Barry,
Thanks for the reply. I am leaving the CheckLatency and MaxLatency with
their default values: I am receiving market data, so if something gets to me
after 120 seconds, it is pretty worthless.
My concern is with the HeartBeats I send out which do not make it to the
other side in time. I have limited control over the other side, and I want
to make sure I am not delaying the outgoing heartbeats in any way. If,
indeed, field 52 is the timestamp when the message actually goes out on the
socket (and is not sent out at a later time by some thread), I know that I
am doing all I can on my side.
Thanks.
_____
From: Bishop, Barry [mailto:Bar...@gs...]
Sent: Thursday, January 06, 2005 12:09 PM
To: 'Victor Glava'
Cc: qui...@li...
Subject: RE: [Quickfix-developers] Sending time question
Are you aware of the CheckLatency field in the settings file?
This could help things at your end.
Regards,
barry
-----Original Message-----
From: Victor Glava [mailto:vic...@ch...]
Sent: Thursday, January 06, 2005 4:39 PM
To: qui...@li...
Subject: [Quickfix-developers] Sending time question
I have a question regarding the sending time in the header...
When sending any message to a counter party, like the following:
<20050105-20:15:00, FIX.4.4:GATEWAY->ADAPTER, outgoing>
(8=FIX.4.4^9=74^35=0^34=34^49=ENLON00WK002^52=20050105-20:15:00.831^56=CTC_M
DAdapterEurex^10=162^)
Is the SendingTime (tag 52) the time the message is actually sent out, or is
it the time when it is placed on a queue, meaning there might be a lag
before the message actually goes out...
My concern is that I am noticing a lag between the time the heart-beat
messages goes out and the time it is received at the other end, which, at
times, causes the other side to logout/disconnect. This might be due to
network delay, but I want to be sure it is not something on my end.
I am using 1.9.4.
Thanks.
|
