Matthew Khouzam | 19 Dec 17:23 2014
Picon

Re: [tracecompass-dev] Unreadable traces

Upon further inspection/reflexion, your clock was set at 0, but you are
probably in gmt-5, so you are in 1969 according to the trace, the trace
stores the offset as an unsigned long, so it relies on a rollover. I
would suggest that this is an undocumented feature in lttng. To be more
"correct" it would be nice if the clock offset were a signed long
instead of an unsigned one, this way we can handle the rather common
case of people in the east or west coast (or anywhere west of gmt-0)
working on embedded systems where their clocks are set to 0.

Any thoughts?

Matthew

On 14-12-19 11:11 AM, Matthew Khouzam wrote:
> Hi, this is  a very interesting trace!
>
> The clock offset is 18446744073709551571 ns, or 300 years in the future...
> babeltrace parses the date as follows [1969-12-31 19:13:09.028388290]
> (babeltrace --clock-date)
> tracecompass fails.
>
> How did you take this trace?
> I am kinda more comfortable keeping the trace failing for now, but if
> you have a use case where the clock can be offset by such a large
> number, please share the use case, I have a couple of ideas on how to
> extend it. :)
>
> BR
> Matthew
>
(Continue reading)

Ali Maredia | 18 Dec 19:40 2014

Re: Babeltrace lttng-live API

Thank you so much for the prompt response! I look forward to seeing those features in the next release. If you
have anything up that I can test, I'd love to do so.

Best,

Ali
----- Original Message -----
From: "Jérémie Galarneau" <jeremie.galarneau <at> efficios.com>
To: "Ali Maredia" <ali <at> cohortfs.com>
Cc: lttng-dev <at> lists.lttng.org
Sent: Wednesday, December 17, 2014 6:11:26 PM
Subject: Re: [lttng-dev] Babeltrace lttng-live API

On Wed, Dec 17, 2014 at 5:45 PM, Ali Maredia <ali <at> cohortfs.com> wrote:
> Hello,
>
> I am wondering whether I would be able to implement traces using the lttng-live format using the
Babeltrace API, or whether there is development towards implementing that format in releases soon to come?

Hi Ali,

The plan is to support live trace reading from the public API in the
next Babeltrace release (v1.3) which is scheduled for early 2015.

Regards,
Jérémie

>
> All I see in API.txt (which hasn't changed in a few years) are methods to implement traces that are ctf-format.
>
(Continue reading)

David Engraf | 18 Dec 11:57 2014

[PATCH] Fix compile error on kernel 3.0.101 with CONFIG_PERF_EVENTS


_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
Mathieu Desnoyers | 18 Dec 19:02 2014

[PATCH lttng-tools] Cleanup: relayd: centralize thread stopping function

Rather than relying on having main.c and live.c threads both using the
same notification pipe from different stop_thread implementations,
centralize thread stop in one central function exposed to both main.c
and live.c

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers <at> efficios.com>
---
 src/bin/lttng-relayd/live.c         | 53 +++++++------------------------------
 src/bin/lttng-relayd/lttng-relayd.h |  1 +
 src/bin/lttng-relayd/main.c         | 45 ++++++++++++++++++++-----------
 3 files changed, 39 insertions(+), 60 deletions(-)

diff --git a/src/bin/lttng-relayd/live.c b/src/bin/lttng-relayd/live.c
index beb67b2..fd570e0 100644
--- a/src/bin/lttng-relayd/live.c
+++ b/src/bin/lttng-relayd/live.c
 <at>  <at>  -326,42 +326,12  <at>  <at>  error_unlock:
 	return ret;
 }

-/*
- * Write to writable pipe used to notify a thread.
- */
-static
-int notify_thread_pipe(int wpipe)
-{
-	ssize_t ret;
-
-	ret = lttng_write(wpipe, "!", 1);
-	if (ret < 1) {
(Continue reading)

Mathieu Desnoyers | 18 Dec 02:45 2014

[PATCH lttng-tools 00/13] lttng-tools leak fixes

This patchset fixes various leaks. The main reason for refactoring the
daemon main functions is to make it possible to use valgrind to spot
other leaks without being incumbered by process teardown leaks.

