quickfix-developers — QuickFIX developers forum

[Quickfix-developers] symbol lookup error: ./quickfix-1.13.3/src/C++/.libs/libquickfix.so.14: undefined symbol: _Z10

[Daniel M. <dmo...@ya...>]

My code works without FIX::FileLogFactory but I do not get any messages when
a fix connection is not logged on due to rejection by quickfix (Initiator or
Acceptor). Therefore, I added 

FileLogPath=./fix_log 

And nothing happen. Then I modified the code to 

FIX::ThreadedSocketInitiator 	*initiator = 0;

FIX::FileLogFactory *file_log_factory = 0;
file_log_factory = new FIX::FileLogFactory(*settings);		
initiator = new FIX::ThreadedSocketInitiator ( *this, 
									*storeFactory, 
									*settings, 
									*file_log_factory);	

And I get 
symbol lookup error:./quickfix-1.13.3/src/C++/.libs/libquickfix.so.14:
undefined symbol: _Z10

Same thing with ThreadedSocketAcceptor ?

I am using /quickfix-1.13.3/src/C++. 

What seems to be the issue --
Is there a sample code that works for this?

Thanks




--
View this message in context: http://quickfix.13857.n7.nabble.com/symbol-lookup-error-quickfix-1-13-3-src-C-libs-libquickfix-so-14-undefined-symbol-Z10-tp6483.html
Sent from the QuickFIX - Dev mailing list archive at Nabble.com.

Re: [Quickfix-developers] How to suppress 35=X messages in log file

[Mike G. <mg...@co...>]

On Thu, May 30, 2013 at 2:52 AM, Gunnar Harms <gh...@hi...> wrote:

> Deriving from the FileLogFactory and from FileLog to create my own derived
> FileLog is probably easy but in the onOutgoing method of the Log object the
> message type is not available
>

I've run into this before myself and I agree that it might be nice to have
access to something other than the raw string in the Log interface.
 However, it is also desirable to log the message immediately (and
therefore raw) so that we have a record of it in case an error occurs when
we try to process it.  Maybe the answer is to augment the Log interface to
have onIncoming/onOutgoing and additionally onIncomingRaw/onOutgoingRaw
callbacks.  Then you could implement whatever makes sense for your app.

In the meantime, however: see Message::identifyType() in Message.h, which
takes a raw string and returns the MsgType.

-- 
Mike Gatny
Connamara Systems, LLC

Re: [Quickfix-developers] How to suppress 35=X messages in log file

[Gunnar H. <gh...@hi...>]

When deriving from FIX::FileLog and searching the message string for 35=X is the only way to supress these messages from getting logged w/o changing QF source then I will do that.
I was just thinking that I was overlooking something simple because it seems logical to disable logging of 35=X messages when using FIX::FileLog.
BTW. I like the QF library a lot!

Thank you for your help!

Gunnar

On May 30, 2013, at 3:57 PM, Grant Birchmeier wrote:

> I do agree: in retrospect, it would be nice if the engine would pass Message objects to the logger instead of strings.  If you felt like hacking the engine, that change probably wouldn't be very hard.
> 
> 
> On Thu, May 30, 2013 at 8:17 AM, Gunnar Harms <gh...@hi...> wrote:
> Thanks Grant!
> 
> Actually the application is finished so it is probably not a premature-mindset.
> 
> The thing is that when in Sesison m_sttate->onOutgoing is called the MsgType is already in a variable. but in the flow that follows the MsgType is lost and only message string reaches the Log object's onOutgoing() method.
> 
> It is just weird to look in message string with std::string compare method for the MsgType when it was known a couple steps before.
> it would be better if MsgType is an actual argument of Log object's onOutgoing(); then i would not hesitate to derive from FileLogFactory and FileLog.
> But still it would be better to already in session send methods determine not to put the message to the logger when 35=X
> 
> 
> Regards,
> Gunnar
> 
> 
> 
> On May 30, 2013, at 2:39 PM, Grant Birchmeier wrote:
> 
>> Are you sure it's a waste of time?  Sounds like a premature-optimization mindset.
>> 
>> Attempting it and profiling its performance sounds like an easier option than hacking the engine, and I don't think it would take that many man-hours.
>> 
>> 
>> On Thu, May 30, 2013 at 2:52 AM, Gunnar Harms <gh...@hi...> wrote:
>> QuickFIX Documentation: http://www.quickfixengine.org/quickfix/doc/html/index.html
>> QuickFIX Support: http://www.quickfixengine.org/services.html
>> 
>> 
>> Thanks Mike!
>> 
>> I create an acceptor with a FIX::FileLogFactory because I do want the logging to file of all messages except the 35=X. I have stream and trade sessions. For the trade session i want all logging and for stream session all but the 35=X. So I think I use the appropriate constructor of the acceptor.
>> 
>> Deriving from the FileLogFactory and from FileLog to create my own derived FileLog is probably easy but in the onOutgoing method of the Log object the message type is not available, only the message string and I don't want to search in the message string for 35=X because that is a waste of time and totally not nice.
>> 
>> Regards,
>> Gunnar
>> 
>> 
>> 
>> 
>> Gunnar Harms
>> 
>> On May 29, 2013, at 2:15 PM, Mike Gatny wrote:
>> 
>>> It is typical to set PersistMessages=Y, ResetOnLogon=Y, and use the initator/acceptor constructor that takes no LogFactory argument.  If you want to do more specific filtering, the LogFactory and Log interfaces exist for this reason and are simple to derive from.
>>> 
>>> On May 29, 2013 6:41 AM, "Gunnar Harms" <gh...@hi...> wrote:
>>> QuickFIX Documentation: http://www.quickfixengine.org/quickfix/doc/html/index.html
>>> QuickFIX Support: http://www.quickfixengine.org/services.html
>>> 
>>> Hi All,
>>> 
>>> I want to suppress the 35=X messages in the FIX.4.2-<X>-<Y>.messages.current.log files and log all other messages (administrative, marketdatarequestrejects, etc.).
>>> 
>>> It would be easy if i could derive my own Session class from FIX::Session and override the sendRaw/send methods to not make the m_state.onOutgoing( string ) call for these message types but this ssems not possible (because in acceptor class the sessions are created from calls to a locally defined SessionFactory variable).
>>> 
>>> 
>>> What i could try is  deriving from ScreenLogFactory (say MyScreenLogFactory), override in that class some create() method(s) to create not a ScreenLog but a derivation (say MyScreenLog) of it and in that derived class override onOutgoing(). But then I must search in the message string for subsstrings like 35=X which is a waste of time since in FIX::Session send methods is available the MsgType.
>>> 
>>> Also some periodically running script to truncate the log files is far from optimal for obvious reasons.
>>> 
>>> 
>>> I am perhaps missing something simple because I cannot imagine that writing each 35=X to log file is acceptable for other users of QF (for us it is really a show stopper, log files get toooo big).
>>> 
>>> 
>>> Thanks for reading and for any help,
>>> Gunnar
>>> 
>>> 
>>> 
>>> 
>>> 
>>> ------------------------------------------------------------------------------
>>> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
>>> Get 100% visibility into your production application - at no cost.
>>> Code-level diagnostics for performance bottlenecks with <2% overhead
>>> Download for free and get started troubleshooting in minutes.
>>> http://p.sf.net/sfu/appdyn_d2d_ap1
>>> _______________________________________________
>>> Quickfix-developers mailing list
>>> Qui...@li...
>>> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>> 
>> 
>> ------------------------------------------------------------------------------
>> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
>> Get 100% visibility into your production application - at no cost.
>> Code-level diagnostics for performance bottlenecks with <2% overhead
>> Download for free and get started troubleshooting in minutes.
>> http://p.sf.net/sfu/appdyn_d2d_ap1
>> _______________________________________________
>> Quickfix-developers mailing list
>> Qui...@li...
>> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>> 
>> 
>> 
>> -- 
>> Grant Birchmeier
>> Connamara Systems, LLC
>> Made-To-Measure Trading Solutions.
>> Exactly what you need. No more. No less.
>> http://connamara.com
> 
> 
> 
> 
> -- 
> Grant Birchmeier
> Connamara Systems, LLC
> Made-To-Measure Trading Solutions.
> Exactly what you need. No more. No less.
> http://connamara.com

