Newbie Questions

[Mike Keys]

Hello group. Just got back from McKesson’s Level 1 interface training and have some questions that I hope you can answer. I am the interface engineer for a hospital in Illinois and have been working with the eGate product for almost 9 years. All of our interface development has been done with eGate. We decided this last year to purchase Cloverleaf for a couple of reasons: 1) This is what McKesson uses and it aids troubleshooting and implementation of new interfaces with McKesson products and 2) SUN will eventually EOL eGate 4.5.3 and we will have to migrate anyway.

Now for the questions:

1. We purchased two HP rp4440 servers with HBA’s. Our intentions (based on sales advice) was to SAN attach the servers and install Cloverleaf. We wanted some way to save the messages on the SAN in a data repository. Currently we are only able to go back 14 days on most systems to view or resend HL7 messages. We wanted the capability to save a few months worth. We used the SMAT tool in class but it appears that we will need to load Cloverleaf on a filesystem that is mounted on the SAN space in order to have enough space available to save a few months worth of messages. Do I understand this to be correct?

2. Is there a better way to search messages? SMAT doesn’t offer this (or does it) and I would like to break down the location to save messages into folders that are associated with the thread name. Beyond this, I would like to save the messages based on the month and day, with each message named with the timestamp that it was sent. Is it possible to do this with the product now or will we need to setup an external database to house messages.

3. Building messages. With eGate, I have the ability to build a HL7 message from scratch (not a variant). Then, based on this structure, I can ID/Route (say, just ADT/MFN/ORU) the message and call any translations to run against this structure. I asked this question in class and was told that perhaps this would be better handled in a Tcl script as I could define the message structure in there. Is this correct?

4. Disaster Recovery. Do most facilities run their system in a cold standby setup? Is anyone using HP’s Serviceguard to run Cloverleaf?

I appreciate your responses.

Topic

[Todd Lundstedt]

Mike,

re: item 1–

You would have to know your own message load over time to determine how much space you need to store several months worth of messages.  Take your 14 day size total, multiply by 4 to get 2 months.. that should be simple enough.  SAN space would be determined by your calculations.

re: item 2–

I have been developing a set of scripts to save messages historically.  They are in no way automated at this stage; I am just getting it down to how I want to manage, and where I want to keep files.  Specifically, we save .idx files for all protocol threads on our server (for later parsing by the histogram script posted somewhere on this forum, histSmat.pl I think) and then we save .idx and .msg files for a select group of threads where we think it is beneficial to have saved messages.  My scripts site_thread_datetime stamp each file.  However, they are KSH scripts for Cloverleaf on AIX.  If you aren’t using AIX, or some other real OS, you might have trouble using the scripts as they are, but you could certainly use them to build your own.

[Richard Hart]

Mike.

1. We have a housekeeping script that cycles all our SMAT files (one file for each application incoming message).  The files are cycled at midnight, date stamped and zipped.  We store ALL SMAT files in a repository and lave a months worth under the Cloverleaf site

2. I wrote a Perl script a few years ago that reads the SMAT (.msg) files and extracts messages based on the selection of patient code or patient account.  this is tailored to our message and have some tweaks for old SMAT files where messages were incomplete!  I can share, but you may need to understand Perl!

eg

Usage: /hci/InfoHEALTH/bin/smatList.sh  [-v] [-h] [-p] [-fp] [-a] [-fa] [-m] [-fm] -o [file2] …

Where -h          help!

     -v          verbose mode

     -o    the output file. See Note below

     -p  the Patient Code (MRN). This can be repeated

     -fp   a file containing Patient Codes (MRN), one per line

     -a  the Account Number. This can be repeated

     -fa   a file containing Account Numbers, one per line

     -m  the Medical Staff ID or GP Code (for Master File Updates). This can be repeated

     -fm   a file containing Medical Staff codes, one per line

     file1       is a Cloverleaf SMAT file, compressed or uncompressed

    The output file supplied will have a suffix _xxx.dat, where

    xxx is the file number in the range 1 through 999.

    Each file will contain  a maximum of 1000 records.

3. This is not something we have done, but I have played with the grm commands in a TCL script to change HL7 to FRL without the Xlate, so I guess that this is possible.

   Wrt the routing, we route message on different things and (in the GUI) set the message type to FRL so we can call a TCL proc to determine the routing.  We have used this for years without issue.

4.  We are on AIX and have two servers, with one sever LPAR’d for test and prod (secondary).  We have ‘rc’ scripts that stop and start the Cloverleaf sites as appropriate.

    There are some Smarts so that CPU’s are ‘moved’ from the test LPAR to the prod LPAR when a failover is called.

[Kathy Zwilling]

Richard,

I would be very interested in your Perl script.  Would I be able to get that form you or would you be willing to post it for all of us?

Thank you!

Kathy

[Richard Hart]

I’ve posted our SMAT list tool.

I hope it helps.

https://usspvlclovertch2.infor.com/viewtopic.php?p=10050#10050

[Kathy Zwilling]

Thanks, Richard!

[Russ Ross]

My reply wasn’t meant for this post so I deleted it and move to the correct cloverltech post.

Russ Ross
RussRoss318@gmail.com

[Mike Campbell]

Mike, we are also in the process of converting our eGate interfaces over to Cloverleaf.  I have found that Cloverleaf is a WHOLE lot easier to work with.

That being said, we have discovered some nuances that are interesting.  We have Cerner as our HIM system and interfacing with a lot of different vendors.  If you have questions or want to discuss how we are doing things don’t hesitate to contact me.

[Jim Kosloskey]

Mike,

Item 3: you can construct a variant to match any structure you want I don’t really see a need to use Tcl unless you just want to.

In other words, make a variant for the message as the trading partner dictates it should be structured.

However, with most vendors although they don’t follow the HL/7 standard closely, usually their message structures can be used within the standard variants with few if any modifications to the structure.

If you would like to chat directly about specifics, email me and I will see if I can help.

email: jim.kosloskey@jim-kosloskey.com 29+ years Cloverleaf, 59 years IT - old fart.

[Author]

Replies