lttng-tools.git
3 years agoFix: action executor: deadlock on session_list_lock
Jonathan Rajotte [Thu, 22 Apr 2021 02:01:02 +0000 (22:01 -0400)] 
Fix: action executor: deadlock on session_list_lock

Observed issue
==============

A dead lock is observed during the start-stop test suite for triggers.

Cause
=====

A start session action is executed by the action executor, the
`cmd_start_trace` function is called and effectively holds the
`session_list_lock.`. During `cmd_start_trace` a call to
`notification_thread_command_add_channel` is performed to inform the
notification thread of the new channel presence.

At the same time, a tracer event notification is received by the
notification thread. The actions are queued up and the sample of the
session id take place and a call to `session_lock_list` is performed and
blocks on the lock operation.

The notification thread wait on the `session_list_lock` and the
`session_list_lock` holder, the action executor, waits on the completion
of a command the be run by the notification thread: deadlock.

The backtrace:

 Thread 6 (Thread 0x7f831c8a6700 (LWP 3046458)):
 #0  syscall () at ../sysdeps/unix/sysv/linux/x86_64/syscall.S:38
 #1  0x000000000053b852 in futex (uaddr=0x7f831c8a45e0, op=0, val=0, timeout=0x0, uaddr2=0x0, val3=0) at /home/joraj/lttng/master/install/include/urcu/futex.h:72
 #2  0x000000000053b4f9 in futex_noasync (uaddr=0x7f831c8a45e0, op=0, val=0, timeout=0x0, uaddr2=0x0, val3=0) at /home/joraj/lttng/master/install/include/urcu/futex.h:81
 #3  0x000000000053af10 in lttng_waiter_wait (waiter=0x7f831c8a45d8) at waiter.c:55
 #4  0x000000000046b0f2 in run_command_wait (handle=0xe60520, cmd=0x7f831c8a4588) at notification-thread-commands.c:49
 #5  0x000000000046b270 in notification_thread_command_add_channel (handle=0xe60520, session_name=0x7f8300006c30 "my_triggered_session", uid=1000, gid=1000, channel_name=0x7f82dc00be04 "channel0", key=1, domain=LTTNG_DOMAIN_UST, capacity=2097152) at notification-thread-commands.c:184
 #6  0x00000000004c7f65 in create_channel_per_uid (app=0x7f82d8000bf0, usess=0x7f8300000bb0, ua_sess=0x7f82dc002600, ua_chan=0x7f82dc00bde0) at ust-app.c:3360
 #7  0x00000000004c6f98 in ust_app_channel_send (app=0x7f82d8000bf0, usess=0x7f8300000bb0, ua_sess=0x7f82dc002600, ua_chan=0x7f82dc00bde0) at ust-app.c:3514
 #8  0x00000000004c6bde in ust_app_channel_create (usess=0x7f8300000bb0, ua_sess=0x7f82dc002600, uchan=0x7f8300005a90, app=0x7f82d8000bf0, _ua_chan=0x7f831c8a48b0) at ust-app.c:4771
 #9  0x00000000004c6968 in find_or_create_ust_app_channel (usess=0x7f8300000bb0, ua_sess=0x7f82dc002600, app=0x7f82d8000bf0, uchan=0x7f8300005a90, ua_chan=0x7f831c8a48b0) at ust-app.c:5610
 #10 0x00000000004c4f09 in ust_app_synchronize_all_channels (usess=0x7f8300000bb0, ua_sess=0x7f82dc002600, app=0x7f82d8000bf0) at ust-app.c:5820
 #11 0x00000000004b958c in ust_app_synchronize (usess=0x7f8300000bb0, app=0x7f82d8000bf0) at ust-app.c:5886
 #12 0x00000000004b8500 in ust_app_global_update (usess=0x7f8300000bb0, app=0x7f82d8000bf0) at ust-app.c:5960
 #13 0x00000000004b7ec2 in ust_app_start_trace_all (usess=0x7f8300000bb0) at ust-app.c:5520
 #14 0x0000000000444e86 in cmd_start_trace (session=0x7f8300006c30) at cmd.c:2707
 #15 0x00000000004a5af9 in action_executor_start_session_handler (executor=0x7f8314004410, work_item=0x7f8314005100, item=0x7f83140050b0) at action-executor.c:342
 #16 0x00000000004a537f in action_executor_generic_handler (executor=0x7f8314004410, work_item=0x7f8314005100, item=0x7f83140050b0) at action-executor.c:696
 #17 0x00000000004a4dbc in action_work_item_execute (executor=0x7f8314004410, work_item=0x7f8314005100) at action-executor.c:715
 #18 0x00000000004a37e6 in action_executor_thread (_data=0x7f8314004410) at action-executor.c:797
 #19 0x0000000000486193 in launch_thread (data=0x7f83140044b0) at thread.c:66
 #20 0x00007f8320b60609 in start_thread (arg=<optimized out>) at pthread_create.c:477
 #21 0x00007f8320a87293 in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95

 Thread 5 (Thread 0x7f831d0a7700 (LWP 3046457)):
 #0  __lll_lock_wait (futex=futex@entry=0x5e1c10 <ltt_session_list>, private=0) at lowlevellock.c:52
 #1  0x00007f8320b630a3 in __GI___pthread_mutex_lock (mutex=0x5e1c10 <ltt_session_list>) at ../nptl/pthread_mutex_lock.c:80
 #2  0x00000000004378c3 in session_lock_list () at session.c:156
 #3  0x00000000004a871c in add_action_to_subitem_array (action=0x7f830001a730, subitems=0x7f83140051d0) at action-executor.c:1081
 #4  0x00000000004a8578 in add_action_to_subitem_array (action=0x7f830001a620, subitems=0x7f83140051d0) at action-executor.c:1025
 #5  0x00000000004a4922 in populate_subitem_array_from_trigger (trigger=0x7f830001a950, subitems=0x7f83140051d0) at action-executor.c:1116
 #6  0x00000000004a416e in action_executor_enqueue_trigger (executor=0x7f8314004410, trigger=0x7f830001a950, evaluation=0x7f8314005190, object_creds=0x0, client_list=0x7f8314004980) at action-executor.c:924
 #7  0x0000000000479481 in dispatch_one_event_notifier_notification (state=0x7f831d0a63e8, notification=0x7f8314005160) at notification-thread-events.c:4613
 #8  0x0000000000472324 in handle_one_event_notifier_notification (state=0x7f831d0a63e8, pipe=65, domain=LTTNG_DOMAIN_UST) at notification-thread-events.c:4702
 #9  0x0000000000472271 in handle_notification_thread_event_notification (state=0x7f831d0a63e8, pipe=65, domain=LTTNG_DOMAIN_UST) at notification-thread-events.c:4717
 #10 0x00000000004695a3 in handle_event_notification_pipe (event_source_fd=65, domain=LTTNG_DOMAIN_UST, revents=1, state=0x7f831d0a63e8) at notification-thread.c:591
 #11 0x000000000046849b in thread_notification (data=0xe60520) at notification-thread.c:727
 #12 0x0000000000486193 in launch_thread (data=0xe60610) at thread.c:66
 #13 0x00007f8320b60609 in start_thread (arg=<optimized out>) at pthread_create.c:477
 #14 0x00007f8320a87293 in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95

Solution
========

Instead of using session_find_by_name() which requires the
`session_list_lock`, we introduce `sample_session_id_by_name` that uses
a urcu backed data structure. This allows the sampling of the session
id without holding the session list lock. We accept the small window
where a session object is still accessible but concretely not valid
since the actual execution context will be validated at the moment of
execution. The execution side already handles the possibility that the
session is removed at that point or is not the same session. The
execution side acquires the session_list_lock for validation.

Known drawbacks
=========

None

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I5ad2c57acc0d03d2814dda59f8ecf2d831fd961e

3 years agolttng add-trigger: replace --exclude-names with --exclude-name
Simon Marchi [Tue, 20 Apr 2021 02:13:28 +0000 (22:13 -0400)] 
lttng add-trigger: replace --exclude-names with --exclude-name

Implement the proposal to make the --exclude-names option singular. The
user can specify it multiple times to specify multiple exclusions,
rather than specifying a comma-separated list.

Change-Id: Id95313bb5576a8d6cecb73b61b537ae190554cd0
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agosessiond: refcount `ust_error_accounting_entry`
Francis Deslauriers [Fri, 16 Apr 2021 15:14:21 +0000 (11:14 -0400)] 
sessiond: refcount `ust_error_accounting_entry`

Keep a reference count of the `ust_error_accounting_entry` to reclaim
memory and cleanup handles as we go.