Re: [Quickfix-developers] How to suppress 35=X messages in log file

[Grant B. <gbi...@co...>]

I do agree: in retrospect, it would be nice if the engine would pass
Message objects to the logger instead of strings.  If you felt like hacking
the engine, that change probably wouldn't be very hard.


On Thu, May 30, 2013 at 8:17 AM, Gunnar Harms <gh...@hi...> wrote:

> Thanks Grant!
>
> Actually the application is finished so it is probably not a
> premature-mindset.
>
> The thing is that when in Sesison m_sttate->onOutgoing is called the
> MsgType is already in a variable. but in the flow that follows the MsgType
> is lost and only message string reaches the Log object's onOutgoing()
> method.
>
> It is just weird to look in message string with std::string compare method
> for the MsgType when it was known a couple steps before.
> it would be better if MsgType is an actual argument of Log object's
> onOutgoing(); then i would not hesitate to derive from FileLogFactory and
> FileLog.
> But still it would be better to already in session send methods determine
> not to put the message to the logger when 35=X
>
>
> Regards,
> Gunnar
>
>
>
> On May 30, 2013, at 2:39 PM, Grant Birchmeier wrote:
>
> Are you sure it's a waste of time?  Sounds like a premature-optimization
> mindset.
>
> Attempting it and profiling its performance sounds like an easier option
> than hacking the engine, and I don't think it would take that many
> man-hours.
>
>
> On Thu, May 30, 2013 at 2:52 AM, Gunnar Harms <gh...@hi...> wrote:
>
>> QuickFIX Documentation:
>> http://www.quickfixengine.org/quickfix/doc/html/index.html
>> QuickFIX Support: http://www.quickfixengine.org/services.html
>>
>>
>> Thanks Mike!
>>
>> I create an acceptor with a FIX::FileLogFactory because I do want the
>> logging to file of all messages except the 35=X. I have stream and trade
>> sessions. For the trade session i want all logging and for stream session
>> all but the 35=X. So I think I use the appropriate constructor of the
>> acceptor.
>>
>> Deriving from the FileLogFactory and from FileLog to create my own
>> derived FileLog is probably easy but in the onOutgoing method of the Log
>> object the message type is not available, only the message string and I
>> don't want to search in the message string for 35=X because that is a waste
>> of time and totally not nice.
>>
>> Regards,
>> Gunnar
>>
>>
>>
>>
>>  *Gunnar Harms*
>>
>> On May 29, 2013, at 2:15 PM, Mike Gatny wrote:
>>
>> It is typical to set PersistMessages=Y, ResetOnLogon=Y, and use the
>> initator/acceptor constructor that takes no LogFactory argument.  If you
>> want to do more specific filtering, the LogFactory and Log interfaces exist
>> for this reason and are simple to derive from.
>> On May 29, 2013 6:41 AM, "Gunnar Harms" <gh...@hi...> wrote:
>>
>>> QuickFIX Documentation:
>>> http://www.quickfixengine.org/quickfix/doc/html/index.html
>>> QuickFIX Support: http://www.quickfixengine.org/services.html
>>>
>>> Hi All,
>>>
>>> I want to suppress the 35=X messages in the
>>> FIX.4.2-<X>-<Y>.messages.current.log files and log all other messages
>>> (administrative, marketdatarequestrejects, etc.).
>>>
>>> It would be easy if i could derive my own Session class from
>>> FIX::Session and override the sendRaw/send methods to not make the
>>> m_state.onOutgoing( string ) call for these message types but this ssems
>>> not possible (because in acceptor class the sessions are created from calls
>>> to a locally defined SessionFactory variable).
>>>
>>>
>>> What i could try is  deriving from ScreenLogFactory (say
>>> MyScreenLogFactory), override in that class some create() method(s) to
>>> create not a ScreenLog but a derivation (say MyScreenLog) of it and in that
>>> derived class override onOutgoing(). But then I must search in the message
>>> string for subsstrings like 35=X which is a waste of time since in
>>> FIX::Session send methods is available the MsgType.
>>>
>>> Also some periodically running script to truncate the log files is far
>>> from optimal for obvious reasons.
>>>
>>>
>>> I am perhaps missing something simple because I cannot imagine that
>>> writing each 35=X to log file is acceptable for other users of QF (for us
>>> it is really a show stopper, log files get toooo big).
>>>
>>>
>>> Thanks for reading and for any help,
>>> Gunnar
>>>
>>>
>>>
>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
>>> Get 100% visibility into your production application - at no cost.
>>> Code-level diagnostics for performance bottlenecks with <2% overhead
>>> Download for free and get started troubleshooting in minutes.
>>> http://p.sf.net/sfu/appdyn_d2d_ap1
>>> _______________________________________________
>>> Quickfix-developers mailing list
>>> Qui...@li...
>>> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>>>
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
>> Get 100% visibility into your production application - at no cost.
>> Code-level diagnostics for performance bottlenecks with <2% overhead
>> Download for free and get started troubleshooting in minutes.
>> http://p.sf.net/sfu/appdyn_d2d_ap1
>> _______________________________________________
>> Quickfix-developers mailing list
>> Qui...@li...
>> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>>
>
>
>
> --
> Grant Birchmeier
> *Connamara Systems, LLC*
> *Made-To-Measure Trading Solutions.*
> Exactly what you need. No more. No less.*
> *
> http://connamara.com
>
>
>


