LTTng bugs repository: Issueshttps://bugs.lttng.org/https://bugs.lttng.org/themes/lttng/favicon/a.ico?14249722912024-02-21T17:04:43ZLTTng bugs repository
Redmine Babeltrace - Feature #1410 (New): Add pretty-print sink options to override how to format floatin...https://bugs.lttng.org/issues/14102024-02-21T17:04:43ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>Babeltrace 2, just like its predecessor babeltrace 1.5, uses "%g" to print floating point numbers.</p>
<p>It would be useful to let users optionally override the formatting for floating point numbers, perhaps with a pretty print sink option. They could then choose if exponent notation should be used or not, choose the precision, etc.</p> LTTng-tools - Bug #1383 (New): The "cpu_id" context (available for filters) is not discoverable b...https://bugs.lttng.org/issues/13832023-07-23T16:22:58ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>I recently tried remembering how to filter by cpu_id when enabling an event, and found that there was not clear way to see that $ctx.cpu_id actually exists from the lttng man pages (other than a random example in the lttng-enable-event man page). AFAIU the man pages rely on lttng add-context --list to let the user discover the available contexts. This is likely because the cpu_id context is not available for the "add-context" command because it would be redundant with the implicit context already sampled with the buffers.</p>
<p>We may want to have some way to let users discovers those contexts which are filter-specific.</p>
<p>My use-case is to try to grab only traces for a few logical cpus (0-3) from my 384 logical cpus machine.</p> LTTng-tools - Bug #1379 (New): Document behavior of live mode with per-pid UST buffershttps://bugs.lttng.org/issues/13792023-06-01T18:48:58ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>There is a design limitation of per-pid UST buffers when used in live mode which should be documented.</p>
<p>Basically, because the per-pid buffers are reclaimed soon after the application exits, it may cause the events traced by a short-lived application to never appear in the output of a live trace when per-pid buffers are used.</p>
<p>This is caused by the fact that the live client periodically checks for new buffers, but there is no inherent reference kept on the streams before they disappear.</p>
<p>This means that the information will be available in the disk output of the trace, but it will be missing from the live mode output.</p>
<p>This could eventually be mitigated by keeping references to the streams received by the relay daemon which are of interest to a live client until they are referenced by the live client.</p> LTTng-tools - Bug #1378 (New): Document behavior of snapshot mode with per-pid UST buffershttps://bugs.lttng.org/issues/13782023-06-01T18:01:12ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>We should document a design limitation of the per-pid UST buffers which can lead to unexpected results when used with the "snapshot" feature.</p>
<p>Basically, because the per-pid buffers lifetime is bound by the application using them (and the consumer daemon extracting data from them), they are freed almost immediately after the traced application exits. However, in a scenario where we are interested in capturing a snapshot of the content of flight recorder buffers soon after application exits (e.g. caused by a crash), those buffers are not available anymore a few milliseconds after the application exits.</p>
<p>We should document this design limitation, and eventually think about improvements (feature request ?) to allow a configurable delay after application exit during which the consumer daemon could keep references on the buffers so they are available for snapshot.</p> Userspace RCU - Feature #1368 (New): Integrate RCU implementation from libside into liburcuhttps://bugs.lttng.org/issues/13682023-02-14T14:26:10ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>libside features a multi-domain RCU implementation semantically similar to the Linux kernel's SRCU's implementation. It tracks grace periods with per-cpu counters rather than per-thread state.</p>