Triggers on a sessiond target all applications regardless of the UID of
both the trigger and the application. This means that whenever a new UST
counter is created for a UID, we need to keep it around as long as there
is 1. an application from that UID, or 2. an event notifier registered.

This commit achieves that by keeping a count of the registered event
notifiers. If the count is above zero, we keep a reference on every UID
entries. When the count reaches zero, we put that reference on every
entries.

The event notifier count is guarded by a mutex since it's accessed by
both the notification thread and UST registration thread.

Here is the pseudo code for the relevant error accounting events:
App registration:
  lookup entry for application UID in hashtable
  if found:
    get a reference on that existing entry
  else:
    create a new entry for that UID
    get a reference on that new entry
    if event_notifier_count > 0:
      get an extra reference on the behalf of the event notifiers

Event notifier registration:
  increment event_notifier_count
  if event_notifier_count == 1:
    get a reference on all existing UID entries

App unregistration:
  put reference on application UID entry

Event notifier unregistration:
  decrement event_notifier_count
  if event_notifier_count == 0:
    put ref on all existing UID entries

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I74667d30a5b6975bad7975b82f6819e263199cd0

3 years agoaction list: missing renames from previous name "group"
Jérémie Galarneau [Thu, 22 Apr 2021 18:56:59 +0000 (14:56 -0400)] 
action list: missing renames from previous name "group"

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I9b7379d5f2ed66677666365cf5cf0b98c774f219

3 years agoFix: error-query: uninitialized action_index value
Jérémie Galarneau [Thu, 22 Apr 2021 18:50:55 +0000 (14:50 -0400)] 
Fix: error-query: uninitialized action_index value

action_index will be used uninitialized if the user passes an action
that is not part of the trigger (an API use error).

CID 1452662 (#2 of 2): Uninitialized scalar variable (UNINIT)
14. uninit_use: Using uninitialized value action_index.is_set.

Reported-by: Coverity Scan
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I9afc6730b5462f4bf55e0dba3454efb5b4ca22c9

3 years agoFix: sessiond: leak of config_path on duplicate --config option
Jérémie Galarneau [Thu, 22 Apr 2021 18:47:14 +0000 (14:47 -0400)] 
Fix: sessiond: leak of config_path on duplicate --config option

1452207 Resource leak

The system resource will not be reclaimed and reused, reducing the
future availability of the resource.

CID 1452373 (#3 of 3): Resource leak (RESOURCE_LEAK)
26. overwrite_var: Overwriting config_path in config_path =
utils_expand_path(optarg) leaks the storage that config_path points to.

Reported-by: Coverity Scan
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I70eff0f4de5ac9b84b699e88232669259eab54db

3 years agoFix: tests: base notification client: unchecked sscanf return value
Jérémie Galarneau [Thu, 22 Apr 2021 18:43:02 +0000 (14:43 -0400)] 
Fix: tests: base notification client: unchecked sscanf return value

CID 1407934 (#2 of 2): Unchecked return value (CHECKED_RETURN)
8. check_return: Calling sscanf without checking return value (as is
done elsewhere 17 out of 19 times).

Reported-by: Coverity Scan
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ifcef490aa15aea8ae4846553fbc63c596583d274

3 years agoFix: argpar: Error out when passing an argument to long option that takes no argument
Simon Marchi [Fri, 9 Apr 2021 16:48:30 +0000 (12:48 -0400)] 
Fix: argpar: Error out when passing an argument to long option that takes no argument

If the user defines a long option `--foo` and passes `--foo=arg`, the
current behavior is that the option foo is recognized and the argument
is ignored.  Change that to be an error.

It is currently not possible to pass arguments to short options using an
equal sign, so this doesn't apply to short options.

Equivalent accepted change in argpar: https://review.lttng.org/c/argpar/+/5361

Change-Id: I6fd6271abb7e271704db1e60f680729931a1fbe1
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agoClean-up: sessiond: identation fix
Jérémie Galarneau [Thu, 22 Apr 2021 16:37:37 +0000 (12:37 -0400)] 
Clean-up: sessiond: identation fix

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I10e56e226c7ea3eb6eae6a2ea17f5eae7cff3b6c

3 years agoFix: tests: health thread stall: only stop consumerd when required
Jérémie Galarneau [Thu, 22 Apr 2021 16:32:26 +0000 (12:32 -0400)] 
Fix: tests: health thread stall: only stop consumerd when required

Since a0f8e3109, stop_lttng_consumerd will report a failure when
there is no consumer daemon to kill. This fix ensures it is only
invoked for tests that launch a consumer daemon.

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I6831fbe7563d2e1804f10588494d126fbb4202ff

3 years agoFix: tests: notification: update kernel testcase count
Francis Deslauriers [Thu, 22 Apr 2021 15:07:26 +0000 (11:07 -0400)] 
Fix: tests: notification: update kernel testcase count

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ib65e4ea8ced397bdd3b30f1401121adbfeeff23a

3 years agoFix: tests: integer truncation warning on 32-bit platforms
Jérémie Galarneau [Thu, 22 Apr 2021 14:17:04 +0000 (10:17 -0400)] 
Fix: tests: integer truncation warning on 32-bit platforms

large integer implicitly truncated to unsigned type [-Woverflow]

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I3e479d240c25550f44bbf40944484870da85a401

3 years agoFix: sessiond: triggers: failure to insert trigger is fatal
Jérémie Galarneau [Wed, 21 Apr 2021 04:41:09 +0000 (00:41 -0400)] 
Fix: sessiond: triggers: failure to insert trigger is fatal

Inserting a trigger by name/owner should succeed if it could be added
to the top-level triggers_ht.

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I46518b86b34cb13074723b4f2bef38026d898700

3 years agosessiond: notification-thread: switch logging to debug on GET_TRIGGER
Jérémie Galarneau [Wed, 21 Apr 2021 03:57:11 +0000 (23:57 -0400)] 
sessiond: notification-thread: switch logging to debug on GET_TRIGGER

Not finding a trigger is not necessarily an error (espeacially during
the execution of tests).

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I238e6ca42f65e773e7e88773bc2e4f54b0a78ad5

3 years agoTests: add trigger name/unnamed backwards compatibility test
Jérémie Galarneau [Mon, 19 Apr 2021 21:44:55 +0000 (17:44 -0400)] 
Tests: add trigger name/unnamed backwards compatibility test

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ib19de93e316640fb9ba37d2df621c3c97cd712f6

3 years agolttng-ctl: separate support of named/unnamed trigger registration
Jérémie Galarneau [Tue, 20 Apr 2021 02:38:49 +0000 (22:38 -0400)] 
lttng-ctl: separate support of named/unnamed trigger registration

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I37e78344dd14d00c617cd462914dee287e3b24bb

3 years agosessiond: add support for anonymous triggers
Jérémie Galarneau [Tue, 20 Apr 2021 04:43:22 +0000 (00:43 -0400)] 
sessiond: add support for anonymous triggers

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I5b7fb29700af7ac7b633e5d73fb29f99f55ebfe8

3 years agoFix: lttng-ctl: assertion failure during unregistration of trigger
Jérémie Galarneau [Tue, 20 Apr 2021 16:48:05 +0000 (12:48 -0400)] 
Fix: lttng-ctl: assertion failure during unregistration of trigger

Issue observed
==============

lt-trigger_name: trigger.c:302: int lttng_trigger_serialize(const struct lttng_trigger *, struct lttng_payload *): Assertion `(creds->uid).is_set' failed.

Program terminated with signal SIGABRT, Aborted.
 #0  0x00007fb74129eef5 in raise () from /usr/lib/libc.so.6
 #1  0x00007fb741288862 in abort () from /usr/lib/libc.so.6
 #2  0x00007fb741288747 in __assert_fail_base.cold () from /usr/lib/libc.so.6
 #3  0x00007fb741297646 in __assert_fail () from /usr/lib/libc.so.6
 #4  0x00007fb74169bab7 in lttng_trigger_serialize (trigger=0x5616f6f70060, payload=0x7ffe5819d140) at trigger.c:302
 #5  0x00007fb74169cef0 in lttng_trigger_copy (trigger=0x5616f6f70060) at trigger.c:859
 #6  0x00007fb74164302e in lttng_unregister_trigger (trigger=0x5616f6f70060) at lttng-ctl.c:3350
 #7  0x00005616f50c675f in register_named_trigger () at trigger_name.c:295
 #8  0x00005616f50c6879 in main (argc=1, argv=0x7ffe581a07d8) at trigger_name.c:343

Cause
=====

When creating a trigger instance and using it to unregister an existing
trigger, its credentials are unset (meaning 'default'). Expecting this,
lttng_unregister_trigger() copies the source trigger to change its
credentials to those of the caller.

