tracing domain.
5903 An event rule is always enabled at creation time.
5905 The following examples show how you can combine the previous
5906 command-line options to create simple to more complex event rules.
5908 .Create an event rule targetting a Linux kernel tracepoint (default channel).
5912 $ lttng enable-event --kernel sched_switch
5916 .Create an event rule matching four Linux kernel system calls (default channel).
5920 $ lttng enable-event --kernel --syscall open,write,read,close
5924 .Create event rules matching tracepoints with filter expressions (default channel).
5928 $ lttng enable-event --kernel sched_switch --filter='prev_comm == "bash"'
5933 $ lttng enable-event --kernel --all \
5934 --filter='$ctx.tid == 1988 || $ctx.tid == 1534'
5939 $ lttng enable-event --jul my_logger \
5940 --filter='$app.retriever:cur_msg_id > 3'
5943 IMPORTANT: Make sure to always quote the filter string when you
5944 use man:lttng(1) from a shell.
5947 .Create an event rule matching any user space tracepoint of a given tracepoint provider with a log level range (default channel).
5951 $ lttng enable-event --userspace my_app:'*' --loglevel=TRACE_INFO
5954 IMPORTANT: Make sure to always quote the wildcard character when you
5955 use man:lttng(1) from a shell.
5958 .Create an event rule matching multiple Python loggers with a wildcard and with exclusions (default channel).
5962 $ lttng enable-event --python my-app.'*' \
5963 --exclude='my-app.module,my-app.hello'
5967 .Create an event rule matching any Apache log4j logger with a specific log level (default channel).
5971 $ lttng enable-event --log4j --all --loglevel-only=LOG4J_WARN
5975 .Create an event rule attached to a specific channel matching a specific user space tracepoint provider and tracepoint.
5979 $ lttng enable-event --userspace my_app:my_tracepoint --channel=my-channel
5983 The event rules of a given channel form a whitelist: as soon as an
5984 emitted event passes one of them, LTTng can record the event. For
5985 example, an event named `my_app:my_tracepoint` emitted from a user space
5986 tracepoint with a `TRACE_ERROR` log level passes both of the following
5991 $ lttng enable-event --userspace my_app:my_tracepoint
5992 $ lttng enable-event --userspace my_app:my_tracepoint \
5993 --loglevel=TRACE_INFO
5996 The second event rule is redundant: the first one includes
6000 [[disable-event-rule]]
6001 === Disable an event rule
6003 To disable an event rule that you <<enabling-disabling-events,created>>
6004 previously, use the man:lttng-disable-event(1) command. This command
6005 disables _all_ the event rules (of a given tracing domain and channel)
6006 which match an instrumentation point. The other conditions are not
6007 supported as of LTTng{nbsp}{revision}.
6009 The LTTng tracer does not record an emitted event which passes
6010 a _disabled_ event rule.
6012 .Disable an event rule matching a Python logger (default channel).
6016 $ lttng disable-event --python my-logger
6020 .Disable an event rule matching all `java.util.logging` loggers (default channel).
6024 $ lttng disable-event --jul '*'
6028 .Disable _all_ the event rules of the default channel.
6030 The opt:lttng-disable-event(1):--all-events option is not, like the
6031 opt:lttng-enable-event(1):--all option of man:lttng-enable-event(1), the
6032 equivalent of the event name `*` (wildcard): it disables _all_ the event
6033 rules of a given channel.
6037 $ lttng disable-event --jul --all-events
6041 NOTE: You cannot delete an event rule once you create it.
6045 === Get the status of a tracing session
6047 To get the status of the current tracing session, that is, its
6048 parameters, its channels, event rules, and their attributes:
6050 * Use the man:lttng-status(1) command:
6060 To get the status of any tracing session:
6062 * Use the man:lttng-list(1) command with the tracing session's name:
6067 $ lttng list my-session
6071 Replace `my-session` with the desired tracing session's name.
6074 [[basic-tracing-session-control]]
6075 === Start and stop a tracing session
6077 Once you <<creating-destroying-tracing-sessions,create a tracing
6079 <<enabling-disabling-events,create one or more event rules>>,
6080 you can start and stop the tracers for this tracing session.
6082 To start tracing in the current tracing session:
6084 * Use the man:lttng-start(1) command:
6093 LTTng is very flexible: you can launch user applications before
6094 or after the you start the tracers. The tracers only record the events
6095 if they pass enabled event rules and if they occur while the tracers are
6098 To stop tracing in the current tracing session:
6100 * Use the man:lttng-stop(1) command:
6109 If there were <<channel-overwrite-mode-vs-discard-mode,lost event
6110 records>> or lost sub-buffers since the last time you ran
6111 man:lttng-start(1), warnings are printed when you run the
6112 man:lttng-stop(1) command.
6115 [[enabling-disabling-channels]]
6116 === Create a channel
6118 Once you create a tracing session, you can create a <<channel,channel>>
6119 with the man:lttng-enable-channel(1) command.
6121 Note that LTTng automatically creates a default channel when, for a
6122 given <<domain,tracing domain>>, no channels exist and you
6123 <<enabling-disabling-events,create>> the first event rule. This default
6124 channel is named `channel0` and its attributes are set to reasonable
6125 values. Therefore, you only need to create a channel when you need
6126 non-default attributes.
6128 You specify each non-default channel attribute with a command-line
6129 option when you use the man:lttng-enable-channel(1) command. The
6130 available command-line options are:
6132 [role="growable",cols="asciidoc,asciidoc"]
6133 .Command-line options for the man:lttng-enable-channel(1) command.
6135 |Option |Description
6141 <<channel-overwrite-mode-vs-discard-mode,event loss mode>> instead of
6142 the default _discard_ mode.
6144 |`--buffers-pid` (user space tracing domain only)
6147 Use the per-process <<channel-buffering-schemes,buffering scheme>>
6148 instead of the default per-user buffering scheme.
6150 |+--subbuf-size=__SIZE__+
6153 Allocate sub-buffers of +__SIZE__+ bytes (power of two), for each CPU,
6154 either for each Unix user (default), or for each instrumented process.
6156 See <<channel-subbuf-size-vs-subbuf-count,Sub-buffer count and size>>.
6158 |+--num-subbuf=__COUNT__+
6161 Allocate +__COUNT__+ sub-buffers (power of two), for each CPU, either
6162 for each Unix user (default), or for each instrumented process.
6164 See <<channel-subbuf-size-vs-subbuf-count,Sub-buffer count and size>>.
6166 |+--tracefile-size=__SIZE__+
6169 Set the maximum size of each trace file that this channel writes within
6170 a stream to +__SIZE__+ bytes instead of no maximum.
