doc/man/lttng.1

   1 .TH "LTTNG" "1" "February 9, 2012" "" ""
   2
   3 .SH "NAME"
   4 lttng \(em LTTng 2.0 tracer control command line tool
   5
   6 .SH "SYNOPSIS"
   7
   8 .PP
   9 .nf
  10 lttng [OPTIONS] <COMMAND>
  11 .fi
  12 .SH "DESCRIPTION"
  13
  14 .PP
  15 The LTTng project aims at providing highly efficient tracing tools for Linux.
  16 It's tracers help tracking down performance issues and debugging problems
  17 involving multiple concurrent processes and threads. Tracing across multiple
  18 systems is also possible.
  19
  20 The \fBlttng\fP command line tool from the lttng-tools package is used to control
  21 both kernel and user-space tracing. Every interactions with the tracer should
  22 be done by this tool or by the liblttng-ctl provided with the lttng-tools
  23 package.
  24
  25 LTTng uses a session daemon (lttng-sessiond(8)), acting as a tracing registry,
  26 which permits you to interact with multiple tracers (kernel and user-space)
  27 inside the same container, a tracing session. Traces can be gathered from the
  28 kernel and/or instrumented applications (lttng-ust(3)). Aggregating and reading
  29 those traces is done using the babeltrace(1) text viewer.
  30
  31 In order to trace the kernel, the session daemon needs to be running as root.
  32 LTTng provides the use of a \fBtracing group\fP (default: tracing). Whomever is
  33 in that group can interact with the root session daemon and thus trace the
  34 kernel. Session daemons can co-exist meaning that you can have a session daemon
  35 running as Alice that can be used to trace her applications along side with a
  36 root daemon or even a Bob daemon. We highly recommend to start the session
  37 daemon at boot time for stable and long term tracing.
  38
  39 Every user-space applications instrumented with lttng-ust(3), will
  40 automatically register to the session daemon. This feature gives you the
  41 ability to list available traceable applications and tracepoints on a per user
  42 basis. (See \fBlist\fP command).
  43 .SH "OPTIONS"
  44
  45 .PP
  46 This program follow the usual GNU command line syntax with long options starting with
  47 two dashes. Below is a summary of the available options.
  48 .PP
  49
  50 .TP
  51 .BR "\-h, \-\-help"
  52 Show summary of possible options and commands.
  53 .TP
  54 .BR "\-v, \-\-verbose"
  55 Increase verbosity.
  56 Three levels of verbosity are available which are triggered by putting additionnal v to
  57 the option (\-vv or \-vvv)
  58 .TP
  59 .BR "\-q, \-\-quiet"
  60 Suppress all messages (even errors).
  61 .TP
  62 .BR "\-g, \-\-group NAME"
  63 Set unix tracing group name. (default: tracing)
  64 .TP
  65 .BR "\-n, \-\-no-sessiond"
  66 Don't automatically spawn a session daemon.
  67 .TP
  68 .BR "\-\-sessiond\-path"
  69 Set session daemon full binary path.
  70 .TP
  71 .BR "\-\-list\-options"
  72 Simple listing of lttng options.
  73 .TP
  74 .BR "\-\-list\-commands"
  75 Simple listing of lttng commands.
  76 .SH "COMMANDS"
  77
  78 .TP
  79 \fBadd-context\fP
  80 .nf
  81 Add context to event(s) and/or channel(s).
  82
  83 A context is basically extra information appended to a channel or event. For
  84 instance, you could ask the tracer to add the PID information within the
  85 "sched_switch" kernel event. You can also add performance monitoring unit
  86 counters (perf PMU) using the perf kernel API).
  87
  88 For example, this command will add the context information 'prio' and two perf
  89 counters (hardware branch misses and cache misses), to all events in the trace
  90 data output:
  91
  92 # lttng add-context \-k \-t prio \-t perf:branch-misses \-t perf:cache-misses
  93
  94 Please take a look at the help (\-h/\-\-help) for a detailed list of available
  95 contexts.