Unfortunately, the trigger copy operation expects credentials to be set.

We don't run into this situation typically since the trigger instance
used to perform the unregistration is sourced from a listing or is the
same instance that was used to perform the registration (which sets the
credentials before serializing).

Solution
========

A proper implementation of "copy" is provided for the trigger object
itself. For its condition and action, we still use the same "trick"
of leveraging the serdes code to perform a deep-copy, keeping the change
small

Drawbacks
=========

None really, except that we lose some of the code sharing between
copy and serdes.

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I71b7b075c959bc4935621543c4d379f62b7dabdf

3 years agoRename group action files to list files
Jonathan Rajotte [Tue, 20 Apr 2021 02:06:33 +0000 (22:06 -0400)] 
Rename group action files to list files

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Iea4c2fe0a2d7001e8c05b4bb33c24e5641b193b0

3 years agoRename action group to action list
Jonathan Rajotte [Tue, 20 Apr 2021 01:57:13 +0000 (21:57 -0400)] 
Rename action group to action list

This is necessary to convey that the `group` an ordered sequence of
actions. We use the term `list` to put an emphasis on the ordered nature
of the group.

Files rename follows.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Idb85d8ca5b60947d8dc3d43d98c8d915bf34c3f2

3 years agoRename on-event to event-rule-matches
Jonathan Rajotte [Tue, 20 Apr 2021 01:22:09 +0000 (21:22 -0400)] 
Rename on-event to event-rule-matches

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Icb596abdd0c291643f144ce274919b4094828e95

3 years agoMove on-event*.* to event-rule-matches*.*
Jonathan Rajotte [Tue, 20 Apr 2021 01:03:40 +0000 (21:03 -0400)] 
Move on-event*.* to event-rule-matches*.*

Code renaming follows.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ifb2de118546d39877e77ba30ba8720f75240eb39

3 years agoIntroduce emission site for syscall event rule
Jonathan Rajotte [Fri, 16 Apr 2021 21:24:20 +0000 (17:24 -0400)] 
Introduce emission site for syscall event rule

The tracer support multiple emission site configuration: entry, exit,
both.

On the CLI, the emission site is exposed directly on the event rule type
with the following syntax:

 --type=syscall:entry
 --type=syscall:exit
 --type=syscall:entry+exit

`--type=syscall` is equivalent to `--type=syscall:entry+exit`.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I65faea6872f121b98b781d931b6cc7301f20a0b6

3 years agolttng list-triggers: handle all condition types
Simon Marchi [Thu, 15 Apr 2021 13:07:31 +0000 (09:07 -0400)] 
lttng list-triggers: handle all condition types

At the moment, only the event-rule-matches condition is handled by the
list-triggers commands. Change it to handle all existing condition
types.

Because these other condition types can't be created using the
command-line interface, add a util program that creates and registers
triggers for the purpose of the test. The test triggers with the
event-rule-matches condition are kept as is, created using the
command-line interface.

Change-Id: I6a2b0442800c93f7e8815dce96d1cf1928d5725a
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agoFix: sessiond: trigger with condition not requiring event notifier cannot be listed
Jonathan Rajotte [Mon, 19 Apr 2021 18:04:53 +0000 (14:04 -0400)] 
Fix: sessiond: trigger with condition not requiring event notifier cannot be listed

Observed issue
==============

The lttng-sessiond process aborts when an lttng error query is performed
against a trigger that uses a condition other than "event-match".

Cause
=====

The trigger's condition is a LTTNG_CONDITION_TYPE_SESSION_CONSUMED_SIZE
which does not have a "domain type".

This results in the call to `abort()` in
`event_notifier_error_accounting_get_count` of the default case.

Solution
========

Check if the trigger "needs" or at least "depends" on tracer notifier.

Known drawbacks
=========

None

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I3d5f09617f95315060a611b464d1df95cb80c5bd

3 years agolttng add-trigger: remove support for conditions other than event-rule-matches
Simon Marchi [Thu, 15 Apr 2021 11:49:16 +0000 (07:49 -0400)] 
lttng add-trigger: remove support for conditions other than event-rule-matches

We initially only want to support the event-rule-matches condition, so
remove any trace of the other ones. They'll be added later as needed.

Change-Id: I66b79bea89257187865ebf599ef0341279629ec6
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agolttng {add,list,remove}-trigger: rename user id to owner uid
Simon Marchi [Thu, 15 Apr 2021 11:31:01 +0000 (07:31 -0400)] 
lttng {add,list,remove}-trigger: rename user id to owner uid

Implement the proposal to rename the --user-id option to --owner-uid in
the add-trigger and remove-trigger commands.  Update the output of
list-triggers as well.

Change-Id: I79977a0efe80d96917fcaf55909a78c81047c388
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agolttng add-trigger: replace log level options with --log-level
Simon Marchi [Thu, 15 Apr 2021 02:28:07 +0000 (22:28 -0400)] 
lttng add-trigger: replace log level options with --log-level

Implement the proposal to replace the --loglevel and --loglevel-only
options with a single --log-level option.

The syntax of the argument to --log-level is:

 - `LEVEL`: to specify only this level
 - `LEVEL..`: to specify from this level up to the most critical level
 - `..`: to specify any level. Basically the same as no `--log-level`
         option.

The intent is to keep the door open to fully support ranges with the
`LEVEL1..LEVEL2` syntax.  The loglevel_parse_range_string function and
friends are implement this, but their caller in add_trigger.c,
parse_log_level_string, artificially restricts it to the two cases
shown above.

Change-Id: Id8607b3dae2db5aace263f0c56ada2e822db907c
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agolttng add-trigger: replace event rule type options with --type
Simon Marchi [Wed, 14 Apr 2021 02:21:56 +0000 (22:21 -0400)] 
lttng add-trigger: replace event rule type options with --type

Implement the proposal to replace the --function, --probe,
--userspace-probe, --syscall and --tracepoint options, used to select an
event rule type, with a single --type option.

The --probe, --userspace-probe and --function options used to take an
argument, the name of the location.  For these types, the --location
option is introduced to specify that information.

Change-Id: Ib62e502beb6832a3d26cb135ca6b2746382f2075
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agolttng add-trigger: replace domain options with single --domain
Simon Marchi [Tue, 13 Apr 2021 14:32:21 +0000 (10:32 -0400)] 
lttng add-trigger: replace domain options with single --domain

Implement the proposal to replace the --userspace, --kernel, --log4j,
--jul and --python options, used to select a domain, with a single
--domain option that takes an argument.

Change-Id: I9c5ff77523bb974bc1facb6732f3122c2e30395e
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agolttng add-trigger: some updates to event rule to new syntax
Simon Marchi [Mon, 12 Apr 2021 16:25:50 +0000 (12:25 -0400)] 
lttng add-trigger: some updates to event rule to new syntax

Implement a few of the proposals for the event rule syntax related to
event names that we want to adopt for the add-trigger command.

 - Remove the positional argument (tracepoint name).
 - Add --name option for tracepoints, syscalls and loggers.
 - Add --event-name option for probes.
 - Rename --exclude to --exclude-names.
 - Remove the --all/-a option.  The equivalent now is to omit
   --name, which is the equivalent of --name=*.

Change-Id: I7161570a47716c458a60d679bc229a8c8843b04f
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agolttng add-trigger: rename on-event condition to event-rule-matches
Simon Marchi [Thu, 15 Apr 2021 01:33:46 +0000 (21:33 -0400)] 
lttng add-trigger: rename on-event condition to event-rule-matches

Implement the proposal to rename the on-event condition to
event-rule-matches. This patch only updates the CLI, the API will be
renamed later.

Change-Id: I341d11922998043414028477fb65cfbb5a2d66cd
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agolttng {add,list}-triggers: use `name` instead of `id`
Simon Marchi [Fri, 9 Apr 2021 18:46:42 +0000 (14:46 -0400)] 
lttng {add,list}-triggers: use `name` instead of `id`

The right term is "trigger name", not "trigger id".  Change the option
to add-trigger accordingly, as well as the output of list-trigger.

