Re: [Quickfix-developers] Heartbeat

[Oren M. <or...@qu...>]

Ahh you know, there is already a fix for this.  OptionsXpress reported=20=

this problem a couple months ago and I helped resolve it for them.  I=20
thought I had checked it in but looking at the release notes and CVS, I=20=

certainly did not.  The easiest way to fix this problem is to add a=20
call to next() on the last line of the Session::next( const Message&=20
message ) method.  This gives the session a chance to do administrative=20=

processing in between messages.  I'll check in the fix and make sure it=20=

goes out in the next release.

--oren

On Jul 23, 2004, at 8:54 AM, Yihu Fang wrote:

> QuickFIX Documentation:=20
> http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX FAQ: http://www.quickfixengine.org/quickfix/doc/html/FAQ.html
> QuickFIX Support: http://www.quickfixengine.org/services.html
>
> Joerg,
>
> I don't have the logs with me anymore and will need some effort to=20
> recreate it. Basically, it happens when we evaluate the QF engine by=20=

> sending large amount of messages without throttle from an initiator to=20=

> an acceptor. The initiator will miss the heart beat after a certain=20
> period of time, and then it sends the test request to force a heart=20
> beat. After that miss it declares the session down by log off.
>
> -Yihu
>
> -----Original Message-----
> From: Joerg Thoennes [mailto:Joe...@ma...]
> Sent: Friday, July 23, 2004 4:54 AM
> To: Yihu Fang
> Cc: Patrick Flannery; qui...@li...
> Subject: Re: [Quickfix-developers] Heartbeat
>
> Hi Yihu, hi Patrick,
>
> Yihu wrote:
>
>> We have seen similar missed heartbeat behavior when using=20
>> SocketAcceptor > under heavy sustained message rate. It causes=20
>> reconnect and resend and > other undesired behavior.
>>
>> After we change to ThreadedSocketAcceptor and it has no problem ever=20=

>> since.
>
> Yihu, can you send us log files showing the same scenario using
> SocketAcceptor and ThreadedSocketAcceptor?
>
> Patrick wrote:
>
>> Is it possible that when many messages are being sent or received a
>> heartbeat gets drowned out causing a logout message to be generated?
>>
>> I have both acceptor and initiator ResetOnDiscconect and =
ResetOnLogout
>> set to Y, but I notice when the client starts up there are series of
>> onLogoff and onLogon messages and Reset Sequence number events before=20=

>> a > connection is made?  What could cause this?
>>
>> When the client sends the logout message an onLogout event is=20
>> generated > then an onLogon event the client can send messages as if=20=

>> they never
>> logged off, why would that happen?
>
> Could you please send log file (from both sides of the connecting FIX
> engines if possible) to have a more detailed look at this?
>
> Thanks, J=F6rg
>
> --=20
> Joerg Thoennes
>                                 http://macd.com
> Tel.: +49 (0)241 44597-24      Macdonald Associates GmbH
> Fax : +49 (0)241 44597-10      Lothringer Str. 52, D-52070 Aachen
>
>
>
> -----------------------------------------------------------------
>         Visit our Internet site at http://www.reuters.com
>
> Get closer to the financial markets with Reuters Messaging - for more
> information and to register, visit http://www.reuters.com/messaging
>
> Any views expressed in this message are those of  the  individual
> sender,  except  where  the sender specifically states them to be
> the views of Reuters Ltd.
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by BEA Weblogic Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1 today.
> http://ads.osdn.com/?ad_idG21&alloc_id=10040&op=3Dclick
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>