Home > Healthcare Integration > Upgrading our production Rhapsody Environment to a VM Cluster

Upgrading our production Rhapsody Environment to a VM Cluster

After much work and testing, we will be upgrading our production Rhapsody environment from a VM instance to a VM cluster and updating a large number of interfaces from Rhapsody 3.2 to 4.1.1. Our team has been doing this as a side project for over a year while continuing to develop numerous production interfaces to support our growing health care and business integration needs.

We experienced many technical challenges along the way but have overcome them by using good design principals and best practices.

Update: We had to roll because the VM environment was configured incorrectly. Our data center team is working with Orion Health to resolve the problem.

Update: This July we were successful in getting a windows active / passive cluster configured and our Rhapsody configuration installed on it. So far we have had no issues.

Advertisements
  1. Daniel Wood
    October 17, 2014 at 5:15 am

    Hi Steve,
    I am working with a group that is attempting to set up a clustered Rhapsody environment as well. I am wondering if you have any documentation around how you over came the issues you tackled?

    • October 17, 2014 at 5:22 pm

      Daniel I have not updated my blog lately so it does not reflect reality. We backed off of the clustered VM active / passive configuration because of issues with our data center being able to support it. The cluster worked well when it was configured to meet Orion Health’s requirements but every now and then someone from the data center would change it and things would break. So we are back to a single node VM environment and that breaks every now and again as they decide that the Orion Health requirements really don’t mean what they say.

      Good luck on your endeavor.

  2. Daniel Wood
    October 19, 2014 at 9:55 pm

    Ah no worries.
    Currently we are just using native windows clustered resourcing which is working ok we have just run into one last issue with aliases not following the active node.

    thanks anyway.

    • October 19, 2014 at 11:27 pm

      You are welcome. The documentation from Orion Health was complete so if you follow it your team should not have a problem setting up a VM cluster.

  3. Daniel Wood
    October 20, 2014 at 3:30 am

    Yea no worries we got it all sorted and working this morning!

    I have a few other questions that I wouldnt mind asking you if thats ok?
    Can you shoot me an email? (I have entered it below)

    Cheers

  4. October 20, 2014 at 3:02 pm

    Here you go.

    • Wood, Daniel (OCIO) (DHHS)
      October 20, 2014 at 10:03 pm

      Hi Steve,

      We have 2 issues at the moment which we are looking for solutions for.

      1. Long term storage and easy access to historical data. OOTB Orion supports 30 days of storeage in the active data store which is fine but then for a further 3 months the data is sent to log files and then Zipped up for 3 months. Past this there is no support. We have a requirement to store the data for 7 years and we want an easy way to be able to reply the messages should there be any issues of any kind. Do you have any advice on this one?

      2. I am also interested to know if you put anything in place around the use of firewalls to white list which server can come in and on which ports. As it stand any server with the known details of Rhapsody can send data to Rhapsody!
      Looking forward to seeing you thoughts.

      Cheers,

  5. October 21, 2014 at 2:24 am

    7 years is a long time. The only solution would be to extract metadata from each message and store it along with the message in a database. That way you could locate each message and retrieve it if necessary.

    All of the connections to Rhapsody are either local or via a VPN.

  6. Daniel Wood
    October 21, 2014 at 2:36 am

    7 years is the national policy for storage of records in Australia!
    No worries thanks

    • October 21, 2014 at 7:44 pm

      You have to save all HL7 messages processed by a Rhapsody engine for 7 years? Is this at a health care facility or an IHE?

  7. Daniel Wood
    October 21, 2014 at 9:24 pm

    This is at the department of health

    • October 22, 2014 at 12:59 am

      Thanks you have a significant challenge.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: