Unsolicited Cancel in a pending Cancel state

, ,
1

Imported from previous forum

2

In Fix 4.2:
If we send the sellside a cance replace request, they send back a pending cancel, and then do an Unsolicited Cancel should the sellside first send us an order cancel replace reject? Or does the Cancel 39=4 150=4 override the need for a response from the order cancel replace?

Thanks,

James

3

[ original email was from John Prewett - jprewett@lavatrading.com ]
> In Fix 4.2: If we send the sellside a cance replace request, they send

back a pending cancel, and then do an Unsolicited Cancel should the
sellside first send us an order cancel replace reject? Or does the
Cancel 39=4 150=4 override the need for a response from the order
cancel replace?

Thanks,

James

Hi James,

Here is what I believe would conform to the FIX.4.2 standard from the sell-side:

  1. Sends an ExecReport(PendingReplace) - not the pending cancel that you indicated above.
  2. Sends an ExecReport(Canceled) using the original ClOrdID, indicating an unsolicited cancel of the original order.
  3. Sends a CancelReject as a response to the CancelReplaceRequest.

Note that the CancelReject should come after the unsolicited cancel of the original order.
While it is possible for your application to process the unsolicited cancel and determine that, by nature, the CancelReplaceRequest is presumably defunct, I believe an explicit response is required by the protocol and certainly ties up any loose ends.

While I don’t believe there is a 100% example of this in the FIX.4.2 Appendix D, the closest example would be D9 where a final fill “crosses in the mail” with a CancelReplaceRequest. If you were to substitute the final fill for an unsolicited cancel, you would get an an appropriate example (albeat without the ExecReport-pendingReplace).

I hope this helps.

John Prewett