Change-Id: I5c304aea2c37f983de001cb11c5ba8859f564617
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agolttng add-trigger: parse condition / action name as option arguments
Simon Marchi [Fri, 9 Apr 2021 17:14:44 +0000 (13:14 -0400)] 
lttng add-trigger: parse condition / action name as option arguments

Change the `add-trigger` command's --condition and --action options to
take the condition name and action name as option arguments.  Currently,
when we see the --condition option, we go into the parse_condition
function and expect the first argument to be the condition name (same
idea for the action).  Since the condition/action name is mandatory
after --condition/--action, it is simpler to say that --condition and
--action require an argument.

This will let the user do:

    --condition=on-event

in addition to the current syntax:

    --condition on-event

The code is a bit cleaner as a result.

Change-Id: Ic7ea97c2657e5de6ae18563694ad22bbffbb2aa8
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agoAdd .editorconfig file
Simon Marchi [Wed, 14 Apr 2021 18:49:54 +0000 (14:49 -0400)] 
Add .editorconfig file

Many editors support reading this file to automatically configure indent
/ whitespace settings. The author of this patch would be extremely
grateful if the maintainer of the project would consider merging the
patch in the master branch of the project's repository.

Change-Id: I84ae62fe3e028d0d40421943b3a53207854efed7
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agoAllow users to create trigger with their uid
Francis Deslauriers [Thu, 15 Apr 2021 19:46:18 +0000 (15:46 -0400)] 
Allow users to create trigger with their uid

Currently, the `add-trigger` prints the following if a user uses the
`--user-id` with its own uid:
  Error: Failed to set trigger's user identity.

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ie0a642b3faf2259a892def3cb19a899e61cb4705

3 years agolttng: list-triggers: implement listing of SDT userspace-probe
Francis Deslauriers [Wed, 14 Apr 2021 21:25:27 +0000 (17:25 -0400)] 
lttng: list-triggers: implement listing of SDT userspace-probe

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Id57c265dcdb3b324ffbb274ca97ead9364275cf0

3 years agoTests: cleanup: rework trigger listing tests
Francis Deslauriers [Thu, 15 Apr 2021 13:16:38 +0000 (09:16 -0400)] 
Tests: cleanup: rework trigger listing tests

- Rename `list_triggers()` -> `list_triggers_matches_ok()` to make it
  obvious it's testing something,
- Add `diag` to all test cases,
- Launch only one lttng-sessiond for all test cases to reduce test
  runtime, and
- remove-trigger all triggers at the end of every test cases.

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Icbbd08f521565ef1cae901bb192cf6f8f1cdb7ea

3 years agoTests: event-notifier: multi UID error accounting support
Francis Deslauriers [Wed, 14 Apr 2021 17:09:15 +0000 (13:09 -0400)] 
Tests: event-notifier: multi UID error accounting support

The new test case adds one root trigger and one regular user trigger as
well as apps from both of these users.

The test then confirm that the error accounting of the root trigger
reports what is happening on the regular user's app as well as its own
app.

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Iaec231863de98b14891602bdae010449989de6f9

3 years agoFix: tests: quote variable in case it's not set
Francis Deslauriers [Mon, 19 Apr 2021 18:58:31 +0000 (14:58 -0400)] 
Fix: tests: quote variable in case it's not set

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ib0751699eb9deb02f47739062c99dbb8535c3163

3 years agoust: error accounting: don't print error messages if app is dead
Francis Deslauriers [Wed, 21 Apr 2021 20:04:25 +0000 (16:04 -0400)] 
ust: error accounting: don't print error messages if app is dead

UST applications may exit during the registering process. It's not an
error and should be expected.

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I90f91252849ba362e73dad6c30b9cfdc0185abe4

3 years agosessiond: split event notifier error counter CLI options
Francis Deslauriers [Mon, 12 Apr 2021 15:39:08 +0000 (11:39 -0400)] 
sessiond: split event notifier error counter CLI options

With this commit, users can specify the size of event notifier error
counting buffers for each domain independently using the following new
lttng-sessiond options:
  --event-notifier-error-buffer-size-kernel=
  --event-notifier-error-buffer-size-userspace=

The index allocation is now also per-domain meaning that index
allocation in the kernel domain doesn't affect available indices in the
userspace domain and reversely.

Small changes:
- Add manual page description of the new options,
- Rename `struct error_account_entry` to
  `struct ust_error_accounting_entry`.

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I4758d32c00cb432be377fd67eaffa11b193bad74

3 years agoClean-up: sessiond: error-query: remove unneeded goto
Jérémie Galarneau [Wed, 21 Apr 2021 19:39:09 +0000 (15:39 -0400)] 
Clean-up: sessiond: error-query: remove unneeded goto

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ie0d02a5151b359ffb1e9ce487811d8235c4a3d9d

3 years agoFix: sessiond: action-executor: uninitialized session_name used in logging
Jérémie Galarneau [Wed, 21 Apr 2021 19:36:04 +0000 (15:36 -0400)] 
Fix: sessiond: action-executor: uninitialized session_name used in logging

CID 1452636 (#1 of 1): Uninitialized pointer read (UNINIT)
8. uninit_use_in_call: Using uninitialized value session_name when
calling fprintf

Reported-by: Coverity Scan
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I497c613e8cb1342948b726cc5c196e46d61d83dd

3 years agoClean-up: lttng: add-trigger: silence coverity warning
Jérémie Galarneau [Wed, 21 Apr 2021 19:31:38 +0000 (15:31 -0400)] 
Clean-up: lttng: add-trigger: silence coverity warning

CID 1452634 (#1 of 1): Resource leak (RESOURCE_LEAK)
23. leaked_storage: Variable error going out of scope leaks the storage
it points to.

The code path shown by Coverity does not appear reachable, but this
fixes the warning nonetheless.

Reported-by: Coverity Scan
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I9682616b123de108c5717cee2c17349212f224d5

3 years agoFix: error-query: leak of trigger on malformed error-query comm buffer
Jérémie Galarneau [Wed, 21 Apr 2021 19:28:31 +0000 (15:28 -0400)] 
Fix: error-query: leak of trigger on malformed error-query comm buffer

CID 1452633 (#1 of 1): Resource leak (RESOURCE_LEAK)
10. leaked_storage: Variable trigger going out of scope leaks the
storage it points to

Reported-by: Coverity Scan
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ic3ab0c41b6667fb875992e545fcba046edc071fc

3 years agoFix: error-query: leak of trigger on allocation error
Jérémie Galarneau [Wed, 21 Apr 2021 18:39:52 +0000 (14:39 -0400)] 
Fix: error-query: leak of trigger on allocation error

1452630 Resource leak

The system resource will not be reclaimed and reused, reducing the
future availability of the resource.

In lttng_error_query_action_create: Leak of memory or pointers to system
resources (CWE-404)

CID 1452630 (#1 of 1): Resource leak (RESOURCE_LEAK)
21. leaked_storage: Variable trigger_copy going out of scope leaks the
storage it points to.

Reported-by: Coverity Scan
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I4dd99390f1fd5ad957ab6500971dd2ed67a1c722

3 years agoClean-up: tests: notification-client: unused assignment
Jérémie Galarneau [Wed, 21 Apr 2021 18:34:46 +0000 (14:34 -0400)] 
Clean-up: tests: notification-client: unused assignment

1452629 Unused value

An assigned value that is never used may represent unnecessary
computation, an incorrect algorithm, or possibly the need for cleanup or
refactoring.

In main: A value assigned to a variable is never used. (CWE-563)

CID 1452629 (#1 of 1): Unused value (UNUSED_VALUE)assigned_value:
Assigning value -1 to ret here, but that stored value is overwritten
before it can be used

Reported-by: Coverity Scan
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I0d80a244a7b2b3d6aed57d58ca5435230d664f99

3 years agoFix: relayd: leak of config_path on duplicate --config option
Jérémie Galarneau [Wed, 21 Apr 2021 18:30:02 +0000 (14:30 -0400)] 
Fix: relayd: leak of config_path on duplicate --config option

1452207 Resource leak

The system resource will not be reclaimed and reused, reducing the
future availability of the resource.

In lttng_notification_create_from_payload: Leak of memory or pointers to
system resources (CWE-404)

Reported-by: Coverity Scan
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ib4f6b388f4ff24eedcbff64570bd9fa10a445304

3 years agoFix: lttng-ctl: trigger leak on failure to deserialize evaluation
Jérémie Galarneau [Wed, 21 Apr 2021 18:22:08 +0000 (14:22 -0400)] 
Fix: lttng-ctl: trigger leak on failure to deserialize evaluation

1452207 Resource leak

The system resource will not be reclaimed and reused, reducing the
future availability of the resource.

