Open-source ITSM tools compared

GLPI, iTop and Znuny compared as self-hosted ServiceNow alternatives, with Snipe-IT for asset tracking alone. Editions, CMDB models, integrations and operating effort.

On this page

Which open-source ITSM tool fits your team's ServiceNow replacement

If your team is leaving or avoiding ServiceNow and needs ITIL processes, a CMDB and asset records on its own infrastructure, the first question is whether you want one suite that covers assets and service management, a CMDB-first platform, a helpdesk-centered ITSM continuation, or an asset register alone. In short: choose GLPI for assets and service management in one suite, iTop when the CMDB and impact analysis drive the design, Znuny when you already run OTRS or want a helpdesk-led ITSM tool with long-term support lines, and Snipe-IT only when you need asset tracking and nothing more. None of these is a ServiceNow clone. Each models configuration items, services and workflows differently, so the decision rests on your data model and on how many people you can staff to build and run it.

GLPI covers tickets, problems, changes, assets, contracts and financial management in one GPL-3.0-or-later platform maintained by Teclib'. Its native inventory comes from the GLPI Agent, while network discovery and SNMP inventory need the separate GLPI Inventory plugin, so confirm that plugin's version against your GLPI release before you commit. iTop is AGPL-3.0 and organized around a customizable CMDB, change management and graphical impact analysis. Its free Community edition is positioned for up to 10 agents, with Atelier and Forge tiers above that, so count agents early because tier limits shape both licensing and the staff needed to maintain the model. Znuny suits teams with OTRS history, since it continues OTRS Community Edition 6.0.x under GPL-3.0 with feature and LTS release lines. Snipe-IT is useful when the CMDB need is small, but it will not replace ITIL workflows.

Before you choose, have your team answer three questions: how many configuration item classes and relationships you must model, who will own the data model after go-live, and whether your team can run a PHP or Perl application stack with database upgrades on a fixed schedule. Ask each vendor or implementation partner for the paid support scope, the edition limits that apply to your user and agent counts, and a written migration plan from ServiceNow exports. Compare the answers against the GLPI documentation for assets and the iTop software requirements before you shortlist. A pilot with real CMDB data will show you more than any feature table.

Which ITSM tool fits which scenario

This table matches common ITSM team profiles to the tool most likely to fit, with the supplied fact behind each choice. Read it as a starting filter: confirm edition, support and integration needs against your own environment before shortlisting.

ScenarioBest fitWhy
Asset-first team that needs automated hardware and software inventoryGLPIGLPI has native inventory through the GLPI Agent since version 10, and the agent also covers network discovery and package deployment. Confirm whether network discovery and SNMP need the separate GLPI Inventory plugin in your setup.
CMDB-first team that relies on impact analysis across configuration itemsiTopiTop describes a customizable CMDB with graphical impact analysis, data consistency audits and data synchronization for federation. Confirm that its CMDB model fits your relationships before committing.
Helpdesk-led team that wants a Zammad-style front end for ticket handlingZnuny for ITSM-style ticketing, Zammad for a helpdesk-first front endZnuny describes itself as a ticketing system for customer service, help desk and IT service management, and Zammad is an AGPL-3.0 helpdesk with its own PostgreSQL 15 and Redis 7 prerequisites. Compare the front-end experience and asset depth separately.
Team that needs only asset tracking for hardware, licenses and assignmentsSnipe-IT (illustrative for asset-only scope)Snipe-IT is covered here for asset tracking alone, without ITIL process or CMDB depth. If you also need incidents, changes and problems, choose a full ITSM suite instead.
Team that wants a vendor-backed support tier with a paid optioniTop (Atelier or Forge) or GLPI (Network subscriptions)Combodo sells commercial support and consultancy for iTop, and Teclib' sells GLPI Network subscriptions with level 3 support on the Enterprise tier. Ask each vendor for written support scope, response terms and what a self-hosted plan covers.
Team that wants a long-term support line for an existing OTRS deploymentZnunyZnuny continues the OTRS Community Edition 6.0.30 line under GPL v3 and publishes an LTS release line alongside its feature line. Check the LTS support window with Znuny GmbH before relying on it.

Each ITSM option in brief, with its license

Four options sit on this shortlist, each with a different starting point. GLPI, iTop and Znuny each cover IT service management in some form, while Snipe-IT covers asset tracking alone. GPL and AGPL are open-source licenses recognized by the OSI, and any paid tier is a separate commercial offering layered on top.

GLPI: ITSM with native inventory

GLPI is GPL-3.0-or-later and maintained by Teclib', which also sells GLPI Network subscriptions and cloud hosting. The current major line is 12.0, published October 7, 2026, with 11.0.11 and 10.0.28 still maintained. Its main strength is tickets, problems and changes sitting next to asset and inventory data, with hardware and software inventory from the GLPI Agent. Check early that network discovery and SNMP need the separate GLPI Inventory plugin, and plan for plugin reactivation after major updates.

iTop: CMDB and change depth

iTop is AGPL-3.0 licensed and built by Combodo, with 3.3.0 published September 15, 2026 and 3.2.3-3 published October 16, 2026. Its main strength is a customizable CMDB tied to change management, SLA management and graphical impact analysis, which suits teams that need to see what a failed change touches. Its main constraint is licensing by agent count: the free Community edition is positioned for up to 10 agents, with paid Atelier and Forge tiers above that. Confirm your agent count before you plan, and budget for the MariaDB and GraphViz requirements.

