-
From the spec:
data: Raw data with no format or content restrictions. Data fields are always immediately
preceded by a length field. The length field should specify the number of bytes of the value of
the data field (up to but not including the terminating SOH). Caution: the value of one of
these fields may contain the delimiter (SOH) character. Note that the value specified for this...
2009-11-06 13:04:22 UTC by crogers
-
List of files modified to fix the 64-bit p_thread functions' truncation issues:
1. Acceptor.h
2. HttpServer.h
3. Initiator.h
4. ThreadedSocketAcceptor.h
5. ThreadedSocketAcceptor.cpp
6. ThreadedSocketInitiator.h
7. ThreadedSocketInitiator.cpp
8. Utility.h
9. Utility.cpp.
2009-10-28 15:00:11 UTC by snreddy
-
The treaded sokcet initiator does not use the SocketNodelay setting from the config ini file.
Please amend ThreadedSocketInitiator::doConnect and add the following after the call to int socket = socket_createConnector(); in line 148:
int socket = socket_createConnector();
if ( socket != -1 )
{
if ( m_noDelay )
socket_setsockopt( socket, TCP_NODELAY );
}.
2009-10-19 19:55:46 UTC by troelstm
-
GLOBAL.event.log and GLOBAL.messages.log do not appear to be closed by Initiator when the object is destroyed.
Session destroys its logs:
Session::~Session()
{
// ...
if ( m_pLogFactory && m_state.log() )
m_pLogFactory->destroy( m_state.log() );
// ...
}
but Initiator does not:
Initiator::~Initiator()
{
// ?
}
I added the following to Initiator::~Initiator(...
2009-10-13 10:28:54 UTC by paulvr
-
building with gcc-4.4 requires
#include
in
src/C++/Utility.h
and
src/C++/ConfigLexer.cpp
to avoid
"error: sprintf and EOF not declared".
2009-10-12 21:58:32 UTC by rudi_meier
-
Hi ..
i am searching for C# sample programs on QucikFix engine... Kindly share some information with me on this..
Regards,
Ritesh.
2009-09-30 03:08:51 UTC by riteshkothari
-
I have the very similar problem at ~Dictionary(). If I comment out the destructor:
~Dictionary()
{
//delete m_pUnmanaged;
}
It's not crashing anymore.
Definitely, something is wrong in Dictionary. As I understand it serves for FIX message.
The problem occurs on Windows Server 2003. At WinXP and Vista -- everything works fine.
2009-09-25 14:56:44 UTC by sgazik
-
This patch resolve the issue described in
https://sourceforge.net/tracker/?func=detail&aid=2851230&group_id=37535&atid=1126912
pthread_t on Linux is defined as unsigned long. The variable storing the thread_id within various quickfix classes is defined as an unsigned [int]. Which causes problems specifically during thread joins on exit as the thread_id doesn't match the original id...
2009-09-14 16:22:35 UTC by systradingllc
-
This is a Patch that will resolve issue with Thread ID on 64-bit machines using PTHREAD. Tested on Ubuntu 9.04 with g++ 4.3.3.
2009-09-14 14:37:07 UTC by systradingllc
-
In thread_spawn (for example) "pthread_create" result is a pthread_t type (unsigned long int on 64-bit machines) but is being truncated to a 32 bit unsigned. All thread joins functions then use the incorrect thread id and will sometimes crash the app.
2009-09-04 14:35:20 UTC by deanlavdas