-- 
Grant Birchmeier
*Connamara Systems, LLC*
*Made-To-Measure Trading Solutions.*
Exactly what you need. No more. No less.*
*
http://connamara.com

Re: [Quickfix-developers] How to suppress 35=X messages in log file

[Gunnar H. <gh...@hi...>]

Thanks Grant!

Actually the application is finished so it is probably not a premature-mindset.

The thing is that when in Sesison m_sttate->onOutgoing is called the MsgType is already in a variable. but in the flow that follows the MsgType is lost and only message string reaches the Log object's onOutgoing() method.

It is just weird to look in message string with std::string compare method for the MsgType when it was known a couple steps before.
it would be better if MsgType is an actual argument of Log object's onOutgoing(); then i would not hesitate to derive from FileLogFactory and FileLog.
But still it would be better to already in session send methods determine not to put the message to the logger when 35=X


Regards,
Gunnar



On May 30, 2013, at 2:39 PM, Grant Birchmeier wrote:

> Are you sure it's a waste of time?  Sounds like a premature-optimization mindset.
> 
> Attempting it and profiling its performance sounds like an easier option than hacking the engine, and I don't think it would take that many man-hours.
> 
> 
> On Thu, May 30, 2013 at 2:52 AM, Gunnar Harms <gh...@hi...> wrote:
> QuickFIX Documentation: http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX Support: http://www.quickfixengine.org/services.html
> 
> 
> Thanks Mike!
> 
> I create an acceptor with a FIX::FileLogFactory because I do want the logging to file of all messages except the 35=X. I have stream and trade sessions. For the trade session i want all logging and for stream session all but the 35=X. So I think I use the appropriate constructor of the acceptor.
> 
> Deriving from the FileLogFactory and from FileLog to create my own derived FileLog is probably easy but in the onOutgoing method of the Log object the message type is not available, only the message string and I don't want to search in the message string for 35=X because that is a waste of time and totally not nice.
> 
> Regards,
> Gunnar
> 
> 
> 
> 
> Gunnar Harms
> 
> On May 29, 2013, at 2:15 PM, Mike Gatny wrote:
> 
>> It is typical to set PersistMessages=Y, ResetOnLogon=Y, and use the initator/acceptor constructor that takes no LogFactory argument.  If you want to do more specific filtering, the LogFactory and Log interfaces exist for this reason and are simple to derive from.
>> 
>> On May 29, 2013 6:41 AM, "Gunnar Harms" <gh...@hi...> wrote:
>> QuickFIX Documentation: http://www.quickfixengine.org/quickfix/doc/html/index.html
>> QuickFIX Support: http://www.quickfixengine.org/services.html
>> 
>> Hi All,
>> 
>> I want to suppress the 35=X messages in the FIX.4.2-<X>-<Y>.messages.current.log files and log all other messages (administrative, marketdatarequestrejects, etc.).
>> 
>> It would be easy if i could derive my own Session class from FIX::Session and override the sendRaw/send methods to not make the m_state.onOutgoing( string ) call for these message types but this ssems not possible (because in acceptor class the sessions are created from calls to a locally defined SessionFactory variable).
>> 
>> 
>> What i could try is  deriving from ScreenLogFactory (say MyScreenLogFactory), override in that class some create() method(s) to create not a ScreenLog but a derivation (say MyScreenLog) of it and in that derived class override onOutgoing(). But then I must search in the message string for subsstrings like 35=X which is a waste of time since in FIX::Session send methods is available the MsgType.
>> 
>> Also some periodically running script to truncate the log files is far from optimal for obvious reasons.
>> 
>> 
>> I am perhaps missing something simple because I cannot imagine that writing each 35=X to log file is acceptable for other users of QF (for us it is really a show stopper, log files get toooo big).
>> 
>> 
>> Thanks for reading and for any help,
>> Gunnar
>> 
>> 
>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
>> Get 100% visibility into your production application - at no cost.
>> Code-level diagnostics for performance bottlenecks with <2% overhead
>> Download for free and get started troubleshooting in minutes.
>> http://p.sf.net/sfu/appdyn_d2d_ap1
>> _______________________________________________
>> Quickfix-developers mailing list
>> Qui...@li...
>> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
> 
> 
> ------------------------------------------------------------------------------
> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
> Get 100% visibility into your production application - at no cost.
> Code-level diagnostics for performance bottlenecks with <2% overhead
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap1
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
> 
> 
> 
> -- 
> Grant Birchmeier
> Connamara Systems, LLC
> Made-To-Measure Trading Solutions.
> Exactly what you need. No more. No less.
> http://connamara.com

Re: [Quickfix-developers] How to suppress 35=X messages in log file

[Grant B. <gbi...@co...>]

Are you sure it's a waste of time?  Sounds like a premature-optimization
mindset.

