Role map
What each machine does: domain controller, files, ERP database, web server, printing. And for each, the date its operating system drops out of support.
Picture a dental clinic in Seville with a server in the reception cupboard. It holds the patient management database, the X-ray images, the shared folders and a couple of services somebody installed years ago. Nobody knows exactly what each one does, but everybody notices when it fails. Server management starts by writing that list and carries on with less glamorous work: regular patching, watching disks and databases, checking backups and recording every change. It makes no difference if the hardware is on your premises, with Arsys, IONOS or OVHcloud, or in Azure's Spanish region: the approach is identical and done entirely from a distance.
We do not begin with tools but with the services your business relies on. Alert thresholds and the order of work follow from that.
What each machine does: domain controller, files, ERP database, web server, printing. And for each, the date its operating system drops out of support.
Disk space, load, stopped services, recurring log errors. Every alert has a recipient and an action; any alert that leads to no action is removed.
Windows in the agreed window, Linux with extra care over PHP and the database behind your PrestaShop or WooCommerce store. Always with a snapshot or backup beforehand.
SQL Server under a3ERP or Sage 200, MySQL under the website: transaction log growth, native database backups and indexes that degrade over time.
Drive and RAID status, PSUs and heat readings pulled from iDRAC, iLO or the host's dashboard. When a drive shows early signs of trouble, we open the warranty case while it still works.
Every change recorded with its author, date and reason. It makes an audit against ISO 27001 principles or NIS2 preparation bearable, and lets you switch provider without losing your history.
Early on we mostly listen and map. Once that is done, the schedule is predictable and shared, so your team knows when we will be working.
Personal admin logins for each engineer, reached through a two-factor VPN or similar secure tunnel. Legacy passwords are rotated on the very first day.
Disks, outstanding patches, backups, logs, unsupported versions. Anything urgent is dealt with immediately; everything else is scheduled.
A ranked list with dates: retire an unsupported system, split roles apart, close ports, tidy up permissions.
Maintenance slots that never move, a written monthly summary and a short call with the person who signs off IT spending.
The most dangerous server is the one nobody knows about. An old PC under a desk acting as the clock-in server, a test virtual machine left running with remote desktop exposed to the internet, a NAS still on its factory password. The first review sweeps the network for exactly these, because they are the usual entry point for attacks that end in ransomware.
Yes, and it is among the commonest issues. An outage without a UPS can corrupt the ERP database mid-write. We specify the UPS you need and how to wire it so the server shuts down cleanly; the physical installation is done by your supplier or a colleague following our guide. If cuts are frequent, a data centre is worth considering.
Every month it remains live, the exposure grows, and insurers and auditors will ask about it. We schedule a move of its workloads to a maintained release or a fresh virtual machine. Until then it is fenced off on the network with no access from outside.
The window is agreed with you: it might be Tuesday at 22:00 or Saturday morning. Sensitive dates are avoided, such as month-end close, income tax season or quarterly tax filings. A critical patch for a flaw under active attack, flagged by INCIBE-CERT, does not wait for the window; we offer you a slot the same day.
If the servers are in your office, yes. If you prefer a data centre, we work with providers that have facilities in Spain and the EU, such as Arsys, IONOS or OVHcloud, or with the Spanish regions of Azure and AWS. The choice is documented for your record of processing activities.
You are. The credential vault is one you hold keys to, every admin login is tied to a person, and all documentation is your property. Moving to another provider later means a straightforward handover, not a negotiation.
List the servers you operate and where they are hosted. We will come back with a few questions and a rough monthly figure.
Your request is with us
Expect an answer within one working day. A reported fault that has halted your team is handled first.
No match found. Try another spelling, or go with the closest provincial capital: every job is done remotely, so the location makes no difference to what we deliver anywhere in Spain.