GLPI vs iTop

The running cost of GLPI and iTop is set less by the license line than by who maintains the instance, how many agents and assets it covers, and which support tier you buy. Both products are open source, so the software itself carries no license fee. What you pay for is the subscription or support contract from the vendor, the server capacity, and the staff hours spent on upgrades, plugin upkeep, backups and agent rollout. For GLPI, the vendor path is Teclib' GLPI Network, where the public source lists self-hosted subscriptions from 100 euros per month (Basic, up to 10 agents and 500 assets) to 4,500 euros per month (Enterprise), per the GLPI Network subscriptions page. For iTop, the vendor path is Combodo's agent-tiered editions (Community up to 10 agents, Atelier up to 50, Forge above 50), plus commercial support, training and consultancy. Ask each vendor to price your actual agent and asset counts, not the tier label alone.

One-time costs come first: initial configuration, the CMDB data load, and integrations with directory, monitoring and mail systems. These are usually the largest spend in the first year and are the hardest to estimate from a product page. Recurring costs come from subscriptions, support tiers, agent growth and upgrade work. GLPI adds a specific upgrade cost: its update procedure requires backing up the database, files and the config directory including glpi.key and glpicrypt.key, and since GLPI 11.0 plugins are suspended after a new version is detected and must be reactivated after major updates. Budget staff time for that cycle and for re-testing plugins. On the iTop side, the recurring load is custom data model changes, extension upkeep from the iTop Hub, and keeping MariaDB and PHP versions within the supported ranges. Server sizing also drives cost. Combodo suggests an all-in-one iTop server with 2 vCPU and 4 GB of memory for fewer than 200 tickets a month and under 20 console users, and separate servers of up to 8 vCPU and 16 GB for larger installations, per the iTop software requirements.

Cost rises when agent coverage grows faster than planned, when plugins that exist only in a paid tier are needed, or when the team must build customizations that an upstream release later breaks. It falls when inventory is handled by the native GLPI Agent rather than a separate tool, and when the environment scope is fixed early. Ask every bidder for a written quote that states: the agent and user counts now and at the planned peak; the asset count and the number of CMDB classes in scope; the support level and response expectations; the environments covered (production, staging, disaster recovery); whether backups, upgrades and plugin maintenance are included or billed separately; the database and server sizing assumed; and which items are one-time versus annual. Compare totals over three years, not the first invoice.

GLPI projects tend to stall in the web and plugin layer, not in the ticket configuration. The web server document root must point to GLPI's public directory rather than the installation root, with URL rewriting routing requests through index.php on Apache, Nginx, lighttpd or IIS. Teams that expose the installation root by mistake lose time on confusing errors. Next, check the PHP and database versions against the right source for your release. The GLPI README on the main branch lists PHP 8.3 or later with MariaDB 10.11 or later or MySQL 8.0 or later, while the installation docs list PHP 8.2 or later with MySQL 8.0 or MariaDB 10.6 for GLPI 11.0 and say only MySQL and MariaDB are supported. Confirm which set applies to the release you are deploying before building the server image. Plan for the update path too: back up the database, files and the config directory including glpi.key and glpicrypt.key, and expect to reactivate plugins after major updates, since GLPI 11.0 suspends them when a new version is detected.

GLPI inventory is a second source of delay. The GLPI Agent collects hardware and software inventory and runs on Windows, macOS and Linux, but rolling it out to thousands of endpoints is a distribution and scheduling exercise that your endpoint management team must own. Network discovery and SNMP inventory need the separate GLPI Inventory plugin, so scope that plugin as its own work item with its own test network and credentials. Confirm your rollout method and the agent's target configuration against the GLPI Agent documentation before committing to a date.

iTop projects usually run long at the data model stage, before any data is loaded. iTop's CMDB is fully customizable, so the classes, relationships and lifecycle states you define become the basis for change impact analysis and SLA reporting. Changing them after data arrives means migrating records, which is where most schedule overruns start. On the infrastructure side, check the iTop software requirements for the 3.3.x PHP, MariaDB and GraphViz needs, and size the server against the published guidance for your ticket volume and console user count. GraphViz is easy to forget and blocks impact diagrams until it is installed. Build change and SLA workflows only after the data model is signed off, since approval routing depends on those classes.