Attempting it and profiling its performance sounds like an easier option
than hacking the engine, and I don't think it would take that many
man-hours.


On Thu, May 30, 2013 at 2:52 AM, Gunnar Harms <gh...@hi...> wrote:

> QuickFIX Documentation:
> http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX Support: http://www.quickfixengine.org/services.html
>
>
> Thanks Mike!
>
> I create an acceptor with a FIX::FileLogFactory because I do want the
> logging to file of all messages except the 35=X. I have stream and trade
> sessions. For the trade session i want all logging and for stream session
> all but the 35=X. So I think I use the appropriate constructor of the
> acceptor.
>
> Deriving from the FileLogFactory and from FileLog to create my own derived
> FileLog is probably easy but in the onOutgoing method of the Log object the
> message type is not available, only the message string and I don't want to
> search in the message string for 35=X because that is a waste of time and
> totally not nice.
>
> Regards,
> Gunnar
>
>
>
>
>  *Gunnar Harms*
>
> On May 29, 2013, at 2:15 PM, Mike Gatny wrote:
>
> It is typical to set PersistMessages=Y, ResetOnLogon=Y, and use the
> initator/acceptor constructor that takes no LogFactory argument.  If you
> want to do more specific filtering, the LogFactory and Log interfaces exist
> for this reason and are simple to derive from.
> On May 29, 2013 6:41 AM, "Gunnar Harms" <gh...@hi...> wrote:
>
>> QuickFIX Documentation:
>> http://www.quickfixengine.org/quickfix/doc/html/index.html
>> QuickFIX Support: http://www.quickfixengine.org/services.html
>>
>> Hi All,
>>
>> I want to suppress the 35=X messages in the
>> FIX.4.2-<X>-<Y>.messages.current.log files and log all other messages
>> (administrative, marketdatarequestrejects, etc.).
>>
>> It would be easy if i could derive my own Session class from FIX::Session
>> and override the sendRaw/send methods to not make the m_state.onOutgoing(
>> string ) call for these message types but this ssems not possible (because
>> in acceptor class the sessions are created from calls to a locally defined
>> SessionFactory variable).
>>
>>
>> What i could try is  deriving from ScreenLogFactory (say
>> MyScreenLogFactory), override in that class some create() method(s) to
>> create not a ScreenLog but a derivation (say MyScreenLog) of it and in that
>> derived class override onOutgoing(). But then I must search in the message
>> string for subsstrings like 35=X which is a waste of time since in
>> FIX::Session send methods is available the MsgType.
>>
>> Also some periodically running script to truncate the log files is far
>> from optimal for obvious reasons.
>>
>>
>> I am perhaps missing something simple because I cannot imagine that
>> writing each 35=X to log file is acceptable for other users of QF (for us
>> it is really a show stopper, log files get toooo big).
>>
>>
>> Thanks for reading and for any help,
>> Gunnar
>>
>>
>>
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
>> Get 100% visibility into your production application - at no cost.
>> Code-level diagnostics for performance bottlenecks with <2% overhead
>> Download for free and get started troubleshooting in minutes.
>> http://p.sf.net/sfu/appdyn_d2d_ap1
>> _______________________________________________
>> Quickfix-developers mailing list
>> Qui...@li...
>> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>>
>
>
>
> ------------------------------------------------------------------------------
> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
> Get 100% visibility into your production application - at no cost.
> Code-level diagnostics for performance bottlenecks with <2% overhead
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap1
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>



-- 
Grant Birchmeier
*Connamara Systems, LLC*
*Made-To-Measure Trading Solutions.*
Exactly what you need. No more. No less.*
*
http://connamara.com

Re: [Quickfix-developers] How to suppress 35=X messages in log file

[Gunnar H. <gh...@hi...>]

Thanks Mike!

I create an acceptor with a FIX::FileLogFactory because I do want the logging to file of all messages except the 35=X. I have stream and trade sessions. For the trade session i want all logging and for stream session all but the 35=X. So I think I use the appropriate constructor of the acceptor.

Deriving from the FileLogFactory and from FileLog to create my own derived FileLog is probably easy but in the onOutgoing method of the Log object the message type is not available, only the message string and I don't want to search in the message string for 35=X because that is a waste of time and totally not nice.

Regards,
Gunnar




Gunnar Harms

On May 29, 2013, at 2:15 PM, Mike Gatny wrote:

> It is typical to set PersistMessages=Y, ResetOnLogon=Y, and use the initator/acceptor constructor that takes no LogFactory argument.  If you want to do more specific filtering, the LogFactory and Log interfaces exist for this reason and are simple to derive from.
> 
> On May 29, 2013 6:41 AM, "Gunnar Harms" <gh...@hi...> wrote:
> QuickFIX Documentation: http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX Support: http://www.quickfixengine.org/services.html
> 
> Hi All,
> 
> I want to suppress the 35=X messages in the FIX.4.2-<X>-<Y>.messages.current.log files and log all other messages (administrative, marketdatarequestrejects, etc.).
> 
> It would be easy if i could derive my own Session class from FIX::Session and override the sendRaw/send methods to not make the m_state.onOutgoing( string ) call for these message types but this ssems not possible (because in acceptor class the sessions are created from calls to a locally defined SessionFactory variable).
> 
> 
> What i could try is  deriving from ScreenLogFactory (say MyScreenLogFactory), override in that class some create() method(s) to create not a ScreenLog but a derivation (say MyScreenLog) of it and in that derived class override onOutgoing(). But then I must search in the message string for subsstrings like 35=X which is a waste of time since in FIX::Session send methods is available the MsgType.
> 
> Also some periodically running script to truncate the log files is far from optimal for obvious reasons.
> 
> 
> I am perhaps missing something simple because I cannot imagine that writing each 35=X to log file is acceptable for other users of QF (for us it is really a show stopper, log files get toooo big).
> 
> 
> Thanks for reading and for any help,
> Gunnar
> 
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
> Get 100% visibility into your production application - at no cost.
> Code-level diagnostics for performance bottlenecks with <2% overhead
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap1
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers

Re: [Quickfix-developers] How to suppress 35=X messages in log file

[Mike G. <mg...@co...>]

It is typical to set PersistMessages=Y, ResetOnLogon=Y, and use the
initator/acceptor constructor that takes no LogFactory argument.  If you
want to do more specific filtering, the LogFactory and Log interfaces exist
for this reason and are simple to derive from.
On May 29, 2013 6:41 AM, "Gunnar Harms" <gh...@hi...> wrote:

> QuickFIX Documentation:
> http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX Support: http://www.quickfixengine.org/services.html
>
> Hi All,
>
> I want to suppress the 35=X messages in the
> FIX.4.2-<X>-<Y>.messages.current.log files and log all other messages
> (administrative, marketdatarequestrejects, etc.).
>
> It would be easy if i could derive my own Session class from FIX::Session
> and override the sendRaw/send methods to not make the m_state.onOutgoing(
> string ) call for these message types but this ssems not possible (because
> in acceptor class the sessions are created from calls to a locally defined
> SessionFactory variable).
>
>
> What i could try is  deriving from ScreenLogFactory (say
> MyScreenLogFactory), override in that class some create() method(s) to
> create not a ScreenLog but a derivation (say MyScreenLog) of it and in that
> derived class override onOutgoing(). But then I must search in the message
> string for subsstrings like 35=X which is a waste of time since in
> FIX::Session send methods is available the MsgType.
>
> Also some periodically running script to truncate the log files is far
> from optimal for obvious reasons.
>
>
> I am perhaps missing something simple because I cannot imagine that
> writing each 35=X to log file is acceptable for other users of QF (for us
> it is really a show stopper, log files get toooo big).
>
>
> Thanks for reading and for any help,
> Gunnar
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
> Get 100% visibility into your production application - at no cost.
> Code-level diagnostics for performance bottlenecks with <2% overhead
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap1
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>

[Quickfix-developers] How to suppress 35=X messages in log file

[Gunnar H. <gh...@hi...>]

Hi All,

I want to suppress the 35=X messages in the FIX.4.2-<X>-<Y>.messages.current.log files and log all other messages (administrative, marketdatarequestrejects, etc.).

It would be easy if i could derive my own Session class from FIX::Session and override the sendRaw/send methods to not make the m_state.onOutgoing( string ) call for these message types but this ssems not possible (because in acceptor class the sessions are created from calls to a locally defined SessionFactory variable).


What i could try is  deriving from ScreenLogFactory (say MyScreenLogFactory), override in that class some create() method(s) to create not a ScreenLog but a derivation (say MyScreenLog) of it and in that derived class override onOutgoing(). But then I must search in the message string for subsstrings like 35=X which is a waste of time since in FIX::Session send methods is available the MsgType.

Also some periodically running script to truncate the log files is far from optimal for obvious reasons.


I am perhaps missing something simple because I cannot imagine that writing each 35=X to log file is acceptable for other users of QF (for us it is really a show stopper, log files get toooo big).


Thanks for reading and for any help,
Gunnar

[Quickfix-developers] failure to build with python2.7

[Michael F. <qui...@vi...>]

Both on SuSE 12.3 and ubuntu 10.1

./configure --with-python=/usr/include/python2.7

make[3]: Entering directory `/home/mfischer/quickfix/src/python'
/bin/bash ../../libtool --tag=CXX   --mode=compile g++ -DHAVE_CONFIG_H -I.  -I../.. -I. -I.. -I../..    -g -O2 -Wall -ansi -Wpointer-arith -Wwrite-strings     -I/usr/include/libxml2  -I/usr/include/python2.7  -MT QuickfixPython.lo -MD -MP -MF .deps/QuickfixPython.Tpo -c -o QuickfixPython.lo QuickfixPython.cpp
libtool: compile:  g++ -DHAVE_CONFIG_H -I. -I../.. -I. -I.. -I../.. -g -O2 -Wall -ansi -Wpointer-arith -Wwrite-strings -I/usr/include/libxml2 -I/usr/include/python2.7 -MT QuickfixPython.lo -MD -MP -MF .deps/QuickfixPython.Tpo -c QuickfixPython.cpp  -fPIC -DPIC -o .libs/QuickfixPython.o
QuickfixPython.cpp: In function ‘void SWIG_Python_AddErrorMsg(const char*)’:
QuickfixPython.cpp:872:42: warning: format not a string literal and no format arguments [-Wformat-security]
QuickfixPython.cpp: In function ‘PyObject* _wrap_DataDictionary_getGroup(PyObject*, PyObject*)’:
QuickfixPython.cpp:306005:9: warning: unused variable ‘res’ [-Wunused-variable]

And then it hangs, taking up 100% of cpu. I kill it after more than 1
minute of failing to make progress.

gcc (SUSE Linux) 4.7.2 20130108 [gcc-4_7-branch revision 195012]
gcc (Ubuntu/Linaro 4.6.1-9ubuntu3) 4.6.1

Any other info I can supply to get this working?

Michael
-- 

[Quickfix-developers] sorry for the typo

[Da Xu <exu...@gm...>]

Dear friends,

I am sorry for the typo in my previous email. I meant to write "Hi
developers" not "Hell developers". I wrote that email very late and I found
the typo this morning. I don't know what happened to me when I wrote it.

Best,
Da

[Quickfix-developers] segmentation fault when Session constructor is called

[Da Xu <exu...@gm...>]

Hell developers,

I found when I run either example executor or ordermatch, there is very
likely segmentation fault or std::bad_alloc error. Sometimes even new
sessions are created,
the new sessions are not stable. When the client program sends a message,
the server crashes on a segmentation fault. I found it is when the server
program calls
the constructor of Session class, the segmentation fault(or std::bad_alloc)
happens. Do you have any suggestion?