6172 See <<tracefile-rotation,Trace file count and size>>.
6174 |+--tracefile-count=__COUNT__+
6177 Limit the number of trace files that this channel creates to
6178 +__COUNT__+ channels instead of no limit.
6180 See <<tracefile-rotation,Trace file count and size>>.
6182 |+--switch-timer=__PERIODUS__+
6185 Set the <<channel-switch-timer,switch timer period>>
6186 to +__PERIODUS__+{nbsp}µs.
6188 |+--read-timer=__PERIODUS__+
6191 Set the <<channel-read-timer,read timer period>>
6192 to +__PERIODUS__+{nbsp}µs.
6194 |[[opt-blocking-timeout]]+--blocking-timeout=__TIMEOUTUS__+
6197 Set the timeout of user space applications which load LTTng-UST
6198 in blocking mode to +__TIMEOUTUS__+:
6201 Never block (non-blocking mode).
6204 Block forever until space is available in a sub-buffer to record
6207 __n__, a positive value::
6208 Wait for at most __n__ µs when trying to write into a sub-buffer.
6210 Note that, for this option to have any effect on an instrumented
6211 user space application, you need to run the application with a set
6212 env:LTTNG_UST_ALLOW_BLOCKING environment variable.
6214 |+--output=__TYPE__+ (Linux kernel tracing domain only)
6217 Set the channel's output type to +__TYPE__+, either `mmap` or `splice`.
6221 You can only create a channel in the Linux kernel and user space
6222 <<domain,tracing domains>>: other tracing domains have their own channel
6223 created on the fly when <<enabling-disabling-events,creating event
6228 Because of a current LTTng limitation, you must create all channels
6229 _before_ you <<basic-tracing-session-control,start tracing>> in a given
6230 tracing session, that is, before the first time you run
6233 Since LTTng automatically creates a default channel when you use the
6234 man:lttng-enable-event(1) command with a specific tracing domain, you
6235 cannot, for example, create a Linux kernel event rule, start tracing,
6236 and then create a user space event rule, because no user space channel
6237 exists yet and it's too late to create one.
6239 For this reason, make sure to configure your channels properly
6240 before starting the tracers for the first time!
6243 The following examples show how you can combine the previous
6244 command-line options to create simple to more complex channels.
6246 .Create a Linux kernel channel with default attributes.
6250 $ lttng enable-channel --kernel my-channel
6254 .Create a user space channel with 4 sub-buffers or 1{nbsp}MiB each, per CPU, per instrumented process.
6258 $ lttng enable-channel --userspace --num-subbuf=4 --subbuf-size=1M \
6259 --buffers-pid my-channel
6263 .[[blocking-timeout-example]]Create a default user space channel with an infinite blocking timeout.
6265 <<creating-destroying-tracing-sessions,Create a tracing-session>>,
6266 create the channel, <<enabling-disabling-events,create an event rule>>,
6267 and <<basic-tracing-session-control,start tracing>>:
6272 $ lttng enable-channel --userspace --blocking-timeout=inf blocking-channel
6273 $ lttng enable-event --userspace --channel=blocking-channel --all
6277 Run an application instrumented with LTTng-UST and allow it to block:
6281 $ LTTNG_UST_ALLOW_BLOCKING=1 my-app
6285 .Create a Linux kernel channel which rotates 8 trace files of 4{nbsp}MiB each for each stream
6289 $ lttng enable-channel --kernel --tracefile-count=8 \
6290 --tracefile-size=4194304 my-channel
6294 .Create a user space channel in overwrite (or _flight recorder_) mode.
6298 $ lttng enable-channel --userspace --overwrite my-channel
6302 You can <<enabling-disabling-events,create>> the same event rule in
6303 two different channels:
6307 $ lttng enable-event --userspace --channel=my-channel app:tp
6308 $ lttng enable-event --userspace --channel=other-channel app:tp
6311 If both channels are enabled, when a tracepoint named `app:tp` is
6312 reached, LTTng records two events, one for each channel.
6316 === Disable a channel
6318 To disable a specific channel that you <<enabling-disabling-channels,created>>
6319 previously, use the man:lttng-disable-channel(1) command.
6321 .Disable a specific Linux kernel channel.
6325 $ lttng disable-channel --kernel my-channel
6329 The state of a channel precedes the individual states of event rules
6330 attached to it: event rules which belong to a disabled channel, even if
6331 they are enabled, are also considered disabled.
6335 === Add context fields to a channel
6337 Event record fields in trace files provide important information about
6338 events that occured previously, but sometimes some external context may
6339 help you solve a problem faster. Examples of context fields are:
6341 * The **process ID**, **thread ID**, **process name**, and
6342 **process priority** of the thread in which the event occurs.
6343 * The **hostname** of the system on which the event occurs.
6344 * The current values of many possible **performance counters** using
6346 ** CPU cycles, stalled cycles, idle cycles, and the other cycle types.
6348 ** Branch instructions, misses, and loads.
6350 * Any context defined at the application level (supported for the
6351 JUL and log4j <<domain,tracing domains>>).
6353 To get the full list of available context fields, see
6354 `lttng add-context --list`. Some context fields are reserved for a
6355 specific <<domain,tracing domain>> (Linux kernel or user space).
6357 You add context fields to <<channel,channels>>. All the events
6358 that a channel with added context fields records contain those fields.
6360 To add context fields to one or all the channels of a given tracing
6363 * Use the man:lttng-add-context(1) command.
6365 .Add context fields to all the channels of the current tracing session.
6367 The following command line adds the virtual process identifier and
6368 the per-thread CPU cycles count fields to all the user space channels
6369 of the current tracing session.
6373 $ lttng add-context --userspace --type=vpid --type=perf:thread:cpu-cycles
6377 .Add performance counter context fields by raw ID
6379 See man:lttng-add-context(1) for the exact format of the context field
6380 type, which is partly compatible with the format used in
6385 $ lttng add-context --userspace --type=perf:thread:raw:r0110:test
6386 $ lttng add-context --kernel --type=perf:cpu:raw:r0013c:x86unhalted
6390 .Add a context field to a specific channel.
6392 The following command line adds the thread identifier context field
6393 to the Linux kernel channel named `my-channel` in the current
6398 $ lttng add-context --kernel --channel=my-channel --type=tid
6402 .Add an application-specific context field to a specific channel.