<p><a class="external" href="https://github.com/efficios/libside/blob/master/src/rcu.h">https://github.com/efficios/libside/blob/master/src/rcu.h</a><br /><a class="external" href="https://github.com/efficios/libside/blob/master/src/rcu.c">https://github.com/efficios/libside/blob/master/src/rcu.c</a></p>
<p>It is somewhat different from other urcu flavors because it supports multiple domains, whereas current liburcu flavors only have a single domain per process.</p>
<p>We would have to think thoroughly about the impacts of supporting multiple domains on other parts of liburcu such as the call_rcu worker thread. The worker thread would either have to be able to synchronize with multiple RCU domains, or we would have to require each worker thread to be associated with a single RCU domain.</p>
<p>Similar concerns arise with respect to data structures such as rculfhash which take a urcu flavor as parameter: with per-domain flavor, those would have to be associated with a specific RCU domain in addition to the flavor.</p> LTTng-tools - Feature #1341 (New): Introduce "--shm-path-ust" to clarify use vs documentationhttps://bugs.lttng.org/issues/13412021-12-07T20:53:52ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>Considering that the "--shm-path" argument is documented as applying to all domains, but that currently the kernel domain does not support this argument, it leads us to a situation where users may budget their NVRAM for the space needed only for UST buffers, and eventually upgrade to a new LTTng version which would implement kernel support for shm-path. They would then face issues given the lack of available space.</p>
<p>I would recommend to add a new "--shm-path-ust" argument as an alias to "--shm-path", but document that it only applies to the UST domain. This way, users wishing to budget their NVRAM space only for UST tracing, but still perform kernel tracing in the same session, will be able to do so.</p> LTTng-tools - Bug #1340 (New): Document behavior of per-uid UST buffers with respect to asynchron...https://bugs.lttng.org/issues/13402021-12-07T20:22:28ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>We should document a known design limitation of the per-uid UST buffers:</p>
<p>When using lttng-ust with per-uid buffers, there is a known design limitation<br />regarding process terminated by non-handled kill signals: if this occurs while<br />the buffer is being written to (between reserve and commit), it will cause the<br />rest of the per-cpu buffer to be unreadable. This applies to all events<br />recorded by all applications of the same user id for that particular CPU.<br />Recovering from this involves destroying and re-creating the tracing session.</p>
<p>If this kind of scenario is likely for you, you might want to consider using per-<br />pid buffers (lttng enable-channel -u --buffers-pid), which do not suffer from<br />this design limitation. The downside of per-pid buffers is that it allocates<br />more memory on your system for buffering, and adds extra overhead when<br />used with processes that have a short life-time.</p> Babeltrace - Feature #1300 (New): babeltrace2 lacks knowledge of bitwise enum produced by lttng-m...https://bugs.lttng.org/issues/13002021-03-23T17:41:47ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>This is a follow up on <a class="external" href="https://review.lttng.org/c/babeltrace/+/3045">https://review.lttng.org/c/babeltrace/+/3045</a> now that end users are starting to contact us wondering why up-to-date lttng and babeltrace2 show incomplete information for bitwise enums.</p>
<p>I think we should implement a better stop-gap solution while awaiting CTF2. I recommend that we specialize the babeltrace2 CTF input plugin to detect traces produced by the lttng-modules kernel tracer, for specific known (hardcoded) events and fields which have a bitwise enum semantic, and use this hardcoded knowledge to print the correct bitwise enum to the end user rather than "<unknown>".</p>
<p>This should ensure the current producer/consumer tools work well together (show complete information about those bitwise enums) without requiring to modify tons of documentation about the specific flags needed for bt2 to handle lttng traces appropriately. And this would not lead to misleading scenarios where we print an unknown value as a bitwise enum by mistake.</p>
