# ClOrdID uniqueness

**URL:** <https://forum.fixtrading.org/t/clordid-uniqueness/2519>\
**Category:** General Q&A\
**Tags:** general-qa, xid\_166144, imported\
**Created:** [April 15, 1999, 11:09am UTC](https://forum.fixtrading.org/t/clordid-uniqueness/2519 "1999-04-15T11:09:00Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![fplpo\_x](https://avatars.discourse-cdn.com/v4/letter/f/7ab992/32.png) [@fplpo\_x](https://forum.fixtrading.org/u/fplpo_x)\
**Post date:** [April 15, 1999, 11:09am UTC](https://forum.fixtrading.org/t/clordid-uniqueness/2519/1 "1999-04-15T11:09:00Z")

</div>

Imported from previous forum

---

<div class="post-metadata">

**Author:** ![fplpo\_x](https://avatars.discourse-cdn.com/v4/letter/f/7ab992/32.png) [@fplpo\_x](https://forum.fixtrading.org/u/fplpo_x)\
**Post date:** [April 15, 1999, 11:09am UTC](https://forum.fixtrading.org/t/clordid-uniqueness/2519/2 "1999-04-15T11:09:00Z")

</div>

[original email was from Koji Ishikawa - ishikawa@quick.co.jp]  
Quick, which is a part of technical working  
group in Japan currently, is planning to  
provide FIX message routing service between  
buy side and sell side.

The possible problem is that ClOrdID is  
supposed to be unique by institution, however,  
if mulitple buy side happen to put the same  
value in this field and send orders to the  
same sell side, it will not be unique  
anymore on the sell side. This seems to be  
very likely if everyone uses sequence numbers  
for the field.

Are sell side firms supposed to combine  
"OnBehalfOf" field and "ClOrderID" to  
make up an unique value within their system?

Or is the service vendor, such as Quick,  
supposed to convert ClOrdID to some  
unique value and deliver to the sell side?  
In this case, the buy side and sell side  
probably need to know what the converted  
value is, which may be very confusing.

---

<div class="post-metadata">

**Author:** ![fplpo\_x](https://avatars.discourse-cdn.com/v4/letter/f/7ab992/32.png) [@fplpo\_x](https://forum.fixtrading.org/u/fplpo_x)\
**Post date:** [April 15, 1999, 12:18pm UTC](https://forum.fixtrading.org/t/clordid-uniqueness/2519/3 "1999-04-15T12:18:00Z")

</div>

[original email was from Mark Hinman - [mhinman@easetech.com](mailto:mhinman@easetech.com) ]  
\> Quick, which is a part of technical working  
\> group in Japan currently, is planning to  
\> provide FIX message routing service between  
\> buy side and sell side.  
\>  
\> The possible problem is that ClOrdID is  
\> supposed to be unique by institution, however,  
\> if mulitple buy side happen to put the same  
\> value in this field and send orders to the  
\> same sell side, it will not be unique  
\> anymore on the sell side. This seems to be  
\> very likely if everyone uses sequence numbers  
\> for the field.  
\>  
\> Are sell side firms supposed to combine  
\> "OnBehalfOf" field and "ClOrderID" to  
\> make up an unique value within their system?  
\>  
\> Or is the service vendor, such as Quick,  
\> supposed to convert ClOrdID to some  
\> unique value and deliver to the sell side?  
\> In this case, the buy side and sell side  
\> probably need to know what the converted  
\> value is, which may be very confusing.  
\>  
\>

If a message is received across FIX, then the combination of SenderCompID and ClOrdID (or any other ID) forms the unique identifier. Any internally generated message is inherently unique.

---

<div class="post-metadata">

**Author:** ![scott.atwell](https://avatars.discourse-cdn.com/v4/letter/s/51bf81/32.png) [@scott.atwell](https://forum.fixtrading.org/u/scott.atwell)\
**Post date:** [April 15, 1999, 12:36pm UTC](https://forum.fixtrading.org/t/clordid-uniqueness/2519/4 "1999-04-15T12:36:00Z")

</div>

If a single FIX session represents multiple firms, then the sellside to use OnBehalfOfCompID + ClOrdID. OnBehalfOfCompID will identify the institution and the ClOrdID should be unique for that institution. Do you have a situtation in which for a single OnBehalfOfCompID, the ClOrdID value would not be unique?

\> \> Quick, which is a part of technical working  
\> \> group in Japan currently, is planning to  
\> \> provide FIX message routing service between  
\> \> buy side and sell side.  
\> \>  
\> \> The possible problem is that ClOrdID is  
\> \> supposed to be unique by institution, however,  
\> \> if mulitple buy side happen to put the same  
\> \> value in this field and send orders to the  
\> \> same sell side, it will not be unique  
\> \> anymore on the sell side. This seems to be  
\> \> very likely if everyone uses sequence numbers  
\> \> for the field.  
\> \>  
\> \> Are sell side firms supposed to combine  
\> \> "OnBehalfOf" field and "ClOrderID" to  
\> \> make up an unique value within their system?  
\> \>  
\> \> Or is the service vendor, such as Quick,  
\> \> supposed to convert ClOrdID to some  
\> \> unique value and deliver to the sell side?  
\> \> In this case, the buy side and sell side  
\> \> probably need to know what the converted  
\> \> value is, which may be very confusing.  
\> \>  
\> \>  
\>  
\> If a message is received across FIX, then the combination of SenderCompID and ClOrdID (or any other ID) forms the unique identifier. Any internally generated message is inherently unique.  
\>

---

<div class="post-metadata">

**Author:** ![fplpo\_x](https://avatars.discourse-cdn.com/v4/letter/f/7ab992/32.png) [@fplpo\_x](https://forum.fixtrading.org/u/fplpo_x)\
**Post date:** [April 16, 1999, 1:03am UTC](https://forum.fixtrading.org/t/clordid-uniqueness/2519/5 "1999-04-16T01:03:00Z")

</div>

[original email was from Koji Kubota - [koji.kubota@gs.com](mailto:koji.kubota@gs.com) ]  
In Japan extension working group, which is  
discussing how to use FIX in Japan, we  
have agreed that senders placing orders  
are responsible for putting unique values  
in ClOrdID. We can assume that the ClOrdID  
value is unique for a single OnBehalfOfCompID.

\>\> If a single FIX session represents multiple firms, then the sellside to use OnBehalfOfCompID + ClOrdID. OnBehalfOfCompID will identify the institution and the ClOrdID should be unique for that institution. Do you have a situtation in which for a single OnBehalfOfCompID, the ClOrdID value would not be unique?

It seems that sell side should do the following;  
if OnBehalfOfComID exists then  
use OnBehalfOfCompID + ClOrdID  
else  
use SenderCompID + ClOrdID

A third party between sell and buy side  
should not convert ClOrdID, and it is  
a responsiblity of sell side (receiver)  
to use the logic mentioned above.

Is there anyone in US, who had some  
problem regarding this issue? As far as  
I know of, Bloomberg is doing the switching  
service. Did Bloomberg do anything with  
ClOrdID?

\> \> If a message is received across FIX, then the combination of SenderCompID and ClOrdID (or any other ID) forms the unique identifier. Any internally generated message is inherently unique.  
\> \>