It is 7:50am. The first patient is at 8:15. Someone double-clicks the practice management software and nothing happens — or it opens and throws an error about the server or the connection.
Before you ring anyone, there is one question that changes everything that follows, and most people have never been asked it.
First: where does your software actually run?
Australian practices run their practice management system in one of three ways, and the fault-finding is completely different for each.
On your own server, in your practice. A physical box in a cupboard, a comms room, or under reception. Dental4Windows, Exact and Oasis are commonly set up this way. The software lives on that machine and the workstations connect to it across your network.
Hosted by the vendor. The same software, but the server sits in the vendor's data centre and you connect to it over the internet, often through a remote desktop session. Your "server" is not in the building.
Cloud, in a browser. Praktika, Core Practice and the cloud versions of the major systems. There is no server of yours at all — staff open a web address and log in.
If you are not certain which you have, that in itself is worth resolving before something breaks. Ask whoever supports you, and write the answer down where reception can find it.
If your server is in the practice
This covers most Dental4Windows, Exact and Oasis sites. Work through these in order — the first three take about ninety seconds and solve most of them.
1. Is it one computer or all of them?
Ask someone to try a different machine. It is the single most useful thing you can establish before ringing anyone.
- One computer only — the problem is that machine or its connection. The practice can keep working from the others.
- Every computer — the problem is the server or the network. Nobody works until it is fixed.
2. Is the server actually on?
Look for a power light, any amber or red warning lights instead of green, and whether you can hear the fans. A completely silent server is usually a dead one.
Servers do get switched off by accident — cleaners, tradespeople, a tripped powerboard, a UPS that quietly ran out of battery. It is not a stupid thing to check.
3. Did Windows restart overnight?
Windows installs updates and reboots on its own schedule unless someone has deliberately controlled it. If the server rebooted at 3am and the database service did not come back cleanly, the software will refuse to connect even though the machine looks perfectly healthy.
This is the one that catches practices out most often, and it is entirely preventable. Patching should run in a window you choose, not Microsoft's default.
4. Can the workstations see the server at all?
If the software cannot reach the server, the error often mentions a database or a connection rather than anything obviously network-related.
Quick check with no technical knowledge: on a workstation, open File Explorer and try to open a shared folder that lives on the server. If that fails too, it is a network or server problem rather than a software problem — useful to know before you call.
If your software is hosted or in the cloud
Different problem entirely. There is no server of yours to check, and you cannot fix the application — the vendor holds the administrative access.
What you can establish in two minutes:
- Is the internet working at all? Open any other website. If nothing loads, it is your connection, not the software.
- Is it one computer or all of them? Same question, same value.
- Can you reach it from a phone on mobile data? If the software loads on 4G but not in the practice, the problem is inside your building — your internet service, your firewall, or your Wi-Fi. If it fails on mobile data too, the problem is at the vendor's end.
That last check is the useful one, and almost nobody thinks to do it. It tells you within a minute whether to ring your IT provider or your software vendor.
What local IT still looks after on a cloud system
There is a common assumption that cloud software means there is nothing for an IT provider to do. In practice the vendor is responsible for the application and its data — and everything else in the room is still yours:
- The internet service and a backup path for when it drops
- The firewall, router and Wi-Fi, including which devices are allowed on
- The computers themselves — Windows, updates, encryption, antivirus
- Imaging sensors and scanners, which are physically attached to your machines regardless of where the software runs
- HICAPS and Tyro terminals and their network connection
- Printers and scanners
- Who has access to what, and removing it the day someone leaves
- Your own copy of the logins — so you are not locked out of your own practice if a relationship ends
Cloud removes the server. It does not remove the practice.
The awkward one: vendor admin access
On hosted and cloud systems the vendor usually holds administrative rights, sometimes including remote access into your machines to support their software.
That is normal and often necessary. But it is worth knowing:
- Who at the vendor can access your system, and whether that access is logged
- Whether they can reach the whole computer or only their own application
- What happens to that access if you change providers
You are the one responsible for patient information under the Australian Privacy Principles, whoever is holding the keys. Asking the question is reasonable, and a good vendor will answer it plainly.
So who do you ring?
The honest answer is that most practices ring the software vendor first, wait in a queue, and are told forty minutes later that it is a network problem.
A quicker rule:
| What you found | Who to ring |
|---|---|
| One computer affected, others fine | IT |
| No computers can reach anything online | IT |
| Works on mobile data, not in the practice | IT |
| Fails on mobile data too, cloud system | Vendor |
| Server is off, or showing warning lights | IT |
| Software opens but a specific feature is broken | Vendor |
What to have ready when you call
Whoever you ring, these four answers cut the time to a fix considerably:
- On-premise, hosted, or cloud?
- One computer or all of them?
- The exact wording of the error, photographed on your phone
- Anything that changed since it last worked — a new machine, a power cut, a moved cable
The bigger point
Most of these mornings are not software failures. They are a scheduled reboot, an unnoticed power event, a connection that has been degrading for a fortnight, or a change nobody wrote down. All of them are the sort of thing monitoring catches before a practice does — which is the unglamorous half of managed IT support: watching the boring signals so the interesting failures never happen.
If you have had more than one of these mornings, the problem is not the software. It is that nothing is watching between the times somebody happens to look.
We cover the hours most providers do not — evenings until 9pm, Saturdays, and a Sunday emergency line from 9am to 5pm — because that is when practices discover things are broken. The rest of what that covers is on the dental practice IT page.
If something is down right now, call 03 7057 1878. Work through the checks above first if you can; it will make the call shorter.
If nothing is on fire and you would rather know where you stand before the next one, the free IT health check is exactly that — a plain-English look at your backups, your server and who answers when your provider has gone home. No obligation attached to it.