Dental Office Network Support in San Antonio: When the Server Stays Dark After an Outage, a Sensor Hangs in One Operatory, and the Closet Hits 100 Degrees

Monday, 7:50 a.m., on the north side of San Antonio. CPS Energy had the block dark for forty minutes on Saturday night. The lights are back, the fridge is humming, and the first hygiene patient is already in operatory 2. The front desk opens the practice-management software and gets "cannot connect to the database". In operatory 5 the sensor says "device not found", and operatories 1 through 4 can at least see the schedule. Someone has already unplugged the modem twice.

This page is written for the owner who is standing in that hallway, and for the office manager reading it with them. It gives you a table that reads the symptom for you, a vendor-sourced view of how much network your imaging really needs, a power-down and power-up order for the next CPS Energy outage, and a set of questions that reveal whether whoever supports you now has ever opened your closet door. It stays on dental networks: operatories, sensors, CBCT, the practice-management server, the sterilization-area devices. If you are looking for general office networking, this is not that page.

Is it the internet, the Wi-Fi, the switch or the server? A front-desk triage for San Antonio dental offices

You can tell which of the four layers failed before anyone arrives by asking three things: does the practice-management server respond from any workstation, does anything cloud-based work, and is the problem in one operatory or all of them. Together they point at one layer — demarc, firewall, cabling and switch, or server — and each layer has its own first fix.

The layers, in the order a signal crosses them from the outside world to a sensor:

• • The demarc and the carrier's equipment. The ONT or modem where the internet enters your suite. If it loses power or the line outside is damaged, everything cloud-based dies and everything local lives.
• • The firewall. The first box that belongs to you. It decides what gets in and out and, in a well-built office, which internal networks may talk to each other.
• • The switch and the drops. Operatory PCs, the server, desk phones and Wi-Fi access points all reach the switch over cable runs terminated on a patch panel. One room failing alone usually lives here, or in the room itself.
• • The server. The practice-management database and, in most San Antonio offices, the imaging database. If it is down, every workstation shows the same error and the internet is irrelevant.

Then the table. Run down the left column until you find your Monday.

Three things the table teaches. First, the symptom that costs the most chair time in a San Antonio office is the one-room fault, because it looks like a sensor problem, gets handled as a sensor problem, and lives on for months inside a wall. Second, "the internet is down" and "the server is down" are two different layers with two different first calls, and a provider who answers both by asking you to unplug the modem is not diagnosing anything. Third, two of the rows above are not network faults at all: a sensor on a tired USB extender and an unowned remote-access tool. The table exists so you stop paying for network work when the problem is a cable behind the monitor, and stop rebooting when the problem is a security event.

Practices wiring a new suite instead of repairing an old one will find the installation checklist in the dental office network setup guide; this page does not repeat it.

What should a dental office network in San Antonio look like?

A San Antonio dental office of four to ten operatories should have one managed switch in a cooled, lockable closet, a Cat6 or better drop to every operatory and to the lab and sterilization areas, a firewall that separates clinical, voice, staff and guest networks, and a UPS that carries the carrier's equipment as well as the server. Everything else is detail.

That sentence is deliberately short, because the detail is where offices go wrong, so here is the detail.

Drops, not extenders. Each operatory needs at least one tested network run from the closet to the wall, terminated on a labeled patch panel. A second drop per room is the difference between adding a second monitor PC or a networked device cleanly and adding a small switch under the counter. Three of the symptom-table rows trace back to that under-counter switch.

One switch, in a closet that was built to hold it. A chain of small switches, each added when a room was finished, is the most common thing an engineer finds behind "the office is slow". Replace the chain with a single managed switch that has a port for each drop plus the server, desk phones and Wi-Fi access points, and power over Ethernet wherever a phone or access point draws from it. It also gives monitoring something to watch: port speed, errors, temperature.