  96
  97 If no channel and no event is given (\-c/\-e), the context is added to all
  98 channels (which applies automatically to all events in that channel). Otherwise
  99 the context will be added only to the channel (\-c) and/or event (\-e) indicated.
 100
 101 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
 102 file.
 103 .fi
 104
 105 .B OPTIONS:
 106
 107 .nf
 108 \-h, \-\-help
 109         Show summary of possible options and commands.
 110 \-s, \-\-session NAME
 111         Apply on session name.
 112 \-c, \-\-channel NAME
 113         Apply on channel name.
 114 \-e, \-\-event NAME
 115         Apply on event name.
 116 \-k, \-\-kernel
 117         Apply for the kernel tracer
 118 \-u, \-\-userspace
 119         Apply for the user-space tracer
 120 \-t, \-\-type TYPE
 121         Context type. You can repeat this option on the command line. Please
 122         use "lttng add-context \-h" to list all available types.
 123 .fi
 124
 125 .IP
 126
 127 .IP "\fBcalibrate\fP"
 128 .nf
 129 Quantify LTTng overhead
 130
 131 The LTTng calibrate command can be used to find out the combined average
 132 overhead of the LTTng tracer and the instrumentation mechanisms used. This
 133 overhead can be calibrated in terms of time or using any of the PMU performance
 134 counter available on the system.
 135
 136 For now, the only calibration implemented is that of the kernel function
 137 instrumentation (kretprobes).
 138
 139 * Calibrate kernel function instrumentation
 140
 141 Let's use an example to show this calibration. We use an i7 processor with 4
 142 general-purpose PMU registers. This information is available by issuing dmesg,
 143 looking for "generic registers".
 144
 145 This sequence of commands will gather a trace executing a kretprobe hooked on
 146 an empty function, gathering PMU counters LLC (Last Level Cache) misses
 147 information (see lttng add-context \-\-help to see the list of available PMU
 148 counters).
 149
 150 # lttng create calibrate-function
 151 # lttng enable-event calibrate \-\-kernel \-\-function lttng_calibrate_kretprobe
 152 # lttng add-context \-\-kernel \-t perf:LLC-load-misses \-t perf:LLC-store-misses \\
 153                   \-t perf:LLC-prefetch-misses
 154 # lttng start
 155 # for a in $(seq 1 10); do \\
 156         lttng calibrate \-\-kernel \-\-function;
 157   done
 158 # lttng destroy
 159 # babeltrace $(ls \-1drt ~/lttng-traces/calibrate-function-* | tail \-n 1)
 160
 161 The output from babeltrace can be saved to a text file and opened in a
 162 spreadsheet (e.g. oocalc) to focus on the per-PMU counter delta between
 163 consecutive "calibrate_entry" and "calibrate_return" events. Note that these
 164 counters are per-CPU, so scheduling events would need to be present to account
 165 for migration between CPU. Therefore, for calibration purposes, only events
 166 staying on the same CPU must be considered.
 167
 168 The average result, for the i7, on 10 samples:
 169
 170                           Average     Std.Dev.
 171 perf_LLC_load_misses:       5.0       0.577
 172 perf_LLC_store_misses:      1.6       0.516
 173 perf_LLC_prefetch_misses:   9.0      14.742
 174
 175 As we can notice, the load and store misses are relatively stable across runs
 176 (their standard deviation is relatively low) compared to the prefetch misses.
 177 We can conclude from this information that LLC load and store misses can be
 178 accounted for quite precisely, but prefetches within a function seems to behave
 179 too erratically (not much causality link between the code executed and the CPU
 180 prefetch activity) to be accounted for.
 181 .fi
 182
 183 .B OPTIONS:
 184
 185 .nf
 186 \-h, \-\-help
 187         Show summary of possible options and commands.
 188 \-k, \-\-kernel
 189         Apply for the kernel tracer
 190 \-u, \-\-userspace
 191         Apply for the user-space tracer
 192 \-\-function
 193         Dynamic function entry/return probe (default)
 194 .fi
 195
 196 .IP
 197
 198 .IP "\fBcreate\fP [OPTIONS] [NAME]
 199 .nf
 200 Create tracing session.
 201
 202 A tracing session contains channel(s) which contains event(s). It is domain
 203 agnostic meaning that you can enable channels and events for either the
 204 user-space tracer and/or the kernel tracer. It acts like a container
 205 aggregating multiple tracing sources.
 206
 207 On creation, a \fB.lttngrc\fP file is created in your $HOME directory
 208 containing the current session name. If NAME is omitted, a session name is
 209 automatically created having this form: 'auto-yyyymmdd-hhmmss'.
 210
 211 If no \fB\-o, \-\-output\fP is specified, the traces will be written in
 212 $HOME/lttng-traces.
 213 .fi
 214
 215 .B OPTIONS:
 216
 217 .nf
 218 \-h, \-\-help
 219         Show summary of possible options and commands.
 220 \-\-list-options
 221         Simple listing of options
 222 \-o, \-\-output PATH
 223         Specify output path for traces
 224 .fi
 225
 226 .IP
 227
 228 .IP "\fBdestroy\fP [OPTIONS] [NAME]"
 229 .nf
 230 Teardown tracing session
 231
 232 Free memory on the session daemon and tracer side. It's gone!
 233
 234 If NAME is omitted, the session name is taken from the .lttngrc file.
 235 .fi
 236
 237 .B OPTIONS:
 238
 239 .nf
 240 \-h, \-\-help
 241         Show summary of possible options and commands.
 242 \-\-list-options
 243         Simple listing of options
 244 .fi
 245
 246 .IP
 247
 248 .IP "\fBenable-channel\fP NAME[,NAME2,...] [-k|-u] [OPTIONS]"
 249 .nf
 250 Enable tracing channel
 251
 252 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
 253 file.
 254 .fi
 255
 256 .B OPTIONS:
 257
 258 .nf
 259 \-h, \-\-help
 260         Show this help
 261 \-\-list-options
 262         Simple listing of options
 263 \-s, \-\-session
 264         Apply on session name
 265 \-k, \-\-kernel
 266         Apply to the kernel tracer
 267 \-u, \-\-userspace
 268         Apply to the user-space tracer
 269
 270 \-\-discard
 271         Discard event when subbuffers are full (default)
 272 \-\-overwrite
 273         Flight recorder mode : overwrites events when subbuffers are full
 274 \-\-subbuf-size
 275         Subbuffer size in bytes (default: 4096, kernel default: 262144)
 276 \-\-num-subbuf
 277         Number of subbufers (default: 4)
 278         Needs to be a power of 2 for kernel and ust tracers
 279 \-\-switch-timer
 280         Switch subbuffer timer interval in usec (default: 0)
 281         Needs to be a power of 2 for kernel and ust tracers
 282 \-\-read-timer
 283         Read timer interval in usec (default: 200)
 284 .fi
 285
 286 .IP
 287
 288 .IP "\fBenable-event\fP NAME[,NAME2,...] [-k|-u] [OPTIONS]"
 289 .nf
 290 Enable tracing event
 291
 292 A tracing event is always assigned to a channel. If \fB\-c, \-\-channel\fP is
 293 omitted, a default channel named '\fBchannel0\fP' is created and the event is
 294 added to it. For the user-space tracer, using \fB\-a, \-\-all\fP is the same as
 295 using the wildcard "*".
 296
 297 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
 298 file.
 299 .fi
 300
 301 .B OPTIONS:
 302
 303 .nf
 304 \-h, \-\-help
 305         Show summary of possible options and commands.
 306 \-\-list-options
 307         Simple listing of options
 308 \-s, \-\-session
 309         Apply on session name
 310 \-c, \-\-channel
 311         Apply on channel name
 312 \-a, \-\-all
 313         Enable all tracepoints and syscalls
 314 \-k, \-\-kernel
 315         Apply for the kernel tracer
 316 \-u, \-\-userspace
 317         Apply for the user-space tracer
 318
 319 \-\-tracepoint
 320         Tracepoint event (default)
 321         - userspace tracer supports wildcards at end of string. Don't forget to
 322         quote to deal with bash expansion.
 323         e.g.:
 324         "*"
 325         "app_component:na*"
 326 \-\-loglevel
 327         Tracepoint loglevel
 328 \-\-probe [addr | symbol | symbol+offset]
 329         Dynamic probe. Addr and offset can be octal (0NNN...), decimal (NNN...)
 330         or hexadecimal (0xNNN...)
 331 \-\-function [addr | symbol | symbol+offset]
 332         Dynamic function entry/return probe. Addr and offset can be octal
 333         (0NNN...), decimal (NNN...) or hexadecimal (0xNNN...)
 334 \-\-syscall
 335         System call event
 336         Enabling syscalls tracing (kernel tracer), you will not be able to disable them
 337         with disable-event. This is a known limitation. You can disable the entire
 338         channel to do the trick.
 339 .fi
 340
 341 .IP "\fBdisable-channel\fP NAME[,NAME2,...] [\-k|\-u] [OPTIONS]"
 342 .nf
 343 Disable tracing channel
 344
 345 Disabling a channel makes all event(s) in that channel to stop tracing. You can
 346 enable it back by calling \fBlttng enable-channel NAME\fP again.
 347
 348 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
 349 file.
 350 .fi
 351
 352 .B OPTIONS:
 353
 354 .nf
 355 \-h, \-\-help
 356         Show summary of possible options and commands.
 357 \-\-list-options
 358         Simple listing of options
 359 \-s, \-\-session
 360         Apply on session name
 361 \-k, \-\-kernel
 362         Apply for the kernel tracer
 363 \-u, \-\-userspace
 364         Apply for the user-space tracer
 365 .fi
 366
 367 .IP "\fBdisable-event\fP NAME[,NAME2,...] [\-k|\-u] [OPTIONS]"
 368 .nf
 369 Disable tracing event
 370
 371 The event, once disabled, can be re-enabled by calling \fBlttng enable-event
 372 NAME\fP again.
 373
 374 If \fB\-s, \-\-session\fP is omitted, the session name is taken from the .lttngrc
 375 file.
 376 .fi
 377
 378 .B OPTIONS:
 379
 380 .nf
 381 \-h, \-\-help
 382         Show summary of possible options and commands.
 383 \-\-list-options
 384         Simple listing of options
 385 \-s, \-\-session
 386         Apply on session name
 387 \-k, \-\-kernel
 388         Apply for the kernel tracer
 389 \-u, \-\-userspace
 390         Apply for the user-space tracer
 391 .fi
 392
 393 .IP "\fBlist\fP [\-k|\-u] [SESSION [SESSION_OPTIONS]]"
 394 .nf
 395 List tracing session information.
 396
 397 With no arguments, it will list available tracing session(s).
 398
 399 With the session name, it will display the details of the session including
 400 the trace file path, the associated channels and their state (activated
 401 and deactivated), the activated events and more.
 402
 403 With \-k alone, it will list all available kernel events (except the system
 404 calls events).
 405 With \-u alone, it will list all available user-space events from registered
 406 applications. Here is an example of 'lttng list \-u':
 407
 408 PID: 7448 - Name: /tmp/lttng-ust/tests/hello/.libs/lt-hello
 409       ust_tests_hello:tptest_sighandler (type: tracepoint)
 410       ust_tests_hello:tptest (type: tracepoint)
 411
 412 You can now enable any event listed by using the name :
 413 \fBust_tests_hello:tptest\fP.