In lttng_notification_create_from_payload: Leak of memory or pointers to
system resources (CWE-404)

Reported-by: Coverity Scan
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I5a66af837d46ca4b8e1909012cd28dc8ea990bad

3 years agoFix: tests: discarded notifications: missing `list-triggers` call
Francis Deslauriers [Tue, 20 Apr 2021 13:52:00 +0000 (09:52 -0400)] 
Fix: tests: discarded notifications: missing `list-triggers` call

I get the following test failure when running the
test_notification_notifier_discarded_count test case.

  # Kernel event notifer error counter
  ok 1 - Add trigger my_trigger
  ok 2 - No discarded tracer notification message
  ok 3 - Trigger 'errors: none' notification message
  ok 4 - Tracer notification discarded line printed
  ok 5 - Discarded tracer notification number non-zero as expected
  ok 6 - Remove trigger my_trigger
  ok 7 - No `on-event` kernel notifier enabled as expected
  ok 8 - Add trigger my_trigger
  ok 9 - No discarded tracer notification message
  not ok 10 - Trigger 'errors: none' notification message
   #   Failed test 'Trigger 'errors: none' notification message'
   #   in tests/regression/tools/notification/test_notification_notifier_discarded_count:test_kernel_notifier_discarded_count() at line 108.
  ok 11 - Remove trigger my_trigger

This happens because we test the `list-triggers` output on outdated
content. We need to call `list-triggers` right before testing the output
of it.

In this commit, I also changed the number of iteration to 2000 as it's
sufficient to generate discarded notifications.

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I66d54f9178f9f37ff2fd899278e77fe50e455c10

3 years agoTests: fix: notifier discarded count: typo results in failure
Jérémie Galarneau [Tue, 20 Apr 2021 05:01:47 +0000 (01:01 -0400)] 
Tests: fix: notifier discarded count: typo results in failure

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I94b8c60764e1a0d79b25fd252c1a344d807fa61f

3 years agoCleanup: tests: test_userspace_probe: remove extra whitespaces
Francis Deslauriers [Wed, 14 Apr 2021 21:30:13 +0000 (17:30 -0400)] 
Cleanup: tests: test_userspace_probe: remove extra whitespaces

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I6e890a6c1927feb4a09f46b200294cb2faf8a580

3 years agoCleanup: add `lttng_ht_type_str()`
Francis Deslauriers [Wed, 20 Jan 2021 21:18:09 +0000 (16:18 -0500)] 
Cleanup: add `lttng_ht_type_str()`

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I1f7d9ae029f11a454cbda6b05eb22457d6b0be74

3 years agoFix: lttng-ctl: appending to dynamic buffer invalidates its data member
Francis Deslauriers [Thu, 15 Apr 2021 01:19:41 +0000 (21:19 -0400)] 
Fix: lttng-ctl: appending to dynamic buffer invalidates its data member

Issue
=====
The following commands fail:
  lttng add-trigger --id T0 --condition on-event -u some-event --action snapshot-session ze-session3 --path /some/path
  lttng remove-trigger T0

  Error: Attempt to create buffer view from another view with invalid length (length > space left after offset in source): source size = 0, offset in source = 0, length = 25
  Error: Invalid trigger received as part of command payload

Valgrind complains in the following way:

==706109==
==706109== Invalid write of size 4
==706109==    at 0x489FED7: lttng_unregister_trigger (lttng-ctl.c:3281)
==706109==    by 0x43C175: cmd_remove_trigger (remove_trigger.c:171)
==706109==    by 0x43F56B: handle_command (lttng.c:237)
==706109==    by 0x43E9B1: parse_args (lttng.c:421)
==706109==    by 0x43E158: main (lttng.c:470)
==706109==  Address 0x73d8d20 is 4,688 bytes inside a block of size 16,384 free'd
==706109==    at 0x483DFAF: realloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==706109==    by 0x48C1478: lttng_dynamic_buffer_set_capacity (dynamic-buffer.c:166)
==706109==    by 0x48C138C: lttng_dynamic_buffer_append (dynamic-buffer.c:55)
==706109==    by 0x48E3325: lttng_snapshot_output_serialize (snapshot.c:120)
==706109==    by 0x48B46C3: lttng_action_snapshot_session_serialize (snapshot-session.c:173)
==706109==    by 0x48B1FB2: lttng_action_serialize (action.c:130)
==706109==    by 0x48B2DFE: lttng_action_group_serialize (group.c:165)
==706109==    by 0x48B1FB2: lttng_action_serialize (action.c:130)
==706109==    by 0x48ECE66: lttng_trigger_serialize (trigger.c:372)
==706109==    by 0x489FEA0: lttng_unregister_trigger (lttng-ctl.c:3275)
==706109==    by 0x43C175: cmd_remove_trigger (remove_trigger.c:171)
==706109==    by 0x43F56B: handle_command (lttng.c:237)
==706109==  Block was alloc'd at
==706109==    at 0x483B723: malloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==706109==    by 0x483E017: realloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==706109==    by 0x48C1478: lttng_dynamic_buffer_set_capacity (dynamic-buffer.c:166)
==706109==    by 0x48C138C: lttng_dynamic_buffer_append (dynamic-buffer.c:55)
==706109==    by 0x489FE66: lttng_unregister_trigger (lttng-ctl.c:3263)
==706109==    by 0x43C175: cmd_remove_trigger (remove_trigger.c:171)
==706109==    by 0x43F56B: handle_command (lttng.c:237)
==706109==    by 0x43E9B1: parse_args (lttng.c:421)
==706109==    by 0x43E158: main (lttng.c:470)

`lttng_unregister_trigger` samples the address of the lsm header in the
message payload. However, it does so before calling
`lttng_trigger_serialize()` which may increase the underlying buffer's
size (and cause a realloc()).

Most of the time the message buffer is large enough _or_ its realloc
yields the same address which hid the problem. However, I stumbled on a
case (a trigger which snapshots to a location) where the realloc ends-up
returning a completely different address, causing invalid data to be
sent to the session daemon.

Solution
========
Sample the lsm header address after the serialization of the trigger.

Note
====
An identical fix was done for the `lttng_register_trigger` function in:
  commit b22f4f54e95ae13edda1d4d5efd1e4845a6319c4
  Author: Jérémie Galarneau <jeremie.galarneau@efficios.com>
  Date:   Thu Feb 18 18:13:19 2021 -0500

      Fix: lttng-ctl: appending to dynamic buffer invalidates its data member

I reuse the bug explanation for this commit message.

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ic50c96dcada9e0595b0fab1d2f357c183b53e1de

3 years agoFix: ust-app: error counter index never set
Francis Deslauriers [Wed, 14 Apr 2021 14:50:28 +0000 (10:50 -0400)] 
Fix: ust-app: error counter index never set

Issue
=====
The `struct ust_app_event_notifier_rule::error_counter_index` field is
currently never set so all error counting is done in bucket index 0.

Fix
===
Set the error counter index.

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I951095dee5586111699b0ea0e3bc26c0fb4a535d

3 years agoFix: lttng: -Wshadow error in cmd_snapshot
Francis Deslauriers [Wed, 14 Apr 2021 14:12:04 +0000 (10:12 -0400)] 
Fix: lttng: -Wshadow error in cmd_snapshot

The optarg variable name shadows a variable in an external dependency.
clang returns the following error on my machine:
  commands/snapshot.c:627:16: error: declaration shadows a variable in the global scope [-Werror,-Wshadow]
                          const char *optarg = poptGetOptArg(pc);
                                      ^
  /usr/include/x86_64-linux-gnu/bits/getopt_core.h:36:14: note: previous declaration is here
  extern char *optarg;

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I85dd7b0761ebc2c023d84ba869c0551f91f38a17

3 years agoFix: python: -Wshadow warning
Francis Deslauriers [Wed, 14 Apr 2021 19:14:45 +0000 (15:14 -0400)] 
Fix: python: -Wshadow warning

Fixes warning when building Python Bindings with SWIG 3.0.12.

  lttng_wrap.c: In function ‘SWIG_Python_NewShadowInstance’:
  lttng_wrap.c:2483:65: warning: declaration of ‘swig_this’ shadows a global declaration [-Wshadow]
   SWIG_Python_NewShadowInstance(SwigPyClientData *data, PyObject *swig_this)
                                                                   ^~~~~~~~~
  lttng_wrap.c:2224:18: note: shadowed declaration is here
   static PyObject *swig_this = NULL;
                    ^~~~~~~~~
  lttng_wrap.c: In function ‘SWIG_Python_SetSwigThis’:
  lttng_wrap.c:2558:51: warning: declaration of ‘swig_this’ shadows a global declaration [-Wshadow]
   SWIG_Python_SetSwigThis(PyObject *inst, PyObject *swig_this)
                                                     ^~~~~~~~~
  lttng_wrap.c:2224:18: note: shadowed declaration is here
   static PyObject *swig_this = NULL;

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Icde5f124e914a158eb190f95e3c012b7f800d7d1

3 years agoFix: kernel consumerd: use of uninitialized variable
Jérémie Galarneau [Mon, 19 Apr 2021 15:07:12 +0000 (11:07 -0400)] 
Fix: kernel consumerd: use of uninitialized variable

‘ret_add_channel’ may be used uninitialized in this function [-Wmaybe-uninitialized]

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I1db71aef9a9585d8737347f8375e688994f56c5f

3 years agoClean-up: sessiond: cmd: abort() on invalid query target
Jérémie Galarneau [Mon, 19 Apr 2021 15:06:23 +0000 (11:06 -0400)] 
Clean-up: sessiond: cmd: abort() on invalid query target

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ia02fb651a18d927c3b04e20bb55aa588d071b87c

3 years agoClean-up: sessiond: cmd: fix bogus -Wmaybe-uninitialized
Jérémie Galarneau [Mon, 19 Apr 2021 15:00:23 +0000 (11:00 -0400)] 
Clean-up: sessiond: cmd: fix bogus -Wmaybe-uninitialized

Old gcc versions report:
cmd.c:4751: ‘query_target_action’ may be used uninitialized in this function [-Wmaybe-uninitialized]

For this to be reachable, a query target would have to change suddenly
during the execution of the function.

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Id90ca37e232385f73d54558f2a00d01cbf773a92

3 years agoRemove error count property of lttng_condition_on_event
Jérémie Galarneau [Thu, 15 Apr 2021 17:21:14 +0000 (13:21 -0400)] 
Remove error count property of lttng_condition_on_event

Errors related to dropped tracer messages are not reported through
the error query interface.

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ib5f8bf841f37d98b582209abb44ab8a26c8fbbc6

3 years agolttng: list-triggers: retrieve trigger errors using error queries
Jérémie Galarneau [Thu, 15 Apr 2021 17:14:37 +0000 (13:14 -0400)] 
lttng: list-triggers: retrieve trigger errors using error queries

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I1a80125122e2568491d940c598d8445410bf58d9

3 years agosessiond: implement EXECUTE_ERROR_QUERY command
Jérémie Galarneau [Thu, 15 Apr 2021 17:11:09 +0000 (13:11 -0400)] 
sessiond: implement EXECUTE_ERROR_QUERY command

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I3404e85e18b0b4e5d51b620c3a082736d3190066

3 years agolttng-ctl: Add error query interface
Jérémie Galarneau [Thu, 15 Apr 2021 17:09:35 +0000 (13:09 -0400)] 
lttng-ctl: Add error query interface

Change-Id: Ibb0ba800330976aad7b7c501f43a0d23a0dba027
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agoadd-trigger: rename --fire-* to --rate-policy=*:value
Jonathan Rajotte [Thu, 15 Apr 2021 21:35:20 +0000 (17:35 -0400)] 
add-trigger: rename --fire-* to --rate-policy=*:value

With the move to firing policy to the action and the renaming to rate
policy, a rename of the CLI options is necessary.

We introduce the following format:

  --rate-policy=<type>:<value>

Where type is either: once-after or every.
Value is an unsigned long long value.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I6df264c536293e686033d0baa8fe870c75c28b8d

3 years agoRename firing policy to rate policy
Jonathan Rajotte [Thu, 15 Apr 2021 03:18:25 +0000 (23:18 -0400)] 
Rename firing policy to rate policy

This is necessary since that now that the policy is moved to the actions
it is a bit weird to use the term "firing".

CLI change is done in a later patch.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Id6940a49cd0640a8fa7145dab5b805fcfa515f40

3 years agoaction executor: use an execution context to validate enqueued action before execution
Jonathan Rajotte [Tue, 13 Apr 2021 18:49:52 +0000 (14:49 -0400)] 
action executor: use an execution context to validate enqueued action before execution

Problem
======

Let's walk over the following scenario:

Trigger my_trigger:
  condition: ust event a
  action: start session S

At time T, session S exists.

At T + 1, the event A is hit.

At T + 2, the tracer event notification is received and the work item is
queued. Here session S have an id of 1.

At T + 3, the session S is destroyed and a new session S is created,
with a resulting id of 200.

At T +4, the work item is popped from the queue and begin execution and
will start session S with an id of 200 instead of the session S id 1
that was present at the queuing phase.

We effectively want to assign an execution context to an action at the
queuing phase and validate that the execution context is still valid
upon actual execution.

Solution
=====

Instead of iterating over the trigger's actions, we prepare an array of
work sub-item (action_work_subitem) that effectively is a flattened and
augmented array of the trigger's action. For now only the session id is
an action level execution context.

Each action handler is responsible of validating the execution context
before performing the action.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I72fbe8d52d77d1d696fa7d99b662b77e9dad02d5

3 years agotrigger/action executor: move trigger registration state change to client thread
Jonathan Rajotte [Mon, 12 Apr 2021 20:38:16 +0000 (16:38 -0400)] 
trigger/action executor: move trigger registration state change to client thread

Set the unregistered state of the sessiond side trigger object in the
client thread since we want to minimize the possibility of the
notification thread being stalled due to a long execution of an action
in the action executor thread.

Stalling the client thread is, comparatively, less harmful.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I7de875996b430e49ec50896f8b26a75af7993c74

3 years agonotification-thread: add `GET_TRIGGER` command
Francis Deslauriers [Mon, 8 Mar 2021 22:19:51 +0000 (17:19 -0500)] 
notification-thread: add `GET_TRIGGER` command

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ifc94ea894999b9da796aa4fda0d929d237e12d3e

3 years agoaction executor: execute action only if the associated trigger is registered
Jonathan Rajotte [Mon, 12 Apr 2021 19:41:36 +0000 (15:41 -0400)] 
action executor: execute action only if the associated trigger is registered

It was decided that a trigger removal should have immediate effect and
not result in a waiting on action queued to be executed.

There was two major way to do this:

1) On unregistrer trigger, lock the action executor queue and remove
all work-item related to the trigger ebing unregistered.

2) Mark the trigger as "unregistered" and validate its state when a work
item is ready for execution. If the trigger is marked as "unregistered",
simply consume and skip the work item.

To isolate the un-registration and the execution of a work item
depending on that trigger, we use the trigger lock to have exclusivity
during the execution for that trigger.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I9f42a323e0ec0dc51339f39f07536ec8060dc02a

3 years agotrigger: keep state of if a trigger is currently registered
Jonathan Rajotte [Mon, 12 Apr 2021 18:45:24 +0000 (14:45 -0400)] 
trigger: keep state of if a trigger is currently registered

Since a trigger can be referenced even when is was "unregistered" in
other part of lttng-sessiond, namely the action executor queue, we must
keep track of the registration state.

This will allows us to easily skip any actions to be executed
if the associated trigger is "unregistered" at the moment of
execution. This is implemented in a following patch.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I06c9d437fce975c9c8393e8d1be3e66a24618af3

3 years agoTests: fix: leak of notification-client arguments
Jérémie Galarneau [Sat, 17 Apr 2021 22:47:16 +0000 (18:47 -0400)] 
Tests: fix: leak of notification-client arguments

Valgrind reports the following leaks:

==334435== HEAP SUMMARY:
==334435==     in use at exit: 8,568 bytes in 24 blocks
==334435==   total heap usage: 846 allocs, 822 frees, 1,425,964 bytes allocated
==334435==
==334435== 9 bytes in 1 blocks are definitely lost in loss record 1 of 11
==334435==    at 0x483E77F: malloc (vg_replace_malloc.c:307)
==334435==    by 0x4BCC5BE: strdup (in /usr/lib/libc-2.33.so)
==334435==    by 0x10A48B: main (notification-client.c:123)
==334435==
==334435== 13 bytes in 1 blocks are definitely lost in loss record 2 of 11
==334435==    at 0x483E77F: malloc (vg_replace_malloc.c:307)
==334435==    by 0x4BCC5BE: strdup (in /usr/lib/libc-2.33.so)
==334435==    by 0x10A473: main (notification-client.c:120)
==334435==
==334435== 27 bytes in 1 blocks are definitely lost in loss record 3 of 11
==334435==    at 0x483E77F: malloc (vg_replace_malloc.c:307)
==334435==    by 0x4BCC5BE: strdup (in /usr/lib/libc-2.33.so)
==334435==    by 0x10A444: main (notification-client.c:114)

Free the strdup()'ed' arguments.

Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I21523e847141aef50858d3018f37f0125dee0e2c

3 years agotests: trigger action firing policy
Jonathan Rajotte [Tue, 30 Mar 2021 02:00:37 +0000 (22:00 -0400)] 
tests: trigger action firing policy

Implements the regression tests for the trigger action firing polices.

For now the focus is put on the notify action behavior. In the future,
the ability to fetch the number of time an action was enqueued for
execution and the number of type it was executed should greatly
facilitate this work. For now, use the notify action for minimal
regression testing.

gen-ust-events is modified to allow a caller to specify that an `end`
event should be emitted at the end of the program. This is used to
guaranteed that all previous event tracer notifications should be
received when the `end` event notification is received. Note that
ordering is guaranteed for delivery of tracer notification and
notification for a given process.

notificiation-client is modified to allow reception of N notifications
and to validate that it received the count it expects.

notificiation-client is modified to allow reception of N notifications
and to exit only when an `end` event notification is received. This
allow us validate the firing policy.

The tests are pretty straightforward. We register 2 triggers, one with
an event rule condition on `tp:tptest` and the other one with a event
rule condition on `tp:end`. From there we perform a phase of event hits
were we expect no notification except for the `tp:end` then we force a
single event hit were we expect a notification for `tp:tptest`. For the
fire every N policy, we rinse and repeat to validate the behaviour. For
the once after N, we simply do a third phase were we generate N events
and expect no notification.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ia3c88f38dcca0389165e1416b06768338b2889b6

3 years agoMove firing policy from lttng_trigger to lttng_action
Jonathan Rajotte [Tue, 30 Mar 2021 01:43:12 +0000 (21:43 -0400)] 
Move firing policy from lttng_trigger to lttng_action

After some reflection on the future of the importance of the trigger
feature and the centralization we wish to carry around it, it is
required that the notion of firing policy be moved from the trigger
object to each action object of a trigger.

This is necessary since we plan on introducing tracer side actions, such
as increment value of map. Controlling the firing policy on the tracer
side is not an easy thing to do since for UST tracing a lot a
synchronizations must take place and also we must define the behaviour
when multiple apps are present. Hence, we need a way to ensure that we
are not painting ourself in a corner. The middle ground that was chosen
was to move the notion of firing policy to the action object level. This
allows us to scope the concept to the action and decide for each type if
firing policy can be supported and, as needed, define the behaviour per
action type.

Essentially this patch perform the complete transition. It removes the
notion of firing policy at the trigger level and exposes the firing
policy of each action type if applicable.

CLI
======

For the `add-trigger` command the change essentially boils down to
moving the `--fire-every` and `--fire-once-after` from a top-level
parsing to the parsing of each actions. Yes, for now all actions
supports the `--fire-*` options but it will not be the case in the
future. A side effect of this is that a user could decide to have
different firing policy for each actions, but this also mean that if a
user want to apply the same firing policy across actions, the user needs
to specify it for each actions. This could be solved in the future as
the trigger feature matures and that common ground are found in the
behaviour or implementation of the actions (tracer side action, async
action, sync actions etc.) such that "syntactic sugar" options emerge.

As for the `list-trigger`, we move the firing policy printing to each
actions.

Tests have been updated to reflect the changes.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ib29d4319922096c0c4b3f00782f3bbefb17e2f40

3 years agoaction-executor: consider action firing policy on action execution
Jonathan Rajotte [Wed, 7 Apr 2021 13:19:18 +0000 (09:19 -0400)] 
action-executor: consider action firing policy on action execution

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I83f73e905e87a401187e243ed3beafa41964434a

3 years agotest: unit: snapshot session action
Jonathan Rajotte [Thu, 8 Apr 2021 19:48:39 +0000 (15:48 -0400)] 
test: unit: snapshot session action

Note that the introduced test are mostly toward the firing policy
parameter.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I7685610f963576e50747a1801c4e0cd754e241b8

3 years agoImplement firing policy for the snapshot session action
Jonathan Rajotte [Thu, 8 Apr 2021 17:51:35 +0000 (13:51 -0400)] 
Implement firing policy for the snapshot session action

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ia849bf101cbdf09977adc0781809139a3c769ab2

3 years agotests: units: action: stop session
Jonathan Rajotte [Thu, 8 Apr 2021 13:35:05 +0000 (09:35 -0400)] 
tests: units: action: stop session

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ib8b230857180966085d682a29e54f6baa525258f

3 years agoImplement firing policy for stop session action
Jonathan Rajotte [Thu, 8 Apr 2021 13:34:43 +0000 (09:34 -0400)] 
Implement firing policy for stop session action

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Iaed5d8c3cc59b2284692ed56f1b4a8ba69736fd6

3 years agotests: unit: action: start_session
Jonathan Rajotte [Thu, 8 Apr 2021 13:14:22 +0000 (09:14 -0400)] 
tests: unit: action: start_session

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ie224592426d286b6abd4c7b627853bef71862bb8

3 years agoImplement firing policy for the start session action
Jonathan Rajotte [Thu, 8 Apr 2021 13:14:06 +0000 (09:14 -0400)] 
Implement firing policy for the start session action

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ica842dbd9980e8dbf73e38cbb1136115394c001c

3 years agotests: unit: action: rotate_session
Jonathan Rajotte [Thu, 8 Apr 2021 01:17:13 +0000 (21:17 -0400)] 
tests: unit: action: rotate_session

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ib412196dea5cb9617b4c56e8eb93698bcfaa40d7

3 years agoImplement firing policy for the rotate session action
Jonathan Rajotte [Thu, 8 Apr 2021 01:02:31 +0000 (21:02 -0400)] 
Implement firing policy for the rotate session action

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I0e2b50d6653cca7a33abc6ee68bed237950476a9

3 years agotests: unit: action: notify
Jonathan Rajotte [Fri, 9 Apr 2021 14:59:09 +0000 (10:59 -0400)] 
tests: unit: action: notify

Note: the added tests are mostly toward testing of the firing policy
parameter.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I2aec17175ac2fed4cf21a15f998b24dfd353e77b

3 years agoImplement firing policy for the notify action
Jonathan Rajotte [Fri, 9 Apr 2021 01:40:12 +0000 (21:40 -0400)] 
Implement firing policy for the notify action

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I40b25033868fd4ec9cb6764f3c47ef3e006bec3e

3 years agoMove test_action to its own file
Jonathan Rajotte [Fri, 9 Apr 2021 14:58:34 +0000 (10:58 -0400)] 
Move test_action to its own file

Further unit testing on action objects will be conducted in the
test_action.c file.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ib17aed9eb02592d5c13bc2875d4203b81ab04cf3

3 years agotests: unit: add firing policy tests
Jonathan Rajotte [Tue, 6 Apr 2021 15:09:09 +0000 (11:09 -0400)] 
tests: unit: add firing policy tests

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ib92bd40b2c93f4e90ec430f2ac70ef2a5ce8f891

3 years agoIntroduce firing policy object
Jonathan Rajotte [Fri, 2 Apr 2021 01:20:23 +0000 (21:20 -0400)] 
Introduce firing policy object

Change-Id: I436ce58d083cd70178a51d0d9ebfddc814f2cd15
Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agoFix: test: start-stop trigger: test execution is invalid
Jonathan Rajotte [Tue, 30 Mar 2021 01:38:54 +0000 (21:38 -0400)] 
Fix: test: start-stop trigger: test execution is invalid

Observed issue
==============

The test expects the notification client to be a background process but
`&` is not used.

The notification client is expected to sync with its launcher via the
sync file but the invocation of mktemp already creates the file.

Solution
=======

Use `&` and `mktemp -u`.

While there a little cleanup in notification_client for error code path
was done.

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Iba80413d6ca36989cd967895d3330860cb4ef614

3 years agoFix: sessiond: session destroy hang in per-uid when context cannot be added
Jonathan Rajotte [Mon, 15 Mar 2021 21:52:24 +0000 (17:52 -0400)] 
Fix: sessiond: session destroy hang in per-uid when context cannot be added

Observed issue
==============

The system_test CI jobs hang on the perf test suite during the destroy
command steps of the ust perf raw subtest.