The server, sized for the heaviest thing it does. In most San Antonio practices the practice-management server is also the imaging server. Henry Schein One's Dentrix G7.9 System Requirements put the network line at "1 Gbps", a figure that applies to servers and workstations alike, and the same document says that the server's "disk space, memory, and networking throughput requirements should be evaluated and optimized to ensure that the overall response to the workstations is acceptable". The dental office server and workstation setup service owns the server side; this page only insists that the server's network card and the switch port it lands on both run at 1 Gb, because a 100 Mb link anywhere on that path slows every room at once.

Four networks, one switch. Clinical devices on one network, desk phones on a voice network with a priority rule, staff tablets and laptops on a third, and guests on a fourth that is throttled and cannot reach a single clinical device. This is a firewall and switch configuration, not four sets of cable.

A UPS that includes the carrier's box. The device most often left on a bare outlet in a San Antonio closet is the carrier's ONT or modem. When power blinks and the lights stay on, that box is the one that reboots.

Documentation you could hand to a stranger. A labeled patch panel, a photo of the closet, the carrier's circuit number and support line taped inside the door, and test results for every run. When an engineer is working with you remotely on a Monday morning, that photo and that label are the difference between five minutes and an hour.

The components of a network project, and what a buildout includes, are covered on the dental office network design page, the owner of that subject. This article does not restate them and claims nothing about what a given project contains.

Why do imaging and phones fight on the same network in a dental office, and how is it fixed?

Imaging and phones fight because a CBCT export or a backup is a large, patient transfer that will take every bit of a link it is given, and a phone call is a small, impatient stream that breaks the moment it has to wait. On one shared uplink with no rule between them, the call loses. The fix is separation and priority, not a faster circuit.

Put differently: the problem is not that your internet is too slow. It is that nobody told the switch which traffic matters more. A voice VLAN with a priority rule, and clinical devices on their own network, lets the two coexist on the same switch. The VoIP phone systems for dentists page covers the phone side; this page stays on the network the phones ride on.

The same logic applies inside imaging. Not all modalities load the network the same way, and the honest way to put them in proportion is to use what the vendors publish rather than invent a number. Three published pages give the proportions below.

Two rows in that table carry a lesson for an orthodontic, endodontic, oral surgery or pediatric office, where the imaging stack is larger: the CBCT row and the scanner row. If your practice is one of those, IT for specialty dental practices in San Antonio works through imaging one practice type at a time. For the software and imaging side of the same problem in this market, the dental software and imaging support in San Antonio page is the owner, and its own framing is the right one: many software problems in San Antonio are cabling problems. This article is the network half of that sentence.

When CPS Energy cuts power to a dental office, what happens to the network and what should come back first?

When CPS Energy drops power to a dental office, the closet shuts down in whatever order the batteries decide, and when power returns the devices boot in whatever order they were plugged in. The practice-management server often comes up before the switch and firewall are ready, and the carrier's box may not re-register at all. A written power-down and power-up order prevents the second outage.

San Antonio's record is the reason to write that order down rather than hope. CPS Energy serves "more than 996,000 electric and 399,000 natural gas customers in San Antonio and portions of seven adjoining counties". During the February 2021 winter storm, KSAT reported that "Roughly 251,606 customers" were eligible for a credit for having been without power for longer than 24 hours. In February 2022, Texas Public Radio reported that "As many as 98,000 CPS Energy customers experienced some kind of power interruption throughout the day" from wind and ice. And in August 2023 the city "recorded its 60th 100-degree day of the year", after a stretch of "23 days in a row" at or above 100 degrees. Winter outages, ice, and summers that cook a closet: three different events, one closet.

The order below is general good practice for a small server closet, written for a dental office. It is not any one provider's procedure, and your provider should confirm it against your equipment before you tape it to the closet door. Imaging units such as a CBCT or panoramic machine follow their manufacturer's shutdown instructions, which this page does not restate.

Two patterns in the table deserve a sentence each. The server goes down first and comes up fourth: first because it is the device that an abrupt power loss damages, fourth because a server that boots before the switch is ready may decide the network does not exist and sulk for the rest of the morning. The carrier's box goes down last and comes up first for the opposite reason: it is cheap to keep alive and slow to re-register, so everything upstream waits on it.