Mathieu Desnoyers (13):
  Fix: agents ht leaks on destroy session
  Fix: leak on UST buffer registry creation error
  Fix: ust app leak on UST buffer creation error
  Missing error handling: ust_app_ht_alloc should return error status
  Missing error handling: consumerd parse_args should return errors to
    caller
  Misleading error handling: utils_create_pid_file() should return 0 on
    success
  Cleanup: lttng-consumerd: remove stale TODO comment
  Cleanup: consumerd: use PERROR rather than perror
  Cleanup: consumerd: lines over 80 columns
  Missing error handling: consumer_signal_init should return its error
    status
  Refactor sessiond main/cleanup/ht-cleanup
  Refactor consumerd main/cleanup
  Refactor relayd main/set_options/cleanup

 src/bin/lttng-consumerd/lttng-consumerd.c | 286 +++++++----
 src/bin/lttng-relayd/live.c               | 127 +++--
 src/bin/lttng-relayd/live.h               |   5 +-
 src/bin/lttng-relayd/main.c               | 207 ++++----
 src/bin/lttng-sessiond/ht-cleanup.c       |  41 +-
 src/bin/lttng-sessiond/lttng-sessiond.h   |   4 +
 src/bin/lttng-sessiond/main.c             | 781 +++++++++++++++++++-----------
 src/bin/lttng-sessiond/trace-ust.c        |   2 +
(Continue reading)

Ali Maredia | 17 Dec 23:45 2014

Babeltrace lttng-live API

Hello,

I am wondering whether I would be able to implement traces using the lttng-live format using the Babeltrace
API, or whether there is development towards implementing that format in releases soon to come? 

All I see in API.txt (which hasn't changed in a few years) are methods to implement traces that are
ctf-format.   

Best,

Ali
David Engraf | 17 Dec 09:01 2014

[PATCH] Update compaction instrumentation for 3.12.x stable kernels


_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
craig harmer | 14 Dec 04:08 2014
Picon

bug #857 -- babeltrace may core dump if index file is too short (with patch)

hi all,

i opened a bug at bugs.lttng.org about a problem that i had where babeltrace would exit, possibly with a core dump, if one the trace's index files was too short.


i submitted a patch for the problem also fixes a few other places that short reads are not handled properly.

there hasn't been any activity on the bug since i submitted the patch about 1 month ago, so i'm wondering if i need to submit the patch to this list as well.

cheers,

craig.
_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
Matthew Khouzam | 13 Dec 01:54 2014
Picon

Bug in LTTng-UST

Hi, it appears as though if a tracepoint is in a so, loading it,
unloading it and reloading it makes it no longer visible.

To confirm this you can try out this repo : 
https://github.com/TheMatthew/yquake2 
(I know, it's not a serious program... but the bug is serious, I think)
You can get the demo files here
http://www.fileplanet.com/6584/0/fileinfo/Quake-2-Demo-v3.14
make the program (make from the git dir)
copy the contents of the baseq2 in the demo into the release/baseq2
directory

start a ust session
run the program.
Ignore the baddr deadlocks :)
The tracepoint is not visible, it is located in the game.so file. 

Chances are though that the problem is on my side, but the evidence
(single registration ) seems to indicate that this is an actual problem.
Jérémie Galarneau | 12 Dec 20:50 2014

[RELEASE] LTTng-Tools 2.6.0-rc4

Hi all,

The lttng-tools project provides a session daemon (lttng-sessiond) that
acts as a tracing registry, the "lttng" command line for tracing
control, an lttng-ctl library for tracing control and an lttng-relay daemon
to provide network streaming and live reading capabilities.

This is version 2.6.0 RC4 which includes 65 new fixes.

Of particular note are fixes pertaining to erroneous RCU usage which could
cause lockups or corruption under heavy loads and a data loss fix which
was discussed on this mailing list [1].

Note that this is expected to be the last release candidate before final.

2014-12-12 lttng-tools 2.6.0-rc4 (International Day Of Heavy Metal)
* Fix: don't destroy the sockets if the snapshot was successful
* Fix: Missing RCU read locks in syscall_list_channel()
* Fix: Missing RCU read lock in kernel_destroy_session()
* Fix: channel names are not validated
* Fix: Apply consumer URI changes to all domains
* Test fix: Don't invoke the system's lttng client
* Tests: Add a test to check disabled events are loaded in the correct state
* Fix: Don't leave events enabled if they were saved in a disabled state
* Fix: "Any" loglevel's value is -1 and not 0
* Test fix: LTTNG_SESSION_CONFIG_XSD_PATH expects an absolute path
* Fix: Memory leak on error in alloc_argv_from_user_opts()
* Fix: Unreachable error logging in set_option()
* Fix: OOM leaks in sessiond modprobe.c
* Fix: test_utils_expand_path.c: out of memory error handling
* Fix: test test_ust_data.c: check OOM
* Fix: test libc-wrapper: check malloc OOM
* Fix: lttng-ctl: use zmalloc(), missing OOM check
* Fix: filter-parser.y: use zmalloc(), missing OOM check
* Fix: utils.c: check str*dup OOM
* Fix: rculfhash: use zmalloc()
* Fix: ini.c: use zmalloc()
* Fix: lttng lttng.c: check strdup OOM
* Fix: lttng conf.c: use zmalloc()
* Fix: lttng view.c missing strdup OOM check
* Fix: use zmalloc() in lttng view.c
* Fix: use zmalloc() in lttng list.c
* Fix: oom error check, realloc oom leak
* Fix: use zmalloc in lttng enable_events.c
* Fix: missing strdup oom check in lttng create.c
* Fix: lttng add_context.c: use zmalloc
* Fix: add missing str*dup oom check in sessiond main.c
* Fix: add missing sessiond kernel-consumer missing str*dup oom checks
* Fix: add missing strdup OOM check
* Fix: Live tracing does not honor live timer after first tracefile
with tracefile rotation
* Fix: Handle empty daemon configuration file lines
* Docs: Add -f/--config option to lttng-sessiond man page
* Fix: relayd: unbalanced RCU read-side lock/unlock
* Fix: relayd: unbalanced RCU read lock/unlock
* Fix: lttng-relayd: unhandled out of memory error
* Fix: relayd viewer stream: unhandled out of memory error
* Fix: Missing rcu_read_lock in cmd_list_events()
* Fix: Missing rcu_read_lock in list_lttng_agent_events()
* Fix: Missing rcu_read_lock in cmd_list_channels()
* Fix: Missing rcu_read_locks in cmd_list_domains()
* Fix: Missing rcu_read_lock in cmd_snapshot_list_outputs()
* Fix: Missing rcu_read_lock in save_domains()
* Fix: Missing rcu_read_lock in ust_app_get_nb_stream()
* Fix: UST subbuffers silently dropped on moderate trace traffic
* Fix: Missing rcu_read_lock in get_session_max_subbuf_size()
* Fix: create/destroy a splice_pipe per stream
* Fix: return EINVAL if agent registration fails
* Note find_ust_app_context must be called with RCU read lock
* Fix: check userspace perf counter name when looking up contexts
* Fix: empty indexes_ht before destroying it.
* Fix: removed useless key from relay_index
* Fix: Ambiguous agent event filter bytecode ownership
* Fix: filter bytecode and string memory leak on error
* Fix: HT must not be destroyed with a rcu_read_lock held
* Fix: agent events HT should be destroyed from the cleanup thread
* Fix: missing rcu_read_lock in trace_ust_destroy_session()
* Fix: missing rcu_read_lock when calling trace_ust_find_agent()
* Fix: missing rcu_read_lock in cmd_start_trace()
* Fix: filter attach vs event enable race
* Fix: Mark libxml2 as a mandatory dependency
* Fix: Warn the user when enabling an event in a new domain
* Fix: Don't report an error when listing a session with no channels
* Fix: Add missing URCU_TLS access to error_log_time
* Update maintainer section of the man pages
* Update maintainer section of README.md

As always, please report any issue you may encounter to bugs.lttng.org
or on this mailing list.

Project website: https://lttng.org

Download links:
https://lttng.org/files/lttng-tools/lttng-tools-2.6.0-rc4.tar.bz2
GPG sig: https://lttng.org/files/lttng-tools/lttng-tools-2.6.0-rc4.tar.bz2.asc

Enjoy!
Jérémie

[1] http://lists.lttng.org/pipermail/lttng-dev/2014-November/023832.html

--

-- 
Jérémie Galarneau
EfficiOS Inc.
http://www.efficios.com

_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
Jérémie Galarneau | 12 Dec 20:49 2014

[RELEASE] LTTng-Tools 2.5.3

Hi all,

The lttng-tools project provides a session daemon (lttng-sessiond) that
acts as a tracing registry, the "lttng" command line for tracing
control, an lttng-ctl library for tracing control and an lttng-relay daemon
to provide network streaming and live reading capabilities.

