Maintenance & infrastructure
Organise website maintenance, backups and incidents
Website and application maintenance preserves operation in a changing environment. Servers, dependencies, external services and user needs continue to evolve after delivery. A clear arrangement explains what is monitored, what is backed up and who responds when something fails.
Separate hosting, maintenance and new development
Hosting supplies the environment in which the website operates. Maintenance may cover updates, checks and corrections under an agreed scope. New functionality and redesigned journeys are another type of work. Record these boundaries so a subscription is not mistaken for unlimited intervention.
Inventory the website, database, files, scheduled work, transactional email and connected services. Name account owners and technical contacts. An apparent website fault may originate elsewhere. Without this map, diagnosis can begin with searching for credentials rather than solving the problem.
Verify backup coverage and restoration
A useful backup covers the elements needed to recover, including data, files and configuration as appropriate to the architecture. Copying website files does not necessarily include the database. Choose frequency and retention according to activity and acceptable data loss, then check the copies actually produced.
Rehearse restoration in an isolated environment. Verify that the site opens, expected files exist and a representative journey works. Retain instructions and access for repeating the operation. Estimate recovery time from these checks and dependencies rather than treating an archive download time as the complete recovery time.
Prepare updates with suitable checks
Updates can affect operating systems, application components, extensions and integrations. Their urgency varies. Review official release information and compatibility for the environment concerned. The process should allow an urgent fix while retaining essential operational checks.
Where possible, test changes outside production using representative configuration and protected data. Check critical journeys afterwards: login, enquiries, purchases or background processing. Prepare a usable backup and a rollback approach appropriate to any changes made to data.
Monitor the functions the organisation needs
An accessible homepage does not prove that email or synchronisation works. Choose meaningful signals, such as application failures, blocked jobs, available storage or delivery errors. Alerts should reach someone able to respond and include enough detail to start investigating.
Collect logs for a defined purpose, with suitable access and retention. Agree which events require immediate attention and which belong in a scheduled review. Large numbers of ignored alerts can obscure the incidents that matter, so monitoring needs an operating process rather than just a technical installation.
Document incident handling and future handover
Specify the reporting channel, required information and agreed support hours. A screenshot, time, attempted action and user role help more than a general report that the site is broken. Distinguish response time from resolution time, which depends on causes and other parties involved.
Record identified causes, interventions and checks after incidents. Improve instructions when access or an alert was missing. Keep account inventories, operating documentation, backups and dependencies current so another team can take over if required.
Your checklist
- Define included work and exclusions.
- Identify account owners and technical contacts.
- Test complete recovery in an isolated environment.
- Check critical journeys after updates.
- Agree reporting channels, support hours and handover.
Frequently asked questions
Doesn’t the hosting provider already make backups?
It may, but check coverage, frequency, retention and restoration conditions. Confirm that the required files and data are included and that the recovery procedure has been tested.
Does maintenance guarantee no downtime?
No. It organises prevention, detection and response within a defined scope. Any availability or timing commitments need to be explicit and consistent with the infrastructure and external services involved.
A project to explore
Pare-Brise Éco
The Pare-Brise Éco page describes infrastructure management alongside websites and applications. It connects technical operation to real workflows and several components beyond the public website.
Explore the project