Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update documentation for using TotalView with Flux #265

Merged
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
27 changes: 20 additions & 7 deletions jobs/debugging.rst
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
.. _debugging:

==============
Debugging Jobs
Debugging Jobs
==============

Flux supports parallel debuggers such as Rogue Wave Software (RWS)'s
Expand All @@ -11,8 +11,8 @@ Flux supports parallel debuggers such as Rogue Wave Software (RWS)'s
Parallel Debugging using TotalView
----------------------------------

Debugging your MPI job from the beginning of
parallel program execution:
Debugging your MPI job from the beginning
of parallel program execution:

.. code-block:: console

Expand All @@ -26,10 +26,13 @@ Attaching to an already running job:
$ PID=$!
$ totalview -pid ${PID} /proc/${PID}/exe

You can also just type ``totalview`` without ``-pid`` option and
then use its ``Start a Debugging Session``
GUI option to select the process of this ``flux-job`` command:
``A running program (attach)``.
You can also just type ``totalview`` without ``-pid`` option.

For the TotalView Classic user interface, select the process of this
``flux-job`` command with
:menuselection:`Start a Debugging Session --> A running program (attach)`.

For new UI: use :menuselection:`Attach To Process`.

.. note::
You use TotalView with the newly invoked ``flux job attach``
Expand All @@ -47,6 +50,16 @@ can be handy when you debug a large-scale job. Please refer to

.. _TotalView user guide: https://docs.roguewave.com/en/totalview/current/html/

Exiting TotalView without completing a full run of your code may not clean
up the Flux job. In that case you will need to cancel the flux job manually.

.. code-block:: console

$ flux cancel <JOBID>

.. note::
Use ``flux jobs`` to find <JOBID>

----------------------------------------------
Better Handling of Flux's Internal Exec Events
----------------------------------------------
Expand Down
Loading