quickfix-developers

[Quickfix-developers] Callink callstack methods from application code?

[Andrei G. <an...@gm...>]

Hi all,

I am trying to "trace" method calls in order to understand what
happens when the engine transmits and receives messages (i will
probably have to add built-in encryption support for my applications
in the near future). I noticed that there is a CallStack object that
is used to keep a trace of method calls.
I was wondering if this object can be used in the application code so
that I can Inspect the stack trace(CallStack::output(), maybe?), for
example, when control reaches "Application" methods(fromApp, toApp,
....).

I am currently using version 1.10.2  via the .NET API, which does not
seem to export the output method.

Any help will be much appreciated,

Regards,

Andrei

[Quickfix-developers] Duplicate Logout messages

[Dale W. <wil...@oc...>]

QuickFIX 1.10.2:  C++; Windows XP

I have a question about the proper sequence of messages during logout 
(and whether QuckFIX is doing "the right thing.")

This shows up during a unit test for our application.
The application is acting as an initiator
 The test is trivial:  logon, wait for the onLogon callback; then 
logout, wait for the onLogout callback

A condensed version of the log produced by this test is:

15:00:47.893000 (FIX.4.2:CLIENT->EXSIM) session->logon()
15:00:47.893000 (FIX.4.2:CLIENT->EXSIM) Application::toAdmin 
8=FIX.4.2|9=66|35=A|34=1|49=CLIENT|52=20050915-20:00:47.000|56=EXSIM|98=0|108=30|10=056|
15:00:47.908000 (FIX.4.2:CLIENT->EXSIM) Application::fromAdmin 
8=FIX.4.2|9=66|35=A|34=1|49=EXSIM|52=20050915-20:00:47.000|56=CLIENT|98=0|108=30|10=056|
15:00:47.924000 (FIX.4.2:CLIENT->EXSIM) Application::onLogon
15:00:47.924000 (FIX.4.2:CLIENT->EXSIM) session->logout()
15:00:48.033000 (FIX.4.2:CLIENT->EXSIM) Application::toAdmin 
8=FIX.4.2|9=54|35=5|34=2|49=CLIENT|52=20050915-20:00:48.000|56=EXSIM|10=018|
15:00:48.033000 (FIX.4.2:CLIENT->EXSIM) Application::onLogout
15:00:48.033000 (FIX.4.2:CLIENT->EXSIM) Application::fromAdmin 
8=FIX.4.2|9=54|35=5|34=2|49=EXSIM|52=20050915-20:00:48.000|56=CLIENT|10=018|
15:00:48.049000 (FIX.4.2:CLIENT->EXSIM) Application::toAdmin 
8=FIX.4.2|9=54|35=5|34=3|49=CLIENT|52=20050915-20:00:48.000|56=EXSIM|10=019|

Two things I see:
#1: the onLogout callback happens before the Logout reply from the 
counterparty, and
#2: an "extra" Logout message is being sent: presumably in response to 
the Logout from the counterparty.

Is this correct behavior?

Dale

PS: #1 led to an interesting race condition when we started the next 
test which managed to send a  login *before* the logout reply came from 
the counterparty, but that's another issue.

Re: [Quickfix-developers] Duplicate Logout messages

[Dale W. <wil...@oc...>]

I have some more info on the problem below-- enough to convince myself 
that it is an error in QuickFIX, and not a misunderstanding on my part.

The second logout message is sent because the SessionState::m_sentLogout 
is being set to false too soon.  When the Logout arrives from the other 
end, this bool is false so, QuickFIX assumes that the other end is 
initiating the logout process.

sentLogout(false) is being called from Session::disconnect which also 
calls Application::onLogout.  Therefore it appears the two symptoms I 
observed below are related -- both seem to happen because 
Session::disconnect is being called too soon.

I'll go ahead and put this information in Bugzilla.

Dale

Dale Wilson wrote:

> QuickFIX Documentation: 
> http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX Support: http://www.quickfixengine.org/services.html
>
> QuickFIX 1.10.2:  C++; Windows XP
>
> I have a question about the proper sequence of messages during logout 
> (and whether QuckFIX is doing "the right thing.")
>
> This shows up during a unit test for our application.
> The application is acting as an initiator
> The test is trivial:  logon, wait for the onLogon callback; then 
> logout, wait for the onLogout callback
>
> A condensed version of the log produced by this test is:
>
> 15:00:47.893000 (FIX.4.2:CLIENT->EXSIM) session->logon()
> 15:00:47.893000 (FIX.4.2:CLIENT->EXSIM) Application::toAdmin 
> 8=FIX.4.2|9=66|35=A|34=1|49=CLIENT|52=20050915-20:00:47.000|56=EXSIM|98=0|108=30|10=056| 
>
> 15:00:47.908000 (FIX.4.2:CLIENT->EXSIM) Application::fromAdmin 
> 8=FIX.4.2|9=66|35=A|34=1|49=EXSIM|52=20050915-20:00:47.000|56=CLIENT|98=0|108=30|10=056| 
>
> 15:00:47.924000 (FIX.4.2:CLIENT->EXSIM) Application::onLogon
> 15:00:47.924000 (FIX.4.2:CLIENT->EXSIM) session->logout()
> 15:00:48.033000 (FIX.4.2:CLIENT->EXSIM) Application::toAdmin 
> 8=FIX.4.2|9=54|35=5|34=2|49=CLIENT|52=20050915-20:00:48.000|56=EXSIM|10=018| 
>
> 15:00:48.033000 (FIX.4.2:CLIENT->EXSIM) Application::onLogout
> 15:00:48.033000 (FIX.4.2:CLIENT->EXSIM) Application::fromAdmin 
> 8=FIX.4.2|9=54|35=5|34=2|49=EXSIM|52=20050915-20:00:48.000|56=CLIENT|10=018| 
>
> 15:00:48.049000 (FIX.4.2:CLIENT->EXSIM) Application::toAdmin 
> 8=FIX.4.2|9=54|35=5|34=3|49=CLIENT|52=20050915-20:00:48.000|56=EXSIM|10=019| 
>
>
> Two things I see:
> #1: the onLogout callback happens before the Logout reply from the 
> counterparty, and
> #2: an "extra" Logout message is being sent: presumably in response to 
> the Logout from the counterparty.
>
> Is this correct behavior?
>
> Dale
>
> PS: #1 led to an interesting race condition when we started the next 
> test which managed to send a  login *before* the logout reply came 
> from the counterparty, but that's another issue.
>
>
>
>
> -------------------------------------------------------
> SF.Net email is sponsored by:
> Tame your development challenges with Apache's Geronimo App Server. 
> Download
> it for free - -and be entered to win a 42" plasma tv or your very own
> Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers



-- 
-----------------------------------------------------
  Dale Wilson, Senior Software Engineer      
  Object Computing, Inc. (OCI)
  http://www.ociweb.com/  http://www.theaceorb.com/
----------------------------------------------------