6404 The following command line adds the `cur_msg_id` context field of the
6405 `retriever` context retriever for all the instrumented
6406 <<java-application,Java applications>> recording <<event,event records>>
6407 in the channel named `my-channel`:
6411 $ lttng add-context --kernel --channel=my-channel \
6412 --type='$app:retriever:cur_msg_id'
6415 IMPORTANT: Make sure to always quote the `$` character when you
6416 use man:lttng-add-context(1) from a shell.
6419 NOTE: You cannot remove context fields from a channel once you add it.
6424 === Track process IDs
6426 It's often useful to allow only specific process IDs (PIDs) to emit
6427 events. For example, you may wish to record all the system calls made by
6428 a given process (Ă la http://linux.die.net/man/1/strace[strace]).
6430 The man:lttng-track(1) and man:lttng-untrack(1) commands serve this
6431 purpose. Both commands operate on a whitelist of process IDs. You _add_
6432 entries to this whitelist with the man:lttng-track(1) command and remove
6433 entries with the man:lttng-untrack(1) command. Any process which has one
6434 of the PIDs in the whitelist is allowed to emit LTTng events which pass
6435 an enabled <<event,event rule>>.
6437 NOTE: The PID tracker tracks the _numeric process IDs_. Should a
6438 process with a given tracked ID exit and another process be given this
6439 ID, then the latter would also be allowed to emit events.
6441 .Track and untrack process IDs.
6443 For the sake of the following example, assume the target system has 16
6447 <<creating-destroying-tracing-sessions,create a tracing session>>,
6448 the whitelist contains all the possible PIDs:
6451 .All PIDs are tracked.
6452 image::track-all.png[]
6454 When the whitelist is full and you use the man:lttng-track(1) command to
6455 specify some PIDs to track, LTTng first clears the whitelist, then it
6456 tracks the specific PIDs. After:
6460 $ lttng track --pid=3,4,7,10,13
6466 .PIDs 3, 4, 7, 10, and 13 are tracked.
6467 image::track-3-4-7-10-13.png[]
6469 You can add more PIDs to the whitelist afterwards:
6473 $ lttng track --pid=1,15,16
6479 .PIDs 1, 15, and 16 are added to the whitelist.
6480 image::track-1-3-4-7-10-13-15-16.png[]
6482 The man:lttng-untrack(1) command removes entries from the PID tracker's
6483 whitelist. Given the previous example, the following command:
6487 $ lttng untrack --pid=3,7,10,13
6490 leads to this whitelist:
6493 .PIDs 3, 7, 10, and 13 are removed from the whitelist.
6494 image::track-1-4-15-16.png[]
6496 LTTng can track all possible PIDs again using the
6497 opt:lttng-track(1):--all option:
6501 $ lttng track --pid --all
6504 The result is, again:
6507 .All PIDs are tracked.
6508 image::track-all.png[]
6511 .Track only specific PIDs
6513 A very typical use case with PID tracking is to start with an empty
6514 whitelist, then <<basic-tracing-session-control,start the tracers>>, and
6515 then add PIDs manually while tracers are active. You can accomplish this
6516 by using the opt:lttng-untrack(1):--all option of the
6517 man:lttng-untrack(1) command to clear the whitelist after you
6518 <<creating-destroying-tracing-sessions,create a tracing session>>:
6522 $ lttng untrack --pid --all
6528 .No PIDs are tracked.
6529 image::untrack-all.png[]
6531 If you trace with this whitelist configuration, the tracer records no
6532 events for this <<domain,tracing domain>> because no processes are
6533 tracked. You can use the man:lttng-track(1) command as usual to track
6534 specific PIDs, for example:
6538 $ lttng track --pid=6,11
6544 .PIDs 6 and 11 are tracked.
6545 image::track-6-11.png[]
6550 [[saving-loading-tracing-session]]
6551 === Save and load tracing session configurations
6553 Configuring a <<tracing-session,tracing session>> can be long. Some of
6554 the tasks involved are:
6556 * <<enabling-disabling-channels,Create channels>> with
6557 specific attributes.
6558 * <<adding-context,Add context fields>> to specific channels.
6559 * <<enabling-disabling-events,Create event rules>> with specific log
6560 level and filter conditions.
6562 If you use LTTng to solve real world problems, chances are you have to
6563 record events using the same tracing session setup over and over,
6564 modifying a few variables each time in your instrumented program
6565 or environment. To avoid constant tracing session reconfiguration,
6566 the man:lttng(1) command-line tool can save and load tracing session
6567 configurations to/from XML files.
6569 To save a given tracing session configuration:
6571 * Use the man:lttng-save(1) command:
6576 $ lttng save my-session
6580 Replace `my-session` with the name of the tracing session to save.
6582 LTTng saves tracing session configurations to
6583 dir:{$LTTNG_HOME/.lttng/sessions} by default. Note that the
6584 env:LTTNG_HOME environment variable defaults to `$HOME` if not set. Use
6585 the opt:lttng-save(1):--output-path option to change this destination
6588 LTTng saves all configuration parameters, for example:
6590 * The tracing session name.
6591 * The trace data output path.
6592 * The channels with their state and all their attributes.
6593 * The context fields you added to channels.
6594 * The event rules with their state, log level and filter conditions.
6596 To load a tracing session:
6598 * Use the man:lttng-load(1) command:
6603 $ lttng load my-session
6607 Replace `my-session` with the name of the tracing session to load.
6609 When LTTng loads a configuration, it restores your saved tracing session
6610 as if you just configured it manually.
6612 See man:lttng(1) for the complete list of command-line options. You
6613 can also save and load all many sessions at a time, and decide in which
6614 directory to output the XML files.
6617 [[sending-trace-data-over-the-network]]
6618 === Send trace data over the network
6620 LTTng can send the recorded trace data to a remote system over the
6621 network instead of writing it to the local file system.
6623 To send the trace data over the network:
6625 . On the _remote_ system (which can also be the target system),
6626 start an LTTng <<lttng-relayd,relay daemon>> (man:lttng-relayd(8)):
6635 . On the _target_ system, create a tracing session configured to
6636 send trace data over the network:
6641 $ lttng create my-session --set-url=net://remote-system
6645 Replace `remote-system` by the host name or IP address of the
6646 remote system. See man:lttng-create(1) for the exact URL format.
6648 . On the target system, use the man:lttng(1) command-line tool as usual.
6649 When tracing is active, the target's consumer daemon sends sub-buffers
6650 to the relay daemon running on the remote system instead of flushing
6651 them to the local file system. The relay daemon writes the received
6652 packets to the local file system.
