We are using DG REGISTER PATIENT. We confirmed that 1. HL7 PING works from WorldVistA to Mirth using the setup below; 2. that CALL^%ZISTCP successfully communicates with Mirth. So we know it's not a Mirth issue.
Here's the Logical Link:
Select HL LOGICAL LINK NODE: MIRTH TCP
ANOTHER ONE:
STANDARD CAPTIONED OUTPUT? Yes// (Yes)
Include COMPUTED fields: (N/Y/R/B): NO// - No record number (IEN), no Computed
Fields
NODE: MIRTH TCP LLP TYPE: TCP
STATE: Enabled TIME STARTED: FEB 17, 2010@12:59:43
SHUTDOWN LLP ?: NO QUEUE SIZE: 10
TCP/IP ADDRESS: 127.0.0.1 TCP/IP PORT: 12346
TCP/IP SERVICE TYPE: CLIENT (SENDER)
Here are the protocols that DG REGISTER PATIENT fires:
NAME: VAFC ADT-A04 SERVER
ITEM TEXT: This protocol fires off of the PIMS Registration option
TYPE: event driver CREATOR: WVEHR,PATCH INSTALLER
PACKAGE: REGISTRATION
DESCRIPTION: This server protocol fires when a patient is registered. It
generates a Health Level Seven (HL7) register a patient (event code A04)
message.
TIMESTAMP: 60991,21593 SENDING APPLICATION: VAFC PIMS
TRANSACTION MESSAGE TYPE: ADT EVENT TYPE: A04
PROCESSING ID: P ACCEPT ACK CODE: NE
APPLICATION ACK TYPE: NE VERSION ID: 2.3
SUBSCRIBERS: RG ADT-A04 TRIGGER
NAME: RG ADT-A04 TRIGGER
ITEM TEXT: MPI/PD ADT-A04 Trigger Protocol
TYPE: subscriber CREATOR: WVEHR,PATCH INSTALLER
PACKAGE: CLINICAL INFO RESOURCE NETWORK
DESCRIPTION: This protocol is used as an internal process of the A04
messages. The protocol will read in the message and reformat to the HL7 v2.4
standard with Application and Commit acknowledgements turned on.
TIMESTAMP: 59147,62050 RECEIVING APPLICATION: RG ADT
EVENT TYPE: A04 LOGICAL LINK: MIRTH TCP
RESPONSE MESSAGE TYPE: ACK PROCESSING ROUTINE: Q
SENDING FACILITY REQUIRED?: YES RECEIVING FACILITY REQUIRED?: YES
ROUTING LOGIC: D PROC^RGADTP
--
Sam Habiel
Hi Sam,
There could actually be several areas that you could check, but I think you would want to start with answering the question of where this is getting hung in the process, ie, is it creating the A04 message, but just not sending to the mirth?
Carolyn, I actually figured it out. I posted on hardhats. Bascially the issue was that the subscriber I decided to hijack has a Routing Logic of 'Q', and this caused it to ignore the Logical Link field, so it never bothered the link at all. Still investigating receiving messages. Sam
Detailed answer on hardhats regarding set-up. Document showing set-up to follow soon.
Sam, can you provide a link to the message? I don't see it...
On a possibly related topic, the experimental code I put up has survived some testing this past week at a customer site, so I've created a bzr branch and a KIDS build for it. The code touches HLCSTCP1, HLCSTCP2, HLCSTCP3, and HLCSTCP4. Here's the bug report, the branch, and a diff (click Expand All to see everything). The code should improve reliability both for sending (outbound HL7 interface) and receiving (inbound HL7 interface), and will be part of future GT.M integration and OpenVista Server releases (unless more problems are found, of course!)
Jon, patching OV to the latest HL7 patches does not require any other dependencies other than the pathces be applied in order. Can we make the mods you have made based on the latest routines from an up to date patching? I think that would be better in the long term.
I certainly agree that it will be better in the long term. Unfortunately, I don't have the bandwidth to push an update like that through right now. I know it seems like a small change, but there's a lot of work involved getting it tested and pushed "upstream" into OpenVista Server. It would be much easier to make such a change when we upgrade all of the packages to the latest FOIA. My patch only touches a very specific code path, so it minimizes the amount of re-testing that needs to be done.
That said, I tried to make my patch as easy to understand as possible, so if you want to integrate it on top of the latest FOIA routines, you're welcome to do that. If you do, please report back on how well it works afterward -- is there anything new the VA added or additionally fixed, etc.
Jon, Sorry for the delay. MPI intervened in my life momentarily. Hopefully we will continue working on HL7 today. Here's the link: http://groups.google.jo/group/hardhats/msg/b248c364645f88fa