Bug #452


Starting a trace with an unready remote lttng-relayd leaves the trace in an odd state

Added by Daniel U. Thibault over 11 years ago. Updated almost 11 years ago.

Won't fix
Target version:
Start date:
Due date:
% Done:


Estimated time:


Here I'm running an lttng session with the intent of streaming the trace to another machine. But the other machine's lttng-relayd has not been started yet (or the IP is wrong, etc.).

$ sudo -H lttng create rsession -U net://
Session rsession created.
Traces will be written in net://
$ sudo -H lttng enable-event -k sched_switch
kernel event sched_switch created in channel channel0
$ sudo -H lttng start
Error: Unable to connect to lttng-relayd
$ sudo -H lttng start
Warning: Tracing already started for session rsession
$ sudo -H lttng stop
Error: Stoping kernel trace failed
$ sudo -H lttng destroy -a
Session rsession destroyed

The main bug here is we have an "already started" session which "cannot be stopped".

Also, the "Error: Unable to connect to lttng-relayd" message could be accompanied by a further message stating that the session was started anyway (or, for some other errors, that the start was aborted); as it currently stands the user is unsure whether his 'start' command succeeded or not.

"Error: Stoping kernel trace failed" includes an obvious typo. A separate bug will be filed shortly.
Actions #1

Updated by David Goulet about 11 years ago

  • Status changed from New to Confirmed
  • Assignee set to David Goulet
  • Target version set to 2.2
Actions #2

Updated by David Goulet about 11 years ago

  • Target version changed from 2.2 to 2.1 stable

This is actually a 2.1 bug and I can confirm it.

This has been resolved in 2.2 where at the enable-event, the Unable to connect is be printed.

Actions #3

Updated by David Goulet almost 11 years ago

  • Status changed from Confirmed to Won't fix

2.1-stable is not supported anymore unless it's a security fix.


Also available in: Atom PDF