6654 The relay daemon writes trace files to
6655 +$LTTNG_HOME/lttng-traces/__hostname__/__session__+ by default, where
6656 +__hostname__+ is the host name of the target system and +__session__+
6657 is the tracing session name. Note that the env:LTTNG_HOME environment
6658 variable defaults to `$HOME` if not set. Use the
6659 opt:lttng-relayd(8):--output option of man:lttng-relayd(8) to write
6660 trace files to another base directory.
6665 === View events as LTTng emits them (noch:{LTTng} live)
6667 LTTng live is a network protocol implemented by the <<lttng-relayd,relay
6668 daemon>> (man:lttng-relayd(8)) to allow compatible trace viewers to
6669 display events as LTTng emits them on the target system while tracing is
6672 The relay daemon creates a _tee_: it forwards the trace data to both
6673 the local file system and to connected live viewers:
6676 .The relay daemon creates a _tee_, forwarding the trace data to both trace files and a connected live viewer.
6681 . On the _target system_, create a <<tracing-session,tracing session>>
6687 $ lttng create my-session --live
6691 This spawns a local relay daemon.
6693 . Start the live viewer and configure it to connect to the relay
6694 daemon. For example, with http://diamon.org/babeltrace[Babeltrace]:
6699 $ babeltrace --input-format=lttng-live \
6700 net://localhost/host/hostname/my-session
6707 * `hostname` with the host name of the target system.
6708 * `my-session` with the name of the tracing session to view.
6711 . Configure the tracing session as usual with the man:lttng(1)
6712 command-line tool, and <<basic-tracing-session-control,start tracing>>.
6714 You can list the available live tracing sessions with Babeltrace:
6718 $ babeltrace --input-format=lttng-live net://localhost
6721 You can start the relay daemon on another system. In this case, you need
6722 to specify the relay daemon's URL when you create the tracing session
6723 with the opt:lttng-create(1):--set-url option. You also need to replace
6724 `localhost` in the procedure above with the host name of the system on
6725 which the relay daemon is running.
6727 See man:lttng-create(1) and man:lttng-relayd(8) for the complete list of
6728 command-line options.
6732 [[taking-a-snapshot]]
6733 === Take a snapshot of the current sub-buffers of a tracing session
6735 The normal behavior of LTTng is to append full sub-buffers to growing
6736 trace data files. This is ideal to keep a full history of the events
6737 that occurred on the target system, but it can
6738 represent too much data in some situations. For example, you may wish
6739 to trace your application continuously until some critical situation
6740 happens, in which case you only need the latest few recorded
6741 events to perform the desired analysis, not multi-gigabyte trace files.
6743 With the man:lttng-snapshot(1) command, you can take a snapshot of the
6744 current sub-buffers of a given <<tracing-session,tracing session>>.
6745 LTTng can write the snapshot to the local file system or send it over
6750 . Create a tracing session in _snapshot mode_:
6755 $ lttng create my-session --snapshot
6759 The <<channel-overwrite-mode-vs-discard-mode,event loss mode>> of
6760 <<channel,channels>> created in this mode is automatically set to
6761 _overwrite_ (flight recorder mode).
6763 . Configure the tracing session as usual with the man:lttng(1)
6764 command-line tool, and <<basic-tracing-session-control,start tracing>>.
6766 . **Optional**: When you need to take a snapshot,
6767 <<basic-tracing-session-control,stop tracing>>.
6769 You can take a snapshot when the tracers are active, but if you stop
6770 them first, you are sure that the data in the sub-buffers does not
6771 change before you actually take the snapshot.
6778 $ lttng snapshot record --name=my-first-snapshot
6782 LTTng writes the current sub-buffers of all the current tracing
6783 session's channels to trace files on the local file system. Those trace
6784 files have `my-first-snapshot` in their name.
6786 There is no difference between the format of a normal trace file and the
6787 format of a snapshot: viewers of LTTng traces also support LTTng
6790 By default, LTTng writes snapshot files to the path shown by
6791 `lttng snapshot list-output`. You can change this path or decide to send
6792 snapshots over the network using either:
6794 . An output path or URL that you specify when you create the
6796 . An snapshot output path or URL that you add using
6797 `lttng snapshot add-output`
6798 . An output path or URL that you provide directly to the
6799 `lttng snapshot record` command.
6801 Method 3 overrides method 2, which overrides method 1. When you
6802 specify a URL, a relay daemon must listen on a remote system (see
6803 <<sending-trace-data-over-the-network,Send trace data over the network>>).
6808 === Use the machine interface
6810 With any command of the man:lttng(1) command-line tool, you can set the
6811 opt:lttng(1):--mi option to `xml` (before the command name) to get an
6812 XML machine interface output, for example:
6816 $ lttng --mi=xml enable-event --kernel --syscall open
6819 A schema definition (XSD) is
6820 https://github.com/lttng/lttng-tools/blob/stable-2.10/src/common/mi-lttng-3.0.xsd[available]
6821 to ease the integration with external tools as much as possible.
6825 [[metadata-regenerate]]
6826 === Regenerate the metadata of an LTTng trace
6828 An LTTng trace, which is a http://diamon.org/ctf[CTF] trace, has both
6829 data stream files and a metadata file. This metadata file contains,
6830 amongst other things, information about the offset of the clock sources
6831 used to timestamp <<event,event records>> when tracing.
6833 If, once a <<tracing-session,tracing session>> is
6834 <<basic-tracing-session-control,started>>, a major
6835 https://en.wikipedia.org/wiki/Network_Time_Protocol[NTP] correction
6836 happens, the trace's clock offset also needs to be updated. You
6837 can use the `metadata` item of the man:lttng-regenerate(1) command
6840 The main use case of this command is to allow a system to boot with
6841 an incorrect wall time and trace it with LTTng before its wall time
6842 is corrected. Once the system is known to be in a state where its
6843 wall time is correct, it can run `lttng regenerate metadata`.
6845 To regenerate the metadata of an LTTng trace:
6847 * Use the `metadata` item of the man:lttng-regenerate(1) command:
6852 $ lttng regenerate metadata
6858 `lttng regenerate metadata` has the following limitations:
6860 * Tracing session <<creating-destroying-tracing-sessions,created>>
6862 * User space <<channel,channels>>, if any, are using
6863 <<channel-buffering-schemes,per-user buffering>>.
6868 [[regenerate-statedump]]
6869 === Regenerate the state dump of a tracing session
6871 The LTTng kernel and user space tracers generate state dump
6872 <<event,event records>> when the application starts or when you
6873 <<basic-tracing-session-control,start a tracing session>>. An analysis
6874 can use the state dump event records to set an initial state before it
6875 builds the rest of the state from the following event records.
