Re: [Quickfix-developers] Expected response to second Logon reque st?

[Derek K. <DK...@li...>]

The FIX 4.2 spec says "if [Logon] authentication fails, the session acceptor
should shut down the connection," which is what QuickFIX does. The spec is
just not super clear on what an authentication failure is, though according
to the code -- Session::validLogonState() -- it seems QF won't accept a
second Logon without a preceeding Reset or Logout etc.
 
I'm not sure if our client wants us to ignore the second Logon or reply to
it. As they have explained it, their application (an order management
system) has a monitoring tool that watches all FIX sessions. If any session
goes down, it sends a Logon to all sessions. They say all the other FIX
engines they connect to have no problem with this second Logon (again, I'm
not sure if this means they ignore it or reply to it).
 
It seems very odd to me to send a Logon to all sessions when only one goes
down, so I'm still working on getting more clarification from my client.
However, in the meantime, I was hoping to better understand if how QuickFIX
handled the situation was defined by the FIX protocol. From what I have read
and what you said, it seems like a gray area.
 
Thanks.
 
Derek

  _____  

From: Oren Miller [mailto:or...@qu...] 
Sent: Friday, June 23, 2006 2:13 PM
To: Derek Kusiak
Cc: 'qui...@li...'
Subject: Re: [Quickfix-developers] Expected response to second Logon
request?


I can't find a really clear cut answer on this.  It does say that every
connection starts with a logon authentication, followed by message
processing, followed by logout.  Other than for resetting sequence numbers,
there is no mention anywhere of ever sending a logon in between.  Any idea
why they are doing this?  What is their intention?  Do they want you to send
another logon in response?  I'm wondering why exactly they want to do this. 

--oren

On Jun 23, 2006, at 12:26 PM, Derek Kusiak wrote:


QuickFIX Documentation:
http://www.quickfixengine.org/quickfix/doc/html/index.html
<http://www.quickfixengine.org/quickfix/doc/html/index.html> 
QuickFIX Support: http://www.quickfixengine.org/services.html
<http://www.quickfixengine.org/services.html> 


I have a situation with QuickFIX 1.10.2 (FIX 4.2) where the initiator
sending two Logon (35=A) requests. The first (expected) Logon arrives and
the session begins normally. Then, later in the day, for reasons I have no
controller over, the acceptor sends a SECOND unexpected Logon request
without a preceding Logout. QuickFIX doesn't like this second Logon. It
replies by saying "Logon state is not valid for message" followed by
"Dropped Connection." It generates a Logout event and ends the session.

My client (the initiator) is saying other FIX engines she connects do not
behave this way when they get a second Logon -- they just output a warning
or ignore the second Logon altogether.

Can anyone tell me where in the FIX spec this behavior is defined? I'm
trying to figure out if this is a QuickFIX bug or the correct FIX response
to a second Logon.

Thanks. 

Derek 

Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job
easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk
<http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642>
&kid=120709&bid=263057&dat=121642
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
<mailto:Qui...@li...> 
https://lists.sourceforge.net/lists/listinfo/quickfix-developers
<https://lists.sourceforge.net/lists/listinfo/quickfix-developers>