Mirth Connect Support and Managed Services

Server uptime monitoring is not interface support. An engine can be running perfectly while a channel silently drops results, a queue backs up behind an unreachable endpoint, or a certificate expires on an outbound connection. Genuine Mirth support means watching the message layer, not the host, and having someone who can read a channel script at three in the morning without first spending two hours working out what it does. That distinction is the entire difference between a hosting contract and an interface support contract.

We publish support pricing rather than quoting it privately, so you can rule us in or out before spending time on a call. Tiers are set by estate size and response-time requirement rather than by feature gating, and the engineering hours in each tier are usable for maintenance, enhancement or new channel work rather than reserved solely for incidents. All tiers include a Business Associate Agreement executed before any access to an environment carrying protected health information.

A service level agreement is only meaningful if severity is defined before an incident rather than negotiated during one. We classify by clinical and operational impact rather than by system, because a single failing results interface can matter more than an entire non-clinical integration being offline. Response commitments below are to engineering engagement, not to an automated acknowledgement, and resolution targets are agreed per environment during onboarding rather than promised generically on a web page.

Support contracts fail on scope ambiguity more often than on service quality. A retainer that appears to cover everything and then excludes the thing you needed is worse than a narrower agreement stated plainly. The boundaries below are how we write engagements, and they are visible before signature rather than discovered at the first invoice. Where something falls outside the retainer we quote it as a separate piece of work rather than declining it.

The value of a support retainer is measured in incidents that never reached a clinician, which makes it inherently hard to demonstrate and easy to underfund. Our approach weights prevention heavily: alerting thresholds tuned per channel, queue depth watched rather than just error counts, and certificate and credential expiry tracked before it causes an outage. Silent failure is the expensive kind, and most of the incidents we respond to in newly onboarded environments were visible in the data days before anyone noticed.

The Mirth licensing position changed in March 2025, when NextGen Healthcare moved Mirth Connect to a single commercial licence at version 4.6. Version 4.5.2 remains available under the Mozilla Public License 2.0 but receives no updates, security patches or bug fixes. Open Integration Engine is the maintained community fork of that codebase. Every organization running Mirth now faces a decision, and support engagements increasingly involve making and executing it rather than simply keeping the current version alive.

Most support engagements begin with an environment nobody fully understands. The original developer has left, the channels have been modified repeatedly without documentation, and the organization has become reluctant to change anything in case it breaks. This is normal rather than unusual, and it is a large part of our work. What we will not do is commit to a service level on an environment we have not assessed, because that is a promise made on someone else’s assumptions.

Onboarding is where a support contract either becomes useful or becomes a phone number nobody trusts. Ours runs as a defined sequence rather than a handover call, because assuming knowledge is what produces the first bad incident. In practice it takes two to four weeks depending on estate size and how much documentation already exists, and it is included in the retainer rather than billed as a separate implementation project.