<p>Hardcoded producer detection/event/fields is of course a last resort solution awaiting proper self-description in the upcoming CTF2 spec.</p> Babeltrace - Bug #1221 (New): Babeltrace 2 sometimes fail on live trace with lttng-clear featurehttps://bugs.lttng.org/issues/12212020-02-17T20:52:19ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>Babeltrace 2 sometimes fail on live trace with lttng-clear feature</p>
<p>This may happen in a scenario where per-pid buffers are used, an application has just quit, and babeltrace 2 observes a stream in make_viewer_streams, but then fails to open the associated index or data file on the first get_next_index because it has been unlinked by clear.</p>
<p>The babeltrace output is:</p>
<pre>
12-04 14:01:48.455 11884 11884 E PLUGIN/CTF/MSG-ITER create_msg_stream_end@msg-iter.c:2464 [lttng-live] Cannot create stream for stream message: notit-addr=0x55b89f957980
12-04 14:01:48.455 11884 11884 W LIB/MSG-ITER bt_self_component_port_input_message_iterator_next@iterator.c:906 Component input port message iterator's "next" method failed: iter-addr=0x55b89f93f510, iter-upstream-comp-name="lttng-live", iter-upstream-comp-log-level=WARNING, iter-upstream-comp-class-type=SOURCE, iter-upstream-comp-class-name="lttng-live", iter-upstream-comp-class-partial-descr="Connect to an LTTng relay daemon", iter-upstream-port-type=OUTPUT, iter-upstream-port-name="out", status=ERROR
12-04 14:01:48.455 11884 11884 E PLUGIN/FLT.UTILS.MUXER muxer_upstream_msg_iter_next@muxer.c:444 [muxer] Error or unsupported status code: status-code=-1
12-04 14:01:48.455 11884 11884 E PLUGIN/FLT.UTILS.MUXER validate_muxer_upstream_msg_iters@muxer.c:975 [muxer] Cannot validate muxer's upstream message iterator wrapper: muxer-msg-iter-addr=0x55b89f950950, muxer-upstream-msg-iter-wrap-addr=0x55b89f9415d0
12-04 14:01:48.455 11884 11884 E PLUGIN/FLT.UTILS.MUXER muxer_msg_iter_next@muxer.c:1371 [muxer] Cannot get next message: comp-addr=0x55b89f93ebc0, muxer-comp-addr=0x55b89f93ec40, muxer-msg-iter-addr=0x55b89f950950, msg-iter-addr=0x55b89f93f3a0, status=ERROR
12-04 14:01:48.455 11884 11884 W LIB/MSG-ITER bt_self_component_port_input_message_iterator_next@iterator.c:906 Component input port message iterator's "next" method failed: iter-addr=0x55b89f93f3a0, iter-upstream-comp-name="muxer", iter-upstream-comp-log-level=WARNING, iter-upstream-comp-class-type=FILTER, iter-upstream-comp-class-name="muxer", iter-upstream-comp-class-partial-descr="Sort messages from multiple inpu", iter-upstream-port-type=OUTPUT, iter-upstream-port-name="out", status=ERROR
12-04 14:01:48.455 11884 11884 W LIB/GRAPH consume_graph_sink@graph.c:596 Component's "consume" method failed: status=ERROR, comp-addr=0x55b89f93ed30, comp-name="pretty", comp-log-level=WARNING, comp-class-type=SINK, comp-class-name="pretty", comp-class-partial-descr="Pretty-print messages (`text` fo", comp-class-is-frozen=0, comp-class-so-handle-addr=0x55b89f94c740, comp-class-so-handle-path="/usr/local/lib/babeltrace2/plugins/babeltrace-plugin-text.so", comp-input-port-count=1, comp-output-port-count=0
12-04 14:01:48.456 11884 11884 E CLI cmd_run@babeltrace2.c:2574 Graph failed to complete successfully
ERROR: [Babeltrace CLI] (babeltrace2.c:2574)
Graph failed to complete successfully
CAUSED BY [Babeltrace library] (graph.c:596)
Component's "consume" method failed: status=ERROR, comp-addr=0x55b89f93ed30,
comp-name="pretty", comp-log-level=WARNING, comp-class-type=SINK,
comp-class-name="pretty", comp-class-partial-descr="Pretty-print messages
(`text` fo", comp-class-is-frozen=0, comp-class-so-handle-addr=0x55b89f94c740,
comp-class-so-handle-path="/usr/local/lib/babeltrace2/plugins/babeltrace-plugin-text.so",
comp-input-port-count=1, comp-output-port-count=0
CAUSED BY [Babeltrace library] (iterator.c:906)
Component input port message iterator's "next" method failed:
iter-addr=0x55b89f93f3a0, iter-upstream-comp-name="muxer",
iter-upstream-comp-log-level=WARNING, iter-upstream-comp-class-type=FILTER,
iter-upstream-comp-class-name="muxer",
iter-upstream-comp-class-partial-descr="Sort messages from multiple inpu",
iter-upstream-port-type=OUTPUT, iter-upstream-port-name="out", status=ERROR