Power-event checklist for a San Antonio dental office

Before the event, which in this city means before the first conservation call of summer and before the first freeze warning:

1. 1. List what is on the UPS and what is not, with the runtime the UPS reports under the closet's real load. "Everything" is a common answer and a wrong one.
2. 1. Put the carrier's equipment on the UPS if it is not there already, and the phone gateway if you have one on site.
3. 1. Configure the server to shut itself down cleanly when the UPS reaches a set runtime, and test it once on a Saturday.
4. 1. Tape the power-up order inside the closet door, with the carrier's circuit number and support line, and a photo of the closet running normally.
5. 1. Decide who does what if the office is dark for a day. Where printed schedules for the coming two days and a list of patient phone numbers will be kept in case the server is down, who calls patients, and from which phone.
6. 1. Know what your failover path protects, if you have one. When the carrier circuit fails, a cellular failover keeps cloud practice-management, e-claims and VoIP running. It does nothing for a closet with no power or a server that was shut down.
7. 1. Ask the question in writing: which of my systems keep working if the circuit drops, and which keep working if the power drops. The two lists are different.

After the event:

1. 1. Bring the closet up in the order above, with a pause between each device, not all at once.
2. 1. Check the server's own screen before the workstations. A disk check or a login prompt waiting there explains every "cannot connect" error in the office.
3. 1. Test one operatory fully — sensor, chart, imaging, a phone call — before opening the rest of the day.
4. 1. Confirm the backup ran. The row at the bottom of the symptom table is the one owners find out about a month later.
5. 1. Write down what failed and why, and have the closet walked through before the next event. The second outage is the one you had a chance to prevent.

The wider continuity picture, including what is backed up and how recovery is tested, belongs to the dental backup and disaster recovery service page and, for San Antonio practices, to our San Antonio dental IT guide. This page's job ends at the closet door.

Older buildings, summer heat and closets: when the fix for a dental office network is new cabling or a new switch, not a reboot

When the same operatory drops every few weeks, or the whole office drops on hot afternoons, the fault is physical, and the hundredth reboot will not change it. Older San Antonio suites were often wired for a different tenant, and a closet chosen for convenience rather than cooling fails in a summer like 2023's. The fix is a cabling project or a new switch, quoted as a project.

Three physical causes account for most of it.

Wiring pulled for a phone system. A run that was installed for an earlier tenant's phones may be the wrong category of cable for a gigabit link, or it may have been spliced, stapled or run beside a power line on its way to the operatory. The jack in the wall looks like a network jack, which is how a practice can be "already wired" and still have a room that links at 100 Mb. The only way to know is a test, run by run, with the result kept. If your provider has never tested the runs, the honest answer to "is it the cabling?" is "nobody knows".

A closet chosen for the floor plan, not for the equipment. In many older suites the switch lives in a storage room with a door that stays closed, under a shelf of supplies, next to a water heater, or in a corner of the sterilization area. On a 100-degree afternoon in a building whose air handling was designed decades ago, that room gets hotter than the equipment is rated for, and the symptom is a drop that fixes itself when the room cools after close. The monitoring dashboard should show the closet temperature, and the closet should have ventilation or cooling of its own, or the equipment should move.

Rooms added one at a time. An office that expanded from four operatories to eight across fifteen years tends to have a network that expanded the same piecemeal way: a small switch here, a run daisy-chained from the next room there. Each addition worked on the day. The result is a chain in which any link can take down the rooms behind it. Collapsing the chain into one managed switch with a home-run to every room is the most common project an engineer proposes in an older San Antonio building, and the one owners most often wish they had done a year earlier.

The dental software and imaging support in San Antonio page frames the same reality from the software side and is the owner for older-building software problems. On the network side, the dental office network design page owns what a project contains. What this page adds is the diagnosis: if the symptom is physical and periodic, ask for a test and a project quote, not another reboot. If you are instead opening a new office and get to choose the closet before the walls close, the companion piece on opening a dental practice in San Antonio is where to begin.