6876 http://tracecompass.org/[Trace Compass] is a notable example of an
6877 application which uses the state dump of an LTTng trace.
6879 When you <<taking-a-snapshot,take a snapshot>>, it's possible that the
6880 state dump event records are not included in the snapshot because they
6881 were recorded to a sub-buffer that has been consumed or overwritten
6884 You can use the `lttng regenerate statedump` command to emit the state
6885 dump event records again.
6887 To regenerate the state dump of the current tracing session, provided
6888 create it in snapshot mode, before you take a snapshot:
6890 . Use the `statedump` item of the man:lttng-regenerate(1) command:
6895 $ lttng regenerate statedump
6899 . <<basic-tracing-session-control,Stop the tracing session>>:
6908 . <<taking-a-snapshot,Take a snapshot>>:
6913 $ lttng snapshot record --name=my-snapshot
6917 Depending on the event throughput, you should run steps 1 and 2
6918 as closely as possible.
6920 NOTE: To record the state dump events, you need to
6921 <<enabling-disabling-events,create event rules>> which enable them.
6922 LTTng-UST state dump tracepoints start with `lttng_ust_statedump:`.
6923 LTTng-modules state dump tracepoints start with `lttng_statedump_`.
6927 [[persistent-memory-file-systems]]
6928 === Record trace data on persistent memory file systems
6930 https://en.wikipedia.org/wiki/Non-volatile_random-access_memory[Non-volatile random-access memory]
6931 (NVRAM) is random-access memory that retains its information when power
6932 is turned off (non-volatile). Systems with such memory can store data
6933 structures in RAM and retrieve them after a reboot, without flushing
6934 to typical _storage_.
6936 Linux supports NVRAM file systems thanks to either
6937 http://pramfs.sourceforge.net/[PRAMFS] or
6938 https://www.kernel.org/doc/Documentation/filesystems/dax.txt[DAX]{nbsp}+{nbsp}http://lkml.iu.edu/hypermail/linux/kernel/1504.1/03463.html[pmem]
6939 (requires Linux 4.1+).
6941 This section does not describe how to operate such file systems;
6942 we assume that you have a working persistent memory file system.
6944 When you create a <<tracing-session,tracing session>>, you can specify
6945 the path of the shared memory holding the sub-buffers. If you specify a
6946 location on an NVRAM file system, then you can retrieve the latest
6947 recorded trace data when the system reboots after a crash.
6949 To record trace data on a persistent memory file system and retrieve the
6950 trace data after a system crash:
6952 . Create a tracing session with a sub-buffer shared memory path located
6953 on an NVRAM file system:
6958 $ lttng create my-session --shm-path=/path/to/shm
6962 . Configure the tracing session as usual with the man:lttng(1)
6963 command-line tool, and <<basic-tracing-session-control,start tracing>>.
6965 . After a system crash, use the man:lttng-crash(1) command-line tool to
6966 view the trace data recorded on the NVRAM file system:
6971 $ lttng-crash /path/to/shm
6975 The binary layout of the ring buffer files is not exactly the same as
6976 the trace files layout. This is why you need to use man:lttng-crash(1)
6977 instead of your preferred trace viewer directly.
6979 To convert the ring buffer files to LTTng trace files:
6981 * Use the opt:lttng-crash(1):--extract option of man:lttng-crash(1):
6986 $ lttng-crash --extract=/path/to/trace /path/to/shm
6992 [[notif-trigger-api]]
6993 === Get notified when a channel's buffer usage is too high or too low
6995 With LTTng's $$C/C++$$ notification and trigger API, your user
6996 application can get notified when the buffer usage of one or more
6997 <<channel,channels>> becomes too low or too high. You can use this API
6998 and enable or disable <<event,event rules>> during tracing to avoid
6999 <<channel-overwrite-mode-vs-discard-mode,discarded event records>>.
7001 .Have a user application get notified when an LTTng channel's buffer usage is too high.
7003 In this example, we create and build an application which gets notified
7004 when the buffer usage of a specific LTTng channel is higher than
7005 75{nbsp}%. We only print that it is the case in the example, but we
7006 could as well use the API of <<liblttng-ctl-lttng,`liblttng-ctl`>> to
7007 disable event rules when this happens.
7009 . Create the application's C source file:
7017 #include <lttng/domain.h>
7018 #include <lttng/action/action.h>
7019 #include <lttng/action/notify.h>
7020 #include <lttng/condition/condition.h>
7021 #include <lttng/condition/buffer-usage.h>
7022 #include <lttng/condition/evaluation.h>
7023 #include <lttng/notification/channel.h>
7024 #include <lttng/notification/notification.h>
7025 #include <lttng/trigger/trigger.h>
7026 #include <lttng/endpoint.h>
7028 int main(int argc, char *argv[])
7030 int exit_status = 0;
7031 struct lttng_notification_channel *notification_channel;
7032 struct lttng_condition *condition;
7033 struct lttng_action *action;
7034 struct lttng_trigger *trigger;
7035 const char *tracing_session_name;
7036 const char *channel_name;
7039 tracing_session_name = argv[1];
7040 channel_name = argv[2];
7043 * Create a notification channel. A notification channel
7044 * connects the user application to the LTTng session daemon.
7045 * This notification channel can be used to listen to various
7046 * types of notifications.
7048 notification_channel = lttng_notification_channel_create(
7049 lttng_session_daemon_notification_endpoint);
7052 * Create a "high buffer usage" condition. In this case, the
7053 * condition is reached when the buffer usage is greater than or
7054 * equal to 75 %. We create the condition for a specific tracing
7055 * session name, channel name, and for the user space tracing
7058 * The "low buffer usage" condition type also exists.
7060 condition = lttng_condition_buffer_usage_high_create();
7061 lttng_condition_buffer_usage_set_threshold_ratio(condition, .75);
7062 lttng_condition_buffer_usage_set_session_name(
7063 condition, tracing_session_name);
7064 lttng_condition_buffer_usage_set_channel_name(condition,
7066 lttng_condition_buffer_usage_set_domain_type(condition,
7070 * Create an action (get a notification) to take when the
7071 * condition created above is reached.
7073 action = lttng_action_notify_create();
7076 * Create a trigger. A trigger associates a condition to an
7077 * action: the action is executed when the condition is reached.
7079 trigger = lttng_trigger_create(condition, action);
7081 /* Register the trigger to LTTng. */
7082 lttng_register_trigger(trigger);
7085 * Now that we have registered a trigger, a notification will be
7086 * emitted everytime its condition is met. To receive this
7087 * notification, we must subscribe to notifications that match
7088 * the same condition.