Thank you for reading.

Best,
Da

Re: [Quickfix-developers] Absence of Symbol in FIX44::NewOrderSingle constructor

[Mike G. <mg...@co...>]

On May 16, 2013 10:13 PM, "Rafael Thome Guimaraes" <p-r...@bv...>
wrote:
>
> Cann anyone tell me in wich conditions a session resets its seqnums?

Depends on your config and your counterparty.  In general, seqnums are
reset at session start.  See the StartTime config option.  See also the
ResetOn{Logon,Logout,Disconnect} config options.

Re: [Quickfix-developers] Absence of Symbol in FIX44::NewOrderSingle constructor

[Rafael T. G. <p-r...@bv...>]

Hi all!

Cann anyone tell me in wich conditions a session resets its seqnums?

Tks,
Thomé


________________________________

Esta mensagem pode conter informação confidencial e/ou privilegiada. Se você não for o destinatário ou a pessoa autorizada a receber esta mensagem, não deverá utilizar, copiar, alterar, divulgar a informação nela contida ou tomar qualquer ação baseada nessas informações. Se você recebeu esta mensagem por engano, por favor avise imediatamente o remetente, respondendo o e-mail e em seguida apague-o. Agradecemos sua cooperação.

This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, change, take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.

Re: [Quickfix-developers] SocketRecvFailed Event

[Hei C. <str...@ya...>]

1. If you use nonblocking socket class (aka SocketInitiator/SocketAcceptor), then your send call just puts your messages on a queue.
2. This can always happen because if another end unplugs the cable, the socket won't be shut down properly.  That's why there is heartbeat in the protocol.  But you can use the blocking socket class to reduce the likelihood that you get a success from sendToTarget but it is actually a failure.


________________________________
 From: yan guan <gy1...@gm...>
To: qui...@li... 
Sent: Wednesday, May 8, 2013 12:35 AM
Subject: [Quickfix-developers] SocketRecvFailed Event
 


Hello,

I have a question about QuickFIX.

1.  The moment before I call sendtoTarget() the network failed, but I can not get the Logout Event instantly. Also sendtoTarget() returns true, and in fact, the message hasn't been sent out. 

2.  Another situation is after sendtoTarget() successfully, the network failed. 

I design my project this way,  wait a period of time for the asynchronous message return. What my difficulty is if timeout and Logout, I can not whether the message is sent out or not. 


Regards,
Yan



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


------------------------------------------------------------------------------
Learn Graph Databases - Download FREE O'Reilly Book
"Graph Databases" is the definitive new guide to graph databases and 
their applications. This 200-page book is written by three acclaimed 
leaders in the field. The early access version is available now. 
Download your free book today! http://p.sf.net/sfu/neotech_d2d_may
_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers

[Quickfix-developers] SocketRecvFailed Event

[yan g. <gy1...@gm...>]

Hello,

I have a question about QuickFIX.

1.  The moment before I call sendtoTarget() the network failed, but I can
not get the Logout Event instantly. Also sendtoTarget() returns true, and
in fact, the message hasn't been sent out.

2.  Another situation is after sendtoTarget() successfully, the network
failed.

I design my project this way,  wait a period of time for the asynchronous
message return. What my difficulty is if timeout and Logout, I can not
whether the message is sent out or not.


Regards,
Yan

Re: [Quickfix-developers] Issue compiling QuickFix v1.13.3 with Python 2.7 ...

[Rodrick B. <rod...@gm...>]

Long time Noel,

it seems like your python path might not be getting set accordingly try
using the python shell and setting your PYTHON to where ever you installed
quick fix to debug

Example:

$ python
>> import sys
>> print sys.path
>> sys.path.append("quickfixpath")
>> import quickfix


Sent from my iPhone

On Apr 7, 2013, at 6:00 PM, Noel Milton Vega <nm...@co...>
wrote:

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

Hello Friends:

I have an issue with compiling QuickFix v1.13.3 with Python 2.7 support on
CentOS6 and Fedora 18 (I tried it on both).
I don't think my issue is O/S related to be honest (as I'll explain).

FYI: I'll pose my question below, but a neater version of it can be found
on StackOverflow.Com
here (easier on the eyes):


http://stackoverflow.com/questions/15860120/issue-compiling-quickfix-v1-13-3-with-python-2-7-on-centos6-and-fedora-18


Here is the sequence of steps I performed, which will show the issue:

user$ ./configure
--with-python=/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/include/python2.7
user$ make
user$ sudo make install

So far so good above. (Well, actually there are a lot of warnings during
the "make" step, but it does eventually succeed.
So next lets try to import a main module in python to see how things work.
We immediately see a problem:

user$ python -c "import quickfix"  # Lets test.
Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/home/nmvega/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/latest/lib/python2.7/site-packages/quickfix.py",
line 7, in <module>
import _quickfix
ImportError: No module named _quickfix

As you can see, python is saying that _quickfix does not exit.
Below are the Python related files that were installed by the "sudo make
install" step. I don't see _quicklib (although I do see a _quicklib.so -
not sure if that's related).

/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/_quickfix.dylib/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/_quickfix.so/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix.py/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix.pyc/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix40.py/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix41.py/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix42.py/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-pack
ages/quickfix43.py/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix44.py/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix50.py/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix50sp1.py/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix50sp2.py/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfixt11.py/usr/local/lib/libquickfix_python.la/usr/local/lib/libquickfix_python.so/usr/local/lib/libquickfix_python.so.10/usr/local/lib/libquickfix_python.so.10.0.0/usr/local/share/quickfix/quickfix.py/usr/local/share/quickfix/quickfix40.py/usr/local/share/quickfix/quickfix41.py/usr/local/share/quickfix/quickfix42
.py/usr/local/share/quickfix/quickfix43.py/usr/local/share/quickfix/quickfix44.py/usr/local/share/quickfix/quickfix50.py/usr/local/share/quickfix/quickfix50sp1.py/usr/local/share/quickfix/quickfix50sp2.py/usr/local/share/quickfix/quickfixt11.py

As a parenthetical note, also note that the following installed symlinks
(during "sudo make install" step) are broken links; meaning that the file
pointed to by the arrow (->) doesn't exist, but the link was made to point
to that anyway. Seems like an additional problem with "make install".

o /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/_quickfix.so
    -> /usr/local/lib/python/_quickfix.so  (This symlink is broken).

o /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/_quickfix.dylib
    -> /usr/local/lib/python/_quickfix.dylib (This symkink is broken, too).