Which devices should never share a dental office's clinical network, and what should the firewall be doing?

The clinical network should carry acquisition PCs, the server, imaging workstations and the front desk, and nothing else. Phones, staff laptops, patient tablets, guest devices, cameras, smart TVs, door access and anything a vendor installed to call home all belong on separate networks behind the firewall. The firewall's job is to keep them apart and to control what reaches the clinical side from outside.

Segmentation is the cheapest security control a dental office has, because it turns a dozen different devices into four groups with four sets of rules. A waiting-room TV that needs the internet does not need the imaging server. A parent's phone on the guest network does not need the practice-management database. A camera system installed by a different contractor should be able to upload its recordings and nothing more.

The last row in the symptom table is the one to read twice. A remote-access tool is how most support is delivered to a dental office, including by your own provider, and that is exactly why every such tool on the clinical network needs a named owner, a reason to exist, and a second factor on its login. The question to ask is simple: list every remote-access tool installed on my computers, who owns each one, and what happens if the owner's credentials leak. A tool nobody can account for comes off the network the same day. In 2026 an owner in this market is right to ask that question.

The remainder of the security picture — endpoint protection, email, backup verification, access controls — belongs to the dental cybersecurity services page. For the Texas-specific checklist, including the state's privacy-training and breach-notice rules, use the Texas dental cybersecurity checklist for 2026; this article leaves the law to that page.

Where has Legend documented network work for San Antonio dental practices?

One Legend Networking project record is public for San Antonio itself; here is what it says. The figures Legend publishes about itself: 22+ years in dental IT (since 2004), 3,500+ practices supported, and a 4.8 rating · 173 Google reviews.

Dated 16 June 2026, the Ray Orthodontics project record states that Legend Networking supported the technology setup for Dr. Andrew Ray's new orthodontic office on Potranco Road in San Antonio, coordinating the systems needed for opening day. That is the scope the record gives, and this page does not add a drop count, a switch model or a closet description to it.

At the Texas level, the project record for Clearview Dental, a practice taking shape in Austin, lists the decisions a practice makes during construction, among them cloud or on-premise systems and structured cabling. It is named here as an Austin, Texas project, not a San Antonio one, and only for the point it makes: cabling is a decision made before the walls close, which is why older buildings carry the problems in the section above.

On the people side, the named person is Ian Lynch — 22 years in dental IT; oversees the NOC, onsite dispatch and client technology roadmaps. Legend also states: In-house engineers; no outsourced first-level triage. Put both to any provider you are weighing: which person is responsible for the monitoring that would catch a switch port dropping to 100 Mb, and who picks up once the first technician is stuck.

Two more published facts matter to a San Antonio owner, and both are stated in the registry's own words. Monitoring and staffed hours are separate things: 24/7 monitoring, and Help desk Mon–Fri 7 AM–7 PM CT, Saturday 8 AM–2 PM CT; 24/7 critical/emergency line. San Antonio is on Central Time, so those hours need no conversion; they are your hours. For pricing, Legend's published figure is Flat monthly plans $399.99–$1,200 per practice per month; it also says Hardware procurement, major cabling projects and software licensing are quoted separately. Taken together, re-pulling six operatories and replacing a switch chain counts as a cabling project plus hardware procurement, priced on its own, and any provider who folds that into "included" should be asked to put the scope in writing.

Legend Networking's headquarters is in Dallas, Texas, and it describes its model as Onsite dispatch by appointment across served dental markets; remote-first nationwide. Before signing, find out what that means for a San Antonio office specifically, and have it put on paper with project work and monthly work kept apart. For the commercial side of that conversation, the dental technology support in San Antonio page is where Legend states what it offers in this city.

What to ask your IT provider about your dental office network

Ask these in writing, and keep the answers with the closet photo. The answers show whether anyone has looked inside your closet or only at your ticket history.

A provider who can answer all thirteen from monitoring and documentation, without a visit, is worth keeping. If they answer fewer than half, your network is being rebooted, not supported.