7090 lttng_notification_channel_subscribe(notification_channel,
7094 * Notification loop. You can put this in a dedicated thread to
7095 * avoid blocking the main thread.
7098 struct lttng_notification *notification;
7099 enum lttng_notification_channel_status status;
7100 const struct lttng_evaluation *notification_evaluation;
7101 const struct lttng_condition *notification_condition;
7102 double buffer_usage;
7104 /* Receive the next notification. */
7105 status = lttng_notification_channel_get_next_notification(
7106 notification_channel, ¬ification);
7109 case LTTNG_NOTIFICATION_CHANNEL_STATUS_OK:
7111 case LTTNG_NOTIFICATION_CHANNEL_STATUS_NOTIFICATIONS_DROPPED:
7113 * The session daemon can drop notifications if
7114 * a monitoring application is not consuming the
7115 * notifications fast enough.
7118 case LTTNG_NOTIFICATION_CHANNEL_STATUS_CLOSED:
7120 * The notification channel has been closed by the
7121 * session daemon. This is typically caused by a session
7122 * daemon shutting down.
7126 /* Unhandled conditions or errors. */
7132 * A notification provides, amongst other things:
7134 * * The condition that caused this notification to be
7136 * * The condition evaluation, which provides more
7137 * specific information on the evaluation of the
7140 * The condition evaluation provides the buffer usage
7141 * value at the moment the condition was reached.
7143 notification_condition = lttng_notification_get_condition(
7145 notification_evaluation = lttng_notification_get_evaluation(
7148 /* We're subscribed to only one condition. */
7149 assert(lttng_condition_get_type(notification_condition) ==
7150 LTTNG_CONDITION_TYPE_BUFFER_USAGE_HIGH);
7153 * Get the exact sampled buffer usage from the
7154 * condition evaluation.
7156 lttng_evaluation_buffer_usage_get_usage_ratio(
7157 notification_evaluation, &buffer_usage);
7160 * At this point, instead of printing a message, we
7161 * could do something to reduce the channel's buffer
7162 * usage, like disable specific events.
7164 printf("Buffer usage is %f %% in tracing session \"%s\", "
7165 "user space channel \"%s\".\n", buffer_usage * 100,
7166 tracing_session_name, channel_name);
7167 lttng_notification_destroy(notification);
7171 lttng_action_destroy(action);
7172 lttng_condition_destroy(condition);
7173 lttng_trigger_destroy(trigger);
7174 lttng_notification_channel_destroy(notification_channel);
7180 . Build the `notif-app` application, linking it to `liblttng-ctl`:
7185 $ gcc -o notif-app notif-app.c -llttng-ctl
7189 . <<creating-destroying-tracing-sessions,Create a tracing session>>,
7190 <<enabling-disabling-events,create an event rule>> matching all the
7191 user space tracepoints, and
7192 <<basic-tracing-session-control,start tracing>>:
7197 $ lttng create my-session
7198 $ lttng enable-event --userspace --all
7203 If you create the channel manually with the man:lttng-enable-channel(1)
7204 command, you can control how frequently are the current values of the
7205 channel's properties sampled to evaluate user conditions with the
7206 opt:lttng-enable-channel(1):--monitor-timer option.
7208 . Run the `notif-app` application. This program accepts the
7209 <<tracing-session,tracing session>> name and the user space channel
7210 name as its two first arguments. The channel which LTTng automatically
7211 creates with the man:lttng-enable-event(1) command above is named
7217 $ ./notif-app my-session channel0
7221 . In another terminal, run an application with a very high event
7222 throughput so that the 75{nbsp}% buffer usage condition is reached.
7224 In the first terminal, the application should print lines like this:
7227 Buffer usage is 81.45197 % in tracing session "my-session", user space
7231 If you don't see anything, try modifying the condition in
7232 path:{notif-app.c} to a lower value (0.1, for example), rebuilding it
7233 (step 2) and running it again (step 4).
7240 [[lttng-modules-ref]]
7241 === noch:{LTTng-modules}
7245 [[lttng-tracepoint-enum]]
7246 ==== `LTTNG_TRACEPOINT_ENUM()` usage
7248 Use the `LTTNG_TRACEPOINT_ENUM()` macro to define an enumeration:
7252 LTTNG_TRACEPOINT_ENUM(name, TP_ENUM_VALUES(entries))
7257 * `name` with the name of the enumeration (C identifier, unique
7258 amongst all the defined enumerations).
7259 * `entries` with a list of enumeration entries.
7261 The available enumeration entry macros are:
7263 +ctf_enum_value(__name__, __value__)+::
7264 Entry named +__name__+ mapped to the integral value +__value__+.
7266 +ctf_enum_range(__name__, __begin__, __end__)+::
7267 Entry named +__name__+ mapped to the range of integral values between
7268 +__begin__+ (included) and +__end__+ (included).
7270 +ctf_enum_auto(__name__)+::
7271 Entry named +__name__+ mapped to the integral value following the
7272 last mapping's value.
7274 The last value of a `ctf_enum_value()` entry is its +__value__+
7277 The last value of a `ctf_enum_range()` entry is its +__end__+ parameter.
7279 If `ctf_enum_auto()` is the first entry in the list, its integral
7282 Use the `ctf_enum()` <<lttng-modules-tp-fields,field definition macro>>
7283 to use a defined enumeration as a tracepoint field.
7285 .Define an enumeration with `LTTNG_TRACEPOINT_ENUM()`.
7289 LTTNG_TRACEPOINT_ENUM(
7292 ctf_enum_auto("AUTO: EXPECT 0")
7293 ctf_enum_value("VALUE: 23", 23)
7294 ctf_enum_value("VALUE: 27", 27)
7295 ctf_enum_auto("AUTO: EXPECT 28")
7296 ctf_enum_range("RANGE: 101 TO 303", 101, 303)
7297 ctf_enum_auto("AUTO: EXPECT 304")
7305 [[lttng-modules-tp-fields]]
7306 ==== Tracepoint fields macros (for `TP_FIELDS()`)
7308 [[tp-fast-assign]][[tp-struct-entry]]The available macros to define
7309 tracepoint fields, which must be listed within `TP_FIELDS()` in
7310 `LTTNG_TRACEPOINT_EVENT()`, are:
7312 [role="func-desc growable",cols="asciidoc,asciidoc"]
