The cooling alarm came through at 2:40am: one computer room air conditioning (CRAC) unit in Hall 2 had tripped and the hall was running on its redundant units. The shift engineer logged it, rang the manufacturer’s after-hours line and was told a technician would call back. At 7am the facilities manager asked the obvious questions. Had anyone confirmed? Was the tech inducted for that hall? Was a change window booked? The answers sat in a call log, a text thread and one person’s memory.
Data centres already know when something breaks. Monitoring and DCIM (data centre infrastructure management) tools flag the fault within seconds. What they do not do is get the right specialist through the door inside the window, on the right induction, and leave proof behind that a customer or auditor can read later.
OpsChase is that layer: every hall and support building on one queue, the regular specialist for each system offered the job first, every attendance followed up until photos and sign-off, and facilities seeing every open job without opening the monitoring stack for a door fault.
From an alarm to a work order
The alarm stays in your monitoring tools. When a person needs to attend, the shift engineer raises a work order against the site, hall and asset: CRAC or computer room air handler (CRAH) unit, chiller, generator set, uninterruptible power supply (UPS), switchboard, fire system, dock door or perimeter security. Attach the alarm screenshot or fault code and the redundancy state, so the contractor knows whether the hall is running on its standby units or has lost cooling outright.
Add the access note on the job: induction required, escort needed, which dock and which door, the approved change window and who on shift will sign the crew in. The crew should not learn the access rules at the gate.
Choose who goes first per system, not per site
A general electrical crew is not the UPS manufacturer, and the company that services the office split systems is not the right call for a hall cooling unit. Set the order of your contractors per system and per hall: the manufacturer or specialist first for critical plant. The order can differ between halls on the same campus.
“On the way” is not a status
A customer asking about an incident does not want to hear that a vendor was on the way. On one queue the status moves through confirmed, attending, attended and proof attached, each with a time against it. Facilities, the shift engineer and the account team see the same job. When night shift hands over, the morning shift reads the job instead of a handover note.
The jobs that still need a person on site
- Cooling. CRAC and CRAH faults, chiller alarms, condenser and cooling tower work go to heating and cooling with the hall, unit ID and redundancy state on the job.
- Power. Switchboard faults, UPS alarms, thermal scans and testing go to electrical; generator defects and load bank tests go to generators with the set ID and fault code.
- Fire. Detection faults, suppression service and extinguisher checks go to fire protection, with the hall and zone named.
- Doors and security. Dock doors and roller shutters on support buildings go to doors and shutters; perimeter, access and camera faults go to security.
- Support buildings. Amenities leaks, roof leaks above plant rooms and pest calls go to plumbing, roofing and pest control on the same queue.
Attend-by times that respect the change window
Every job carries an attend-by time that fits the risk: within hours for a cooling unit when redundancy is reduced, inside the next approved change window for planned work, or by the end of the week for a support-building fault. Late jobs show against your response targets, so the facilities manager sees which hall is still running on reduced redundancy and why.
Contractor compliance before the change window
Keep licence, public liability, induction and site-access details on file for each contractor through contractor compliance, so facilities and the shift team can check them before booking. Attach the safe work method statement for high-risk electrical or work-at-height jobs to the work order. A crew turned away at security for a missing induction is a lost window, not a small delay.
Proof a customer audit can read
After an incident, customers and auditors ask who attended, when, what they did and what they left behind. Set proof templates per system — before and after photos, the vendor’s service report or test certificate, load bank results for generators, and readings for cooling work. Proof of completion lives on the work order, not in a vendor PDF sitting in a shift engineer’s inbox.
Planned maintenance and asset history
Generator runs and load bank tests, UPS battery checks, CRAC filter and belt service, fire system inspections and thermal scans repeat across every hall. With Assets & planned maintenance switched on, run planned maintenance through scheduling on the same queue as breakdowns, and each unit keeps its own job and service history. Group visits into approved windows so the specialist makes one inducted trip instead of three.
What a multi-site operator sees on one queue
With every hall on one queue:
- Alarms that need a person become work orders with the hall, asset and redundancy state.
- Specialist books stay in calling order per system and per hall.
- Shift engineers raise; facilities assigns the trade and regular contractor.
- Attend-by times follow risk and change windows.
- Reports, certificates and photos are attached to the work order.
OpsChase sits beside your monitoring, DCIM and change management tools. Telemetry and change approvals stay there; OpsChase is the contractor follow-up and proof behind each job a person has to attend. If today’s process is a spreadsheet per site and a shared inbox, see how that compares with spreadsheet and email follow-up.
Starting with one hall
Pick the site with the most open vendor jobs. Import its halls and support buildings, the manufacturer and specialist book, and the systems each contractor covers. Set roles for shift engineers and facilities. Set proof templates for cooling, power, generators and fire. Raise the next job that needs a person into OpsChase instead of another phone call, then add the next hall.
Read the data centres industry page or book a demo with your site list and specialist books. Check pricing when you are ready to put every hall on one flow.
Uptime is what a data centre sells. Put every attendance on one queue, get the right specialist inducted and through the door inside the window, and let the next shift start knowing exactly which jobs are still open.