In summary, I think the problems above indicate something wrong with any or
all of "configure", "make" & "make install". Then again, I could be missing
something, too.

Any ideas what is happening, and how to fix this. I know people have
successfully compiled QuickFix with Python support

Thank you very much,
NMV

------------------------------------------------------------------------------
Minimize network downtime and maximize team effectiveness.
Reduce network management and security costs.Learn how to hire
the most talented Cisco Certified professionals. Visit the
Employer Resources Portal
http://www.cisco.com/web/learning/employer_resources/index.html

_______________________________________________
Quickfix-developers mailing list
Qui...@li...
https://lists.sourceforge.net/lists/listinfo/quickfix-developers

[Quickfix-developers] Issue compiling QuickFix v1.13.3 with Python 2.7 ...

[Noel M. V. <nm...@co...>]

Hello Friends:


I have an issue with compiling QuickFix v1.13.3 with Python 2.7 support on CentOS6 and Fedora 18 (I tried it on both).
I don't think my issue is O/S related to be honest (as I'll explain).


FYI: I'll pose my question below, but a neater version of it can be found on StackOverflow.Com
here (easier on the eyes):


   http://stackoverflow.com/questions/15860120/issue-compiling-quickfix-v1-13-3-with-python-2-7-on-centos6-and-fedora-18



Here is the sequence of steps I performed, which will show the issue:
user$ ./configure --with-python=/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/include/python2.7user$ make
user$ sudo make install
So far so good above. (Well, actually there are a lot of warnings 
during the "make" step, but it does eventually succeed.

So next lets try to import a main module in python to see how things work. We immediately see a problem:
user$ python -c "import quickfix"# Lets test.Traceback(most recent call last):File"<string>",line 1,in<module>File"/home/nmvega/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/latest/lib/python2.7/site-packages/quickfix.py",line 7,in<module>import_quickfix ImportError:Nomodulenamed _quickfix
As you can see, python is saying that _quickfix does not exit.
Below are the Python related files that were installed by the "sudo make install" step.
I don't see _quicklib (although I do see a _quicklib.so - not sure if that's related).
/home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/_quickfix.dylib /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/_quickfix.so /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix.py /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix.pyc /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix40.py /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix41.py /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix42.py /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-pack ages/quickfix43.py
 /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix44.py /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix50.py /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix50sp1.py /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfix50sp2.py /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/quickfixt11.py /usr/local/lib/libquickfix_python.la /usr/local/lib/libquickfix_python.so /usr/local/lib/libquickfix_python.so.10/usr/local/lib/libquickfix_python.so.10.0.0/usr/local/share/quickfix/quickfix.py /usr/local/share/quickfix/quickfix40.py /usr/local/share/quickfix/quickfix41.py /usr/local/share/quickfix/quickfix42 .py /usr/local/share/quickfix/quickfix43.py
 /usr/local/share/quickfix/quickfix44.py /usr/local/share/quickfix/quickfix50.py /usr/local/share/quickfix/quickfix50sp1.py /usr/local/share/quickfix/quickfix50sp2.py /usr/local/share/quickfix/quickfixt11.py
As a parenthetical note, also note that the following installed 
symlinks (during "sudo make install" step) are broken links; meaning 
that the file pointed to by the arrow (->) doesn't exist, but the 
link was made to point to that anyway. Seems like an additional problem 
with "make install".
o /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/_quickfix.so ->/usr/local/lib/python/_quickfix.so (Thissymlink isbroken).o /home/user/APPS.d/ENTHOUGHT-PYTHON-IDE.d/x86_64.d/epd_free-7.3-2-rh5-x86_64/lib/python2.7/site-packages/_quickfix.dylib ->/usr/local/lib/python/_quickfix.dylib (Thissymkink isbroken,too).
In summary, I think the problems above indicate something wrong with 
any or all of "configure", "make" & "make install". Then again, I 
could be missing something, too.

Any ideas what is happening, and how to fix this. I know people have 
successfully compiled QuickFix with Python support

Thank you very much,
NMV

[Quickfix-developers] IB Example

[Juan O. <jua...@gm...>]

Anyone has some example on using quickfix with IB ?

Re: [Quickfix-developers] Sequence Reset?

[Ben <ben...@gm...>]

Hei Chan <structurechart@...> writes:

> 
> QuickFIX Documentation: 
http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX Support: http://www.quickfixengine.org/services.html
> 
> I thought that QF would ignore ResetOnLogon if I set it for an initiator, and 
QF 
> would preserve the sequence # in the file system as long as I use 
> FIX::FileStoreFactory and set ResetOnLogout and ResetOnDisconnect to N.
> 
> Anyway, I am going to try out your and Clebson's suggestion.
> 
> Thanks :)
> 
> ----- Original Message ----
> From: "Natala, Benjamin J" <benjamin.j.natala@...>
> To: Clebson Derivan <cderivan@...>; Hei Chan <structurechart@...>
> Cc: "quickfix-developers@..." 
> <quickfix-developers@...>
> Sent: Mon, May 30, 2011 1:29:51 PM
> Subject: RE: [Quickfix-developers] Sequence Reset?
> 
> I believe these lines are the issue. They only apply to Acceptors (servers) 
and 
> you are creating an Initiator (client). You should use RefreshOnLogon = Y and 
> the engine will preserve seqNums.
> 
> ResetOnLogon=N
> ResetOnLogout=N
> ResetOnDisconnect=N
> 
> See the documentation on configuration, 
> http://www.quickfixengine.org/quickfix/doc/html/configuration.html. 
> 
> Regards.
> 



It's been a while, but did this suggestion work for you?