7313 .Available macros to define LTTng-modules tracepoint fields
7315 |Macro |Description and parameters
7318 +ctf_integer(__t__, __n__, __e__)+
7320 +ctf_integer_nowrite(__t__, __n__, __e__)+
7322 +ctf_user_integer(__t__, __n__, __e__)+
7324 +ctf_user_integer_nowrite(__t__, __n__, __e__)+
7326 Standard integer, displayed in base 10.
7329 Integer C type (`int`, `long`, `size_t`, ...).
7335 Argument expression.
7338 +ctf_integer_hex(__t__, __n__, __e__)+
7340 +ctf_user_integer_hex(__t__, __n__, __e__)+
7342 Standard integer, displayed in base 16.
7351 Argument expression.
7353 |+ctf_integer_oct(__t__, __n__, __e__)+
7355 Standard integer, displayed in base 8.
7364 Argument expression.
7367 +ctf_integer_network(__t__, __n__, __e__)+
7369 +ctf_user_integer_network(__t__, __n__, __e__)+
7371 Integer in network byte order (big-endian), displayed in base 10.
7380 Argument expression.
7383 +ctf_integer_network_hex(__t__, __n__, __e__)+
7385 +ctf_user_integer_network_hex(__t__, __n__, __e__)+
7387 Integer in network byte order, displayed in base 16.
7396 Argument expression.
7399 +ctf_enum(__N__, __t__, __n__, __e__)+
7401 +ctf_enum_nowrite(__N__, __t__, __n__, __e__)+
7403 +ctf_user_enum(__N__, __t__, __n__, __e__)+
7405 +ctf_user_enum_nowrite(__N__, __t__, __n__, __e__)+
7410 Name of a <<lttng-tracepoint-enum,previously defined enumeration>>.
7413 Integer C type (`int`, `long`, `size_t`, ...).
7419 Argument expression.
7422 +ctf_string(__n__, __e__)+
7424 +ctf_string_nowrite(__n__, __e__)+
7426 +ctf_user_string(__n__, __e__)+
7428 +ctf_user_string_nowrite(__n__, __e__)+
7430 Null-terminated string; undefined behavior if +__e__+ is `NULL`.
7436 Argument expression.
7439 +ctf_array(__t__, __n__, __e__, __s__)+
7441 +ctf_array_nowrite(__t__, __n__, __e__, __s__)+
7443 +ctf_user_array(__t__, __n__, __e__, __s__)+
7445 +ctf_user_array_nowrite(__t__, __n__, __e__, __s__)+
7447 Statically-sized array of integers.
7450 Array element C type.
7456 Argument expression.
7462 +ctf_array_bitfield(__t__, __n__, __e__, __s__)+
7464 +ctf_array_bitfield_nowrite(__t__, __n__, __e__, __s__)+
7466 +ctf_user_array_bitfield(__t__, __n__, __e__, __s__)+
7468 +ctf_user_array_bitfield_nowrite(__t__, __n__, __e__, __s__)+
7470 Statically-sized array of bits.
7472 The type of +__e__+ must be an integer type. +__s__+ is the number
7473 of elements of such type in +__e__+, not the number of bits.
7476 Array element C type.
7482 Argument expression.
7488 +ctf_array_text(__t__, __n__, __e__, __s__)+
7490 +ctf_array_text_nowrite(__t__, __n__, __e__, __s__)+
7492 +ctf_user_array_text(__t__, __n__, __e__, __s__)+
7494 +ctf_user_array_text_nowrite(__t__, __n__, __e__, __s__)+
7496 Statically-sized array, printed as text.
7498 The string does not need to be null-terminated.
7501 Array element C type (always `char`).
7507 Argument expression.
7513 +ctf_sequence(__t__, __n__, __e__, __T__, __E__)+
7515 +ctf_sequence_nowrite(__t__, __n__, __e__, __T__, __E__)+
7517 +ctf_user_sequence(__t__, __n__, __e__, __T__, __E__)+
7519 +ctf_user_sequence_nowrite(__t__, __n__, __e__, __T__, __E__)+
7521 Dynamically-sized array of integers.
7523 The type of +__E__+ must be unsigned.
7526 Array element C type.
7532 Argument expression.
7535 Length expression C type.
7541 +ctf_sequence_hex(__t__, __n__, __e__, __T__, __E__)+
7543 +ctf_user_sequence_hex(__t__, __n__, __e__, __T__, __E__)+
7545 Dynamically-sized array of integers, displayed in base 16.
7547 The type of +__E__+ must be unsigned.
7550 Array element C type.
7556 Argument expression.
7559 Length expression C type.
7564 |+ctf_sequence_network(__t__, __n__, __e__, __T__, __E__)+
7566 Dynamically-sized array of integers in network byte order (big-endian),
7567 displayed in base 10.
7569 The type of +__E__+ must be unsigned.
7572 Array element C type.
7578 Argument expression.
7581 Length expression C type.
7587 +ctf_sequence_bitfield(__t__, __n__, __e__, __T__, __E__)+
7589 +ctf_sequence_bitfield_nowrite(__t__, __n__, __e__, __T__, __E__)+
7591 +ctf_user_sequence_bitfield(__t__, __n__, __e__, __T__, __E__)+
7593 +ctf_user_sequence_bitfield_nowrite(__t__, __n__, __e__, __T__, __E__)+
7595 Dynamically-sized array of bits.
7597 The type of +__e__+ must be an integer type. +__s__+ is the number
7598 of elements of such type in +__e__+, not the number of bits.
7600 The type of +__E__+ must be unsigned.
7603 Array element C type.
7609 Argument expression.
7612 Length expression C type.
7618 +ctf_sequence_text(__t__, __n__, __e__, __T__, __E__)+
7620 +ctf_sequence_text_nowrite(__t__, __n__, __e__, __T__, __E__)+
7622 +ctf_user_sequence_text(__t__, __n__, __e__, __T__, __E__)+
7624 +ctf_user_sequence_text_nowrite(__t__, __n__, __e__, __T__, __E__)+
7626 Dynamically-sized array, displayed as text.
7628 The string does not need to be null-terminated.
7630 The type of +__E__+ must be unsigned.
7632 The behaviour is undefined if +__e__+ is `NULL`.
7635 Sequence element C type (always `char`).
7641 Argument expression.
7644 Length expression C type.
7650 Use the `_user` versions when the argument expression, `e`, is
7651 a user space address. In the cases of `ctf_user_integer*()` and
7652 `ctf_user_float*()`, `&e` must be a user space address, thus `e` must
7655 The `_nowrite` versions omit themselves from the session trace, but are
7656 otherwise identical. This means the `_nowrite` fields won't be written
7657 in the recorded trace. Their primary purpose is to make some
7658 of the event context available to the
7659 <<enabling-disabling-events,event filters>> without having to
7660 commit the data to sub-buffers.