This is version 2.5.3 which includes 50 new fixes.

A significant portion of these fixes were merged in the last few weeks
as part of the 2.6 release stabilization effort.

Of particular note are fixes pertaining to erroneous RCU usage which could
cause lockups or corruption under heavy loads and a data loss fix which
was discussed on this mailing list [1].

2014-12-12 lttng-tools 2.5.3 (International Day Of Heavy Metal)
* Fix: don't destroy the sockets if the snapshot was successful
* Fix: Missing RCU read lock in kernel_destroy_session()
* Fix: channel names are not validated
* Fix: Apply consumer URI changes to all domains
* Fix: Don't leave events enabled if they were saved in a disabled state
* Test fix: LTTNG_SESSION_CONFIG_XSD_PATH expects an absolute path
* Fix: Memory leak on error in alloc_argv_from_user_opts()
* Fix: Unreachable error logging in set_option()
* Fix: test_utils_expand_path.c: out of memory error handling
* Fix: test test_ust_data.c: check OOM
* Fix: test libc-wrapper: check malloc OOM
* Fix: lttng-ctl: use zmalloc(), missing OOM check
* Fix: filter-parser.y: use zmalloc(), missing OOM check
* Fix: utils.c: check str*dup OOM
* Fix: rculfhash: use zmalloc()
* Fix: ini.c: use zmalloc()
* Fix: lttng lttng.c: check strdup OOM
* Fix: lttng conf.c: use zmalloc()
* Fix: lttng view.c missing strdup OOM check
* Fix: use zmalloc() in lttng view.c
* Fix: use zmalloc() in lttng list.c
* Fix: use zmalloc in lttng enable_events.c
* Fix: Handle empty daemon configuration file lines
* Fix: missing strdup oom check in lttng create.c
* Fix: lttng add_context.c: use zmalloc
* Fix: add missing str*dup oom check in sessiond main.c
* Fix: add missing sessiond kernel-consumer missing str*dup oom checks
* Fix: add missing strdup OOM check
* Fix: Live tracing does not honor live timer after first tracefile
with tracefile rotation
* Docs: Add -f/--config option to lttng-sessiond man page
* Fix: relayd: unbalanced RCU read-side lock/unlock
* Fix: relayd: unbalanced RCU read lock/unlock
* Fix: lttng-relayd: unhandled out of memory error
* Fix: relayd viewer stream: unhandled out of memory error
* Fix: Missing rcu_read_lock in cmd_list_channels()
* Fix: Missing rcu_read_lock in cmd_snapshot_list_outputs()
* Fix: Missing rcu_read_lock in ust_app_get_nb_stream()
* Fix: UST subbuffers silently dropped on moderate trace traffic
* Fix: Missing rcu_read_lock in get_session_max_subbuf_size()
* Fix: create/destroy a splice_pipe per stream
* Fix: return EINVAL if agent registration fails
* Note find_ust_app_context must be called with RCU read lock
* Fix: check userspace perf counter name when looking up contexts
* Fix: empty indexes_ht before destroying it.
* Fix: removed useless key from relay_index
* Fix: filter bytecode and string memory leak on error
* Fix: HT must not be destroyed with a rcu_read_lock held
* Fix: missing rcu_read_lock in cmd_start_trace()
* Fix: filter attach vs event enable race
* Build fix: use ret directly in error handling path
* Fix: Warn the user when enabling an event in a new domain
* Fix: Don't report an error when listing a session with no channels
* Fix: Add missing URCU_TLS access to error_log_time

As always, please report any issue you may encounter to bugs.lttng.org
or on this mailing list.

Project website: https://lttng.org

Download links:
https://lttng.org/files/lttng-tools/lttng-tools-2.5.3.tar.bz2
GPG sig: https://lttng.org/files/lttng-tools/lttng-tools-2.5.3.tar.bz2.asc

Enjoy!
Jérémie

[1] http://lists.lttng.org/pipermail/lttng-dev/2014-November/023832.html

--

-- 
Jérémie Galarneau
EfficiOS Inc.
http://www.efficios.com

_______________________________________________
lttng-dev mailing list
lttng-dev <at> lists.lttng.org
http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev

Gmane