I am having the same problem - my sessionIDs are resetting to 1, despite the 
fact that I now have these flags set:

RefreshOnLogon = Y
ResetOnLogon=N
ResetOnLogout=N
ResetOnDisconnect=N


Also, having read the documentation, I would have thought the ResetOnX = N would 
have been sufficient, but apparently not.

thanks
Ben

Re: [Quickfix-developers] error running tradeclient

[Mike G. <mg...@co...>]

Which version of quickfix are you using? Did you download the source from
the website, or did you pull from svn?

-- 
Mike Gatny
Connamara Systems, LLC

[Quickfix-developers] error running tradeclient

[<dx...@aa...>]

Dear developers,

The example tradeclient seems important to proprietary trading
applications. I have installed it in a Linux environment and compile it
successfully. When I run it, it gets the following error
"Configuration failed: FIX40.xml: Configuration failed: Field BeginString
not defined in fields section"

I try to locate where the error is generated. It turned out that it is
generated at function in DataDictionary.cpp:

int DataDictionary::lookupXMLFieldNumber
( DOMDocument* pDoc, const std::string& name ) const
{ QF_STACK_PUSH(DataDictionary::lookupXMLFieldNumber)

NameToField::const_iterator i = m_names.find(name);
if( i == m_names.end() )
throw ConfigError("Field " + name + " not defined in fields section");
return i->second;

QF_STACK_POP
}

m_names seems not been initialized so when this function is executed,
there is an error.
Does anyone have ideas to fix this problem?

Best Regards,
Da Xu

Re: [Quickfix-developers] Is Session.sendToTarget asynchronous?

[Marcelo B. <bec...@gm...>]

Hi Frank,

Thank you very much for your help. I did appreciate it.

Best regards,

Marcelo Beckmann

On Thu, Mar 14, 2013 at 8:44 PM, K. Frank <kfr...@gm...> wrote:

> QuickFIX Documentation:
> http://www.quickfixengine.org/quickfix/doc/html/index.html
> QuickFIX Support: http://www.quickfixengine.org/services.html
>
> Hello Marcelo!
>
> I cannot give you a definitive answer to your question, but I have some
> information, in line, below.
>
> On Thu, Mar 14, 2013 at 4:49 PM, Marcelo Beckmann
> <bec...@gm...> wrote:
> >
> > Dear all,
> > ...
> > I'm using quick fix C# library, I wonder if Session.sendToTarget(msg,
> sessionId) is asynchronous or blocker method.
>
> I asked a similar question, but it was about the c++ version, I didn't
> fully understand the answer, and I didn't verify the conclusion, so my
> response may not be reliable.
>
> My question and the resulting thread are here:
>
>    http://sourceforge.net/mailarchive/message.php?msg_id=30448623
>
> My supposition is that if you use a SocketInitiator (or SocketAcceptor)
> (and I assume that the same classes exists in the c# version), then
> the call to FIX::Session::sendToTarget (your Session.sendToTarget)
> is synchronous / blocking, that is, the full call is processed by the
> calling thread.
>
> But if you use a ThreadedSocketInitiator (or ThreadedSocketAcceptor),
> then the call is asynchronous / non-blocking, that is, the calling thread
> posts the call to some queue, returning essentially immediately, and
> some other thread plucks the call off of the queue and does the real
> work.
>
> Again, that's my working theory, but don't have any conclusive evidence
> for it.  It would be great if someone who really knows could chime in.
>
> (I did try reading the code, but got lost in what appeared to be an
> infinite
> recursion, so I must not have been looking at it right.)
>
> > ...
> > Best regards,
> >
> > Marcelo Beckmann
>
>
> Good luck, and Happy FIXing!
>
>
> K. Frank
>
>
> ------------------------------------------------------------------------------
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_d2d_mar
> _______________________________________________
> Quickfix-developers mailing list
> Qui...@li...
> https://lists.sourceforge.net/lists/listinfo/quickfix-developers
>



-- 
Marcelo Beckmann

Re: [Quickfix-developers] Is Session.sendToTarget asynchronous?

[K. F. <kfr...@gm...>]

Hello Marcelo!

I cannot give you a definitive answer to your question, but I have some
information, in line, below.

On Thu, Mar 14, 2013 at 4:49 PM, Marcelo Beckmann
<bec...@gm...> wrote:
>
> Dear all,
> ...
> I'm using quick fix C# library, I wonder if Session.sendToTarget(msg, sessionId) is asynchronous or blocker method.

I asked a similar question, but it was about the c++ version, I didn't
fully understand the answer, and I didn't verify the conclusion, so my
response may not be reliable.

My question and the resulting thread are here:

   http://sourceforge.net/mailarchive/message.php?msg_id=30448623

My supposition is that if you use a SocketInitiator (or SocketAcceptor)
(and I assume that the same classes exists in the c# version), then
the call to FIX::Session::sendToTarget (your Session.sendToTarget)
is synchronous / blocking, that is, the full call is processed by the
calling thread.

But if you use a ThreadedSocketInitiator (or ThreadedSocketAcceptor),
then the call is asynchronous / non-blocking, that is, the calling thread
posts the call to some queue, returning essentially immediately, and
some other thread plucks the call off of the queue and does the real
work.

Again, that's my working theory, but don't have any conclusive evidence
for it.  It would be great if someone who really knows could chime in.

(I did try reading the code, but got lost in what appeared to be an infinite
recursion, so I must not have been looking at it right.)

> ...
> Best regards,
>
> Marcelo Beckmann


Good luck, and Happy FIXing!


K. Frank

[Quickfix-developers] Is Session.sendToTarget asynchronous?

[Marcelo B. <bec...@gm...>]

Dear all,

First of all many thanks for the information exchange provided by you.

I'm using quick fix C# library, I wonder if Session.sendToTarget(msg,
sessionId) is asynchronous or blocker method.

Thanks in advance for your support,

Best regards,

-- 
Marcelo Beckmann