 414 .fi
 415
 416 .B OPTIONS:
 417
 418 .nf
 419 \-h, \-\-help
 420         Show summary of possible options and commands.
 421 \-\-list-options
 422         Simple listing of options
 423 \-k, \-\-kernel
 424         Select kernel domain (FIXME : apparition de la notion de "domain" ici)
 425 \-u, \-\-userspace
 426         Select user-space domain.
 427
 428 Session options:
 429 \-c, \-\-channel NAME
 430         List details of a channel
 431 \-d, \-\-domain
 432         List available domain(s)
 433 .fi
 434
 435 .IP "\fBset-session\fP NAME"
 436 .nf
 437 Set current session name
 438
 439 Will change the session name in the .lttngrc file.
 440 .fi
 441
 442 .B OPTIONS:
 443
 444 .nf
 445 \-h, \-\-help
 446         Show summary of possible options and commands.
 447 \-\-list-options
 448         Simple listing of options
 449 .fi
 450
 451 .IP
 452
 453 .IP "\fBstart\fP [OPTIONS] [NAME]"
 454 .nf
 455 Start tracing
 456
 457 It will start tracing for all tracers for a specific tracing session.
 458
 459 If NAME is omitted, the session name is taken from the .lttngrc file.
 460 .fi
 461
 462 .B OPTIONS:
 463
 464 .nf
 465 \-h, \-\-help
 466         Show summary of possible options and commands.
 467 \-\-list-options
 468         Simple listing of options
 469 .fi
 470
 471 .IP
 472
 473 .IP "\fBstop\fP [OPTIONS] [NAME]"
 474 .nf
 475 Stop tracing
 476
 477 It will stop tracing for all tracers for a specific tracing session.
 478
 479 If NAME is omitted, the session name is taken from the .lttngrc file.
 480 .fi
 481
 482 .B OPTIONS:
 483
 484 .nf
 485 \-h, \-\-help
 486         Show summary of possible options and commands.
 487 \-\-list-options
 488         Simple listing of options
 489 .fi
 490
 491 .IP
 492
 493 .IP "\fBversion\fP"
 494 .nf
 495 Show version information
 496 .fi
 497
 498 .B OPTIONS:
 499
 500 .nf
 501 \-h, \-\-help
 502         Show summary of possible options and commands.
 503 \-\-list-options
 504         Simple listing of options
 505 .fi
 506
 507 .IP
 508
 509 .IP "\fBview\fP [SESSION_NAME] [OPTIONS]"