Cause
=====

The system_test are running inside a kvm as root. It turns out that the
PMU (UNHALTED_REFERENCE_CYCLES) the test suite is trying to add is
unavailable on the qemu host.

ustctl_add_context return -1024 since it fails to add the context.

This leads us down the error path for the callstack leading to the
ustctl_add_context call.

1) `ust_app_channel_create` returns `ret` != 0;
2) `find_or_create_ust_app_channel` returns `ret != 0`;
3) `ust_app_synchronize` based on the `ret` value goes directly to the
    end of the function to an error path without passing on the
    `create_ust_app_metadata` function and clean-up structure related to
    the app.

Note that being in per-uid mode, data and metadata
channel/streams/buffer allocation is done on the fly for the first app
during `ust_app_synchronize` and its callee. For the current problematic
scenario, only the data channels have been allocated on the consumer for
the uid at that point. The metadata for that uid is not yet created.

Now that we know more of what is going on during an ""add context""
let's take a look at the actual hang.

The client never complete the destroy command since the consumerd
indicates that the trace chunk for the session is not closed. The trace
chunk still exists despite the fact that a close chunk command has been
issued. This is the case since its refcount never reaches zero and thus
the release does not complete.

In a normal execution without the use of contexts, the release of the
trace hunk (refcount == 0) occurs during the final rotation on destroy.

Upon further comparison between a working execution and a non-working
execution, in a non-working execution the `cmd_rotate_session` does not
issue the rotation for the data channels since the loop detects that no
metadata is present. Which, as we discussed earlier, can happen if we
fail to add the context to the app channel.

[1]
```
  cds_list_for_each_entry(reg, &usess->buffer_reg_uid_list, lnode) {
    struct buffer_reg_channel *reg_chan;
    struct consumer_socket *socket;

    if (!reg->registry->reg.ust->metadata_key) {
      /* Skip since no metadata is present */
      continue;
    }

    ....

    /* Rotate the data channels. */
    cds_lfht_for_each_entry(reg->registry->channels->ht, &iter.iter,
        reg_chan, node.node) {
      ret = consumer_rotate_channel(socket,
          reg_chan->consumer_key,
          usess->uid, usess->gid,
          usess->consumer,
          /* is_metadata_channel */ false);
      if (ret < 0) {
        cmd_ret = LTTNG_ERR_ROTATION_FAIL_CONSUMER;
        goto error;
      }
    }

    ....
  }
```

Solution
========

Move the metadata check after the data channel rotation since it is
possible to have data channels but no metadata channel, although it is a
corner case.

Note that per-pid mode and kernel are not affected by the current bug
since a complete teardown of all objects is done. This only affect
per-uid due to the "on the fly" allocation nature of it since we need to
share the channel/stream/buffers across apps.

Known drawbacks
=========

None.

References
==========
[1] https://github.com/lttng/lttng-tools/blob/3d1384a4add389c38f8554130e8dec2e2d06009d/src/bin/lttng-sessiond/ust-app.c#L7057

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ie8fd1167170706ec9abc5f31f3b33a7306e92cd9

3 years agoFix: backward relayd: path contains a leading "ust" folder
Jonathan Rajotte [Wed, 13 Jan 2021 15:09:06 +0000 (10:09 -0500)] 
Fix: backward relayd: path contains a leading "ust" folder

Observed issue
==============

test_output_path_relayd[lttng-tools-2.13-lttng-tools-2.10-uid] from
the lttng-ivc fails on path validation for the resulting trace.

Here lttng-sessiond is 2.13 and lttng-relayd is 2.10

Traces are generated and the following hierarchy is found in the
lttng-relayd trace folder.

 lttng_home
         └── lttng-traces
*            └── ust
                 └── joraj-alpa
                     ├── auto-20210113-165054
                     │   └── ust
                     │       └── uid

Note the extra "ust" (*) in the hierarchy.

The tests expects:
 lttng_home
         └── lttng-traces
                 └── joraj-alpa
                     ├── auto-20210113-165054
                     │   └── ust
                     │       └── uid

Cause
=====

Introduced by: 5da88b0f58d7f838068037ea449ddfb25d3e85ad [1]

relayd_add_stream now suffixes the domain_name to the pathname.
This is only necessary for cases where the corresponding
lttng-relayd version is greater than 2.10. In other cases, modification
of pathname is not necessary.

Solution
========

Perform domain suffixing only for relayd > 2.10.

Known drawbacks
=========

None.

References
==========

[1] https://github.com/lttng/lttng-tools/commit/5da88b0f58d7f838068037ea449ddfb25d3e85ad

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Iaca4667074aaf7f4f20b50ddb4778fbc02156b5a

3 years agoFix: lttng_destroy_session_no_wait: return 0 on success
Christophe Bedard [Fri, 28 Aug 2020 15:00:30 +0000 (11:00 -0400)] 
Fix: lttng_destroy_session_no_wait: return 0 on success

lttng_destroy_session_no_wait() is supposed to behave like
lttng_destroy_session():

> Return 0 on success else a negative LTTNg error code.

However, it returns LTTNG_OK on success. Make it return 0 instead.

Signed-off-by: Christophe Bedard <christophe.bedard@apex.ai>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: If51f6a2cc3ca77237f7cbac806c5206a807dadf5

3 years agoFix: notification: client with uid != trigger uid assert on gid check
Jonathan Rajotte [Wed, 3 Mar 2021 20:45:11 +0000 (15:45 -0500)] 
Fix: notification: client with uid != trigger uid assert on gid check

The gid was removed from the trigger object since gid is not an exposed
control measure anywhere in lttng except for the "tracing" group and
lttng-sessiond to client communication.

Also based on this:

 /*
 * As for privilieged users, they can register triggers against the objects of
 * other users. They can then subscribe to the notifications associated to their
 * triggers. Privilieged users _can't_ subscribe to the notifications of
 * triggers owned by other users; they must create their own triggers.
 * /

This is why the current check is not bypassed if the notification client
have a UID == 0 (root).

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I690ea355b0b1ab0bfe82b7db47275876afe622f2

3 years agoevent-rule: Normalize pattern for syscall and tracepoint
Jonathan Rajotte [Thu, 4 Mar 2021 21:10:18 +0000 (16:10 -0500)] 
event-rule: Normalize pattern for syscall and tracepoint

Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: If5bae2cd5df76db3f9bd626f95fcace114fcc420

3 years agoconfig: accept "0" and "1" as XML boolean values
Simon Marchi [Fri, 2 Apr 2021 19:52:46 +0000 (15:52 -0400)] 
config: accept "0" and "1" as XML boolean values

These are valid values when validating a "boolean" using a schema (XSD).
Somebody could therefore write a file with

    <started>0</started>

instead of

    <started>false</started>

The file would validate against session.xsd, but we would fail after
that:

    Warning: Invalid boolean value encoutered (0).
    Error: Invalid session configuration

Reference: https://www.w3.org/TR/xmlschema11-2/#boolean

Change-Id: Icea85b13233ec01aa09ef3b578906f4c0e482088
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agoconfig: fix typo in error message
Simon Marchi [Fri, 2 Apr 2021 19:52:27 +0000 (15:52 -0400)] 
config: fix typo in error message

Change-Id: Ie02ccd1e79aecb87fbb9c308ffba09353d50b8f3
Signed-off-by: Simon Marchi <simon.marchi@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
3 years agoCleanup: reuse `lttng_action_type_string()` to print action names
Francis Deslauriers [Fri, 19 Mar 2021 20:11:40 +0000 (16:11 -0400)] 
Cleanup: reuse `lttng_action_type_string()` to print action names

Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I86be1201466a11510ded9da34311650a8aadd1e2

3 years agoTests: array expressions without contant index are invalid
Mathieu Desnoyers [Mon, 25 May 2020 20:19:18 +0000 (16:19 -0400)] 
Tests: array expressions without contant index are invalid

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: I121387b63eb85851be59416c758b7ca12f823f45

3 years agoFix: validate that array expression contains constant
Mathieu Desnoyers [Mon, 25 May 2020 20:15:40 +0000 (16:15 -0400)] 
Fix: validate that array expression contains constant

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Change-Id: Ic7ea2192b3bf0f5c32f5db8fce622f73b182906f

This page took 0.050361 seconds and 4 git commands to generate.