CAUSED BY [Babeltrace library] (iterator.c:906)
Component input port message iterator's "next" method failed:
iter-addr=0x55b89f93f510, iter-upstream-comp-name="lttng-live",
iter-upstream-comp-log-level=WARNING, iter-upstream-comp-class-type=SOURCE,
iter-upstream-comp-class-name="lttng-live",
iter-upstream-comp-class-partial-descr="Connect to an LTTng relay daemon",
iter-upstream-port-type=OUTPUT, iter-upstream-port-name="out", status=ERROR
</pre>
<p>notit->stream is expected not to be NULL when reaching the state STATE_EMIT_MSG_STREAM_END, but if no packet was received, it is indeed still NULL, because it is the packet header being read that moves to the STATE_CHECK_EMIT_MSG_STREAM_BEGINNING state (after_packet_context_state()).</p> LTTng-UST - Bug #1203 (New): Improve documentation of lttng-ust with daemonshttps://bugs.lttng.org/issues/12032019-10-07T15:51:26ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>The lttng-ust(3) man page should be clarified regarding the scenarios where the liblttng-ust-fork.so wrapper is needed.</p>
<p>It should document that it is needed in scenarios where applications using fork(2), clone(2) (with CLONE_VM flag cleared), BSD's rfork(2), daemon(3), without <strong>immediately</strong> following those by an exec(3) family call. This includes scenarios where close(2) is used to close file descriptors before invoking exec(3), even though close(2) is listed as an async-signal-safe function.</p> LTTng - Bug #1192 (New): Web documentation improvement: describe use of CREATE_TRACE_POINTS for l...https://bugs.lttng.org/issues/11922019-08-05T14:36:42ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>Following discussion with David Goulet who is currently instrumenting the tor project with lttng-ust, it appears that the following use-case should be documented more clearly in the web documentation:</p>
<ul>
<li>Instrumentation of applications with LTTng-UST</li>
</ul>
<p>A tracepoint provider header should be included <em>once</em> per application within a compile unit after a #define CREATE_TRACE_POINTS.</p>
<p>It can be included multiple times throughout other compile units of the applications, but make sure the CREATE_TRACE_POINTS macro is not defined before those include, otherwise LTTng-UST will complain about multiple tracepoint probe definitions.</p> LTTng-tools - Bug #1182 (New): relayd mishandles channels configured with tracefile-count == 0 an...https://bugs.lttng.org/issues/11822019-04-02T17:25:50ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>Relayd mistakenly considers that the criterion tracefile-count == 0 means no tracefile rotation should happen for the stream.</p>
<p>However, it's possible that tracefile-size is set to non-zero, but tracefile-count is left at its default value of 0.</p>
<p>The expected behavior would be to have an ever increasing counter (no limit) of tracefiles each "size" in size at most.</p>
<p>I suspect the relayd receiving side may also be broken, because its rotation handling code also has the same incorrect assumptions.</p>
<p>When connecting a live client to such a session, we get on the client side:<br />[error] Remote side has closed connection</p>
<p>And a PERROR on relayd:<br />PERROR - 13:16:28.097800421 [8061/8068]: opening index in read-only: No such file or directory (in lttng_index_file_open() at index.c:261)</p>
<p>Likely caused by confusion between tracefile rotation suffixes in the filename.</p> LTTng-UST - Feature #965 (New): Implement UST statedumphttps://bugs.lttng.org/issues/9652015-10-22T20:43:37ZMathieu Desnoyersmathieu.desnoyers@efficios.com
<p>Initial implementation: <a class="external" href="https://github.com/compudj/lttng-ust-dev/tree/statedump-notifier">https://github.com/compudj/lttng-ust-dev/tree/statedump-notifier</a></p>
<p>Missing tests in lttng-tools for this feature before we can merge it into lttng-ust.</p> LTTng-modules - Feature #964 (New): Implement support for persistent memory buffershttps://bugs.lttng.org/issues/9642015-10-22T20:41:43ZMathieu Desnoyersmathieu.desnoyers@efficios.comLTTng-modules - Feature #963 (New): Implement user-space stack dump contexthttps://bugs.lttng.org/issues/9632015-10-22T20:41:10ZMathieu Desnoyersmathieu.desnoyers@efficios.com