Date/Time: Sat, 15 May 2021 12:21:13 +0000
Post From: TT Based Order Routing Service: TT Interruption
Sierra Chart Engineering - Posts: 96236 | Ending Date: 2022-04-10
See the "Update" at the bottom of this post, because we have tentatively resolved the issue described in this post at this time. 18:50 US Eastern time 2021-02-16.
We no longer consider there to be an urgent issue at this time. Below is the existing text as it was written earlier.:
This message is going to be going out through the Sierra Chart system notification window to all of ours users on the Sierra Chart Order Routing Service. You will also get an email message about this.
TT (Trading Technologies) definitively has serious problems at this time. We have a total of six servers and six independent connections to TT.
Two of these, including Aurora, are unavailable at this time because the TT FIX server is unresponsive to a connection and/or it is not going out of the recovery state if there is a successful connection. This is a TT issue. Not a Sierra Chart issue. But obviously it is affecting the Sierra Chart order routing service.
Affected servers are:
We have closed down the connections to those two servers. Users should be using
Or one of the other servers. Server table:
Order entry is possible on these other servers.
There is nothing you need to do, unless you see Sierra Chart continuously trying to reconnect because Server 1 will be one of the available backup servers by default. So the connection to the backup will happen automatically. So this is one good thing about the Sierra Chart Order Routing Service is that the connectivity is very redundant.
You can control servers through Global Settings >> Data/Trade Service Settings>> Service Setting >> Primary Server/Backup Server. You want to make sure one of those servers contains Server 1.
Now one thing you should be aware of is that if you have working orders that are in an OCO group, cancel those orders and resubmit the OCO. If the OCO order is managed on either Server 2 or 4, there is no longer any OCO management at this time. Meaning when one side of the OCO order pair fills, the other side will not be automatically canceled.
TT has been notified about this problem. We are not very impressed with their response. As if they seem to not even be aware there is a problem. But they said they are looking into it.
Order routing is functional at this time.
The main concern is OCO orders. If you have a bracket order or an OCO order, cancel those orders and resubmit them following the instructions below or just close out your position altogether because if the OCO management is on Server 2 in Aurora or Server 4 in Equinix, those order routing servers have no connection to TT, and cannot manage the OCO. Meaning that when one side Fills, the other side will not be automatically canceled. In this case you would have to manually cancel it.
So you need to cancel those OCO orders, and resubmit them following the instructions here:
Or get out of the position altogether.
Update 2021-02-16 18:50 US Eastern time :
We now have connectivity back to TT on Server 2 and Server 4. But we are not allowing users to connect to Server 2. Only Server 4. We accomplished this, by using the "bypass recovery" option from TT on logon. But we are not fully aware of any unusual consequences of doing this.
So therefore there should be OCO management for those orders in an OCO group which exist on Server 2 and Server 4. But as a precaution we do recommend still canceling any OCO orders that you have submitted earlier, prior to 2021-02-16 17:00 US Eastern time and resubmitting them
Update 2021-02-16 19:05 US Eastern time:
We consider the situation now stable and we believe OCO orders are able to be managed now which are managed by the Sierra Chart order routing service.
We are now allowing users to connect to Server 2, the Aurora order routing server.
For background information about how OCO and bracket orders are managed with this service, refer to:
Update 2021-02-16 19:09 US Eastern time.
Message received from TT:
My apologies for the disruption. Our Operations Team was able to re-direct your connections to another server and they appear to be connected.
We are going to investigate this issue further with our Engineers.
Please let us know if issues persist.
Update 2021-02-16 19:20 US Eastern time.
Once again situation is considered stable. And we are not aware of any outstanding issues. But we have heard that TT position quantities may be inaccurate. You will need to ask your broker about this if a position quantity looks inaccurate.
Despite the problem from TT Sunday evening and today, we are definitely very happy with the Sierra Chart Order Routing architecture.
The redundancy, worked exactly as expected. Users were automatically connected to a backup server upon failed connectivity or upon a failed order submission. All of that worked exactly as expected.
Due to the stable and redundant architecture and flexible design we were able to bypass the recovery process ("Recovery" is a TT FIX concept with their service) and get order routing and order management working once we had the connection restored to TT. And these connections are now over the internal cross connect in the Aurora data center.
Sierra Chart Support - Engineering Level
Your definitive source for support. Other responses are from users. If possible please keep your questions brief and to the point. Please be aware of support policy:
If your question/request has been answered and you do not have anything further, then it is easiest for us if you do not reply again to say thank you.
Date Time Of Last Edit: 2021-02-17 00:28:30