7666 Terms related to LTTng and to tracing in general:
7669 The http://diamon.org/babeltrace[Babeltrace] project, which includes
7670 the cmd:babeltrace command, some libraries, and Python bindings.
7672 <<channel-buffering-schemes,buffering scheme>>::
7673 A layout of sub-buffers applied to a given channel.
7675 <<channel,channel>>::
7676 An entity which is responsible for a set of ring buffers.
7678 <<event,Event rules>> are always attached to a specific channel.
7681 A reference of time for a tracer.
7683 <<lttng-consumerd,consumer daemon>>::
7684 A process which is responsible for consuming the full sub-buffers
7685 and write them to a file system or send them over the network.
7687 <<channel-overwrite-mode-vs-discard-mode,discard mode>>:: The event loss
7688 mode in which the tracer _discards_ new event records when there's no
7689 sub-buffer space left to store them.
7692 The consequence of the execution of an instrumentation
7693 point, like a tracepoint that you manually place in some source code,
7694 or a Linux kernel KProbe.
7696 An event is said to _occur_ at a specific time. Different actions can
7697 be taken upon the occurrence of an event, like record the event's payload
7700 <<channel-overwrite-mode-vs-discard-mode,event loss mode>>::
7701 The mechanism by which event records of a given channel are lost
7702 (not recorded) when there is no sub-buffer space left to store them.
7704 [[def-event-name]]event name::
7705 The name of an event, which is also the name of the event record.
7706 This is also called the _instrumentation point name_.
7709 A record, in a trace, of the payload of an event which occured.
7711 <<event,event rule>>::
7712 Set of conditions which must be satisfied for one or more occuring
7713 events to be recorded.
7715 `java.util.logging`::
7717 https://docs.oracle.com/javase/7/docs/api/java/util/logging/package-summary.html[core logging facilities].
7719 <<instrumenting,instrumentation>>::
7720 The use of LTTng probes to make a piece of software traceable.
7722 instrumentation point::
7723 A point in the execution path of a piece of software that, when
7724 reached by this execution, can emit an event.
7726 instrumentation point name::
7727 See _<<def-event-name,event name>>_.
7730 A http://logging.apache.org/log4j/1.2/[logging library] for Java
7731 developed by the Apache Software Foundation.
7734 Level of severity of a log statement or user space
7735 instrumentation point.
7738 The _Linux Trace Toolkit: next generation_ project.
7740 <<lttng-cli,cmd:lttng>>::
7741 A command-line tool provided by the LTTng-tools project which you
7742 can use to send and receive control messages to and from a
7746 The https://github.com/lttng/lttng-analyses[LTTng analyses] project,
7747 which is a set of analyzing programs that are used to obtain a
7748 higher level view of an LTTng trace.
7750 cmd:lttng-consumerd::
7751 The name of the consumer daemon program.
7754 A utility provided by the LTTng-tools project which can convert
7755 ring buffer files (usually
7756 <<persistent-memory-file-systems,saved on a persistent memory file system>>)
7759 LTTng Documentation::
7762 <<lttng-live,LTTng live>>::
7763 A communication protocol between the relay daemon and live viewers
7764 which makes it possible to see events "live", as they are received by
7767 <<lttng-modules,LTTng-modules>>::
7768 The https://github.com/lttng/lttng-modules[LTTng-modules] project,
7769 which contains the Linux kernel modules to make the Linux kernel
7770 instrumentation points available for LTTng tracing.
7773 The name of the relay daemon program.
7775 cmd:lttng-sessiond::
7776 The name of the session daemon program.
7779 The https://github.com/lttng/lttng-tools[LTTng-tools] project, which
7780 contains the various programs and libraries used to
7781 <<controlling-tracing,control tracing>>.
7783 <<lttng-ust,LTTng-UST>>::
7784 The https://github.com/lttng/lttng-ust[LTTng-UST] project, which
7785 contains libraries to instrument user applications.
7787 <<lttng-ust-agents,LTTng-UST Java agent>>::
7788 A Java package provided by the LTTng-UST project to allow the
7789 LTTng instrumentation of `java.util.logging` and Apache log4j 1.2
7792 <<lttng-ust-agents,LTTng-UST Python agent>>::
7793 A Python package provided by the LTTng-UST project to allow the
7794 LTTng instrumentation of Python logging statements.
7796 <<channel-overwrite-mode-vs-discard-mode,overwrite mode>>::
7797 The event loss mode in which new event records overwrite older
7798 event records when there's no sub-buffer space left to store them.
7800 <<channel-buffering-schemes,per-process buffering>>::
7801 A buffering scheme in which each instrumented process has its own
7802 sub-buffers for a given user space channel.
7804 <<channel-buffering-schemes,per-user buffering>>::
7805 A buffering scheme in which all the processes of a Unix user share the
7806 same sub-buffer for a given user space channel.
7808 <<lttng-relayd,relay daemon>>::
7809 A process which is responsible for receiving the trace data sent by
7810 a distant consumer daemon.
7813 A set of sub-buffers.
7815 <<lttng-sessiond,session daemon>>::
7816 A process which receives control commands from you and orchestrates
7817 the tracers and various LTTng daemons.
7819 <<taking-a-snapshot,snapshot>>::
7820 A copy of the current data of all the sub-buffers of a given tracing
7821 session, saved as trace files.
7824 One part of an LTTng ring buffer which contains event records.
7827 The time information attached to an event when it is emitted.
7830 A set of files which are the concatenations of one or more
7831 flushed sub-buffers.
7834 The action of recording the events emitted by an application
7835 or by a system, or to initiate such recording by controlling
7839 The http://tracecompass.org[Trace Compass] project and application.
7842 An instrumentation point using the tracepoint mechanism of the Linux
7843 kernel or of LTTng-UST.
7845 tracepoint definition::
7846 The definition of a single tracepoint.
7849 The name of a tracepoint.
7851 tracepoint provider::
7852 A set of functions providing tracepoints to an instrumented user
7855 Not to be confused with a _tracepoint provider package_: many tracepoint
7856 providers can exist within a tracepoint provider package.
7858 tracepoint provider package::
7859 One or more tracepoint providers compiled as an object file or as
7863 A software which records emitted events.
7865 <<domain,tracing domain>>::
7866 A namespace for event sources.
7868 <<tracing-group,tracing group>>::
7869 The Unix group in which a Unix user can be to be allowed to trace the
7872 <<tracing-session,tracing session>>::
7873 A stateful dialogue between you and a <<lttng-sessiond,session
7877 An application running in user space, as opposed to a Linux kernel
7878 module, for example.