 510 .nf
 511 View traces of a tracing session
 512
 513 By default, the babeltrace viewer will be used for text viewing.
 514
 515 If SESSION_NAME is omitted, the session name is taken from the .lttngrc file.
 516
 517 .fi
 518
 519 .B OPTIONS:
 520
 521 .nf
 522 \-h, \-\-help
 523         Show this help
 524 \-\-list-options
 525         Simple listing of options
 526 \-t, \-\-trace-path PATH
 527         Trace directory path for the viewer
 528 \-e, \-\-viewer CMD
 529         Specify viewer and/or options to use
 530         This will completely override the default viewers so
 531         please make sure to specify the full command. The trace
 532         directory path of the session will be appended at the end
 533         to the arguments
 534 .fi
 535
 536 .SH "EXIT VALUES"
 537
 538 .IP "0"
 539 Success
 540
 541 .IP "1"
 542 Command error
 543
 544 .IP "2"
 545 Undefined command
 546
 547 .IP "3"
 548 Fatal error
 549
 550 .IP "4"
 551 Command warning
 552
 553 .IP "16"
 554 No session found by the name given
 555
 556 .IP "18"
 557 Error in session creation
 558
 559 .IP "21"
 560 Error in application(s) listing
 561
 562 .IP "28"
 563 Session name already exists
 564
 565 .IP "33"
 566 Kernel tracer unavailable
 567
 568 .IP "35"
 569 Kernel event exists
 570
 571 .IP "37"
 572 Kernel channel exists
 573
 574 .IP "38"
 575 Kernel channel creation failed
 576
 577 .IP "39"
 578 Kernel channel not found
 579
 580 .IP "40"
 581 Kernel channel disable failed
 582
 583 .IP "41"
 584 Kernel channel enable failed
 585
 586 .IP "42"
 587 Kernel context failed
 588
 589 .IP "43"
 590 Kernel enable event failed
 591
 592 .IP "44"
 593 Kernel disable event failed
 594
 595 .IP "53"
 596 Kernel listing events failed
 597
 598 .IP "60"
 599 UST channel disable failed
 600
 601 .IP "61"
 602 UST channel enable failed
 603
 604 .IP "62"
 605 UST adding context failed
 606
 607 .IP "63"
 608 UST event enable failed
 609
 610 .IP "64"
 611 UST event disable failed
 612
 613 .IP "66"
 614 UST start failed
 615
 616 .IP "67"
 617 UST stop failed
 618
 619 .IP "75"
 620 UST event exists
 621
 622 .IP "76"
 623 UST event not found
 624
 625 .IP "77"
 626 UST context exists
 627
 628 .IP "78"
 629 UST invalid context
 630
 631 .IP "79"
 632 Tracing the kernel requires a root lttng-sessiond daemon and "tracing" group
 633 user membership.
 634
 635 .IP "80"
 636 Tracing already started
 637
 638 .IP "81"
 639 Tracing already stopped
 640 .PP
 641 .SH "ENVIRONMENT VARIABLES"
 642
 643 .PP
 644 Note that all command line options override environment variables.
 645 .PP
 646
 647 .PP
 648 .IP "LTTNG_SESSIOND_PATH_ENV"
 649 Allows one to specify the full session daemon binary path to lttng command line
 650 tool. You can also use \-\-sessiond-path option having the same effect.
 651 .SH "SEE ALSO"
 652
 653 .PP
 654 babeltrace(1), lttng-ust(3), lttng-sessiond(8)
 655 .PP
 656 .SH "BUGS"
 657
 658 .PP
 659 No show stopper bugs are known yet in this version.
 660
 661 If you encounter any issues or usability problem, please report it on our
 662 mailing list <lttng-dev@lists.lttng.org> to help improve this project.
 663 .SH "CREDITS"
 664
 665 .PP
 666 lttng is distributed under the GNU General Public License version 2. See the file
 667 COPYING for details.
 668 .PP
 669 A Web site is available at http://lttng.org for more information on the LTTng
 670 project.
 671 .PP
 672 You can also find our git tree at http://git.lttng.org.
 673 .PP
 674 Mailing lists for support and development: <lttng-dev@lists.lttng.org>.
 675 .PP
 676 You can find us on IRC server irc.oftc.net (OFTC) in #lttng.
 677 .PP
 678 .SH "THANKS"
 679
 680 .PP
 681 Thanks to Yannick Brosseau without whom this project would never have been so
 682 lean and mean! Also thanks to the Ericsson teams working on tracing which
 683 helped us greatly with detailed bug reports and unusual test cases.
 684
 685 Thanks to our beloved packager Alexandre Montplaisir-Goncalves (Ubuntu and PPA
 686 maintainer) and Jon Bernard for our Debian packages.
 687
 688 Special thanks to Michel Dagenais and the DORSAL laboratory at Polytechnique de
 689 Montreal for the LTTng journey.
 690 .PP
 691 .SH "AUTHORS"
 692
 693 .PP
 694 lttng-tools was originally written by Mathieu Desnoyers, Julien Desfossez and
 695 David Goulet. More people have since contributed to it. It is currently
 696 maintained by David Goulet <dgoulet@efficios.com>.
 697 .PP