Znuny: help desk heritage with LTS

Znuny is GPL-3.0 licensed and continues OTRS Community Edition 6.0.30 after that edition reached end of life in December 2020. Current lines are the Znuny 7.3.8 feature release and Znuny LTS 6.5.26, each with separate documentation. Its main strength is a ticketing core with a long-term support line, which suits organizations that want a stable Perl-based service desk. Its main constraint is the stack: Perl 5.16.0 or higher on Linux, with Apache and mod_perl2 recommended. Confirm that its configuration item and CMDB features meet your requirements before you commit.

Snipe-IT: asset tracking only

Snipe-IT is AGPL-3.0 licensed and covered here for asset tracking alone, not as an ITSM platform. It offers free self-hosting and paid hosted plans, so the license and the commercial tier are separate decisions. Its main strength is a focused record of hardware and license assignments, which suits teams that only need to know who holds what. Its main constraint is scope: if you need ticket workflows, change control or a full CMDB, pair it with another tool or choose one of the ITSM options above. Confirm its current feature set and hosting terms directly with the project.

How the CMDB and asset data model decides the fit

The first decision is whether your team needs an asset-centric inventory or a configuration-centric CMDB, because the two shape how data enters the tool and how much of it you must maintain. GLPI is built around assets: its native inventory comes from the GLPI Agent, which has been part of GLPI since version 10 and collects hardware and software data from endpoints. Network discovery and SNMP inventory need the separate GLPI Inventory plugin, so if your estate depends on switches, printers and other network devices, add that plugin to the scope and test it on a representative segment before you commit. iTop starts from configuration items and the relationships between them, which suits teams that need to answer questions such as which business services depend on a given database server. Its graphical impact analysis is the feature to test against a real change scenario, not a demo dataset.

Migration effort depends mostly on the source data. Spreadsheets and an existing CMDB both need a field-mapping exercise before anything is imported: which fields become asset attributes, which become configuration item classes, how owners and locations are represented, and which identifiers (serial numbers, hostnames, asset tags) will be the unique keys used for matching and deduplication. iTop provides CSV import and data consistency audits, which are useful for finding orphaned records, duplicates and broken relationships after a load. Run the audit on a sample import first, agree on the rules for rejected rows, and plan who signs off on the data before it becomes the system of record. Confirm the exact import options and audit rules in the current iTop documentation and in a test instance, since these details vary by version.

Before import, decide who owns each class of data, how often it refreshes, and whether discovery or manual entry is authoritative for each field. Ask the vendor or community for the supported import formats in the release you plan to run, and budget the time for cleansing as a one-time cost that a support contract will not remove. Also check how each tool handles sync with your existing sources, since a CMDB that is populated once and never refreshed becomes stale quickly. If you are unsure which model suits your processes, a short illustrative pilot with ten to twenty representative assets and their dependencies will show more than any feature list.

Editions, support and what each one costs to run

GLPI is distributed under GPL-3.0-or-later and maintained by Teclib', which sells GLPI Network subscriptions and cloud hosting. The software itself carries no license fee, so the recurring spend depends on whether you buy a subscription and at what tier. Per the GLPI Network subscriptions page, GLPI Network public cloud costs 19 euros per user per month and private cloud 21 euros per user per month with a 25-user minimum. Self-hosted subscriptions range from 100 euros per month (Basic, up to 10 agents and 500 assets) to 4,500 euros per month (Enterprise), with level 3 support and GLPI Network marketplace plugins at the top.

  • Free self-hosting: The core product, including the GLPI Agent for inventory, carries no license fee. Your costs are hosting, database administration and staff time.
  • Per-user cloud pricing: The public cloud and private cloud figures are priced per user per month, so headcount and the 25-user minimum on private cloud drive the total.
  • Self-hosted tiers: Basic starts at 100 euros per month for up to 10 agents and 500 assets. Enterprise, at 4,500 euros per month, adds level 3 support.
  • Support level: Ask which support level each tier includes, how response is handled and whether plugin support is covered, because marketplace plugins are part of the higher tiers.
  • Quote items: Request the agent and asset counts used for tier selection, the support level, the included plugins, and the monthly total with any setup or migration charge listed separately.

Confirm with Teclib' whether a subscription includes upgrades to new major versions such as GLPI 12.0.0, and whether your inventory plugins are covered.

How to choose and pilot an open-source ITSM tool

This sequence turns a shortlist of GLPI, iTop and Znuny into a tested decision, so you commit based on your own processes, data and sign-in setup rather than on feature lists alone.

  1. Define processes and CMDB scope

    Write down which incident, problem, change and asset lifecycle processes the tool must support in the first phase, and which configuration items belong in the CMDB. Involve the service desk lead, change manager and asset owner, since each will test a different workflow later. Done means a one-page scope document that names the processes, the item classes and the teams that will use each one. Items outside this scope should be parked in a backlog so the pilot stays focused.

  2. Count agents, users and assets

    Count the staff who will work tickets as agents, the requesters who will only submit or view tickets, and the hardware and software records you expect to hold. Licensing and sizing for this comparison depend on these counts. Combodo positions iTop Community for up to 10 agents, Atelier for up to 50 and Forge for more than 50, and GLPI Network self-hosted tiers are sized by agents and assets. Done means a signed-off count per role, which you will bring into every vendor conversation.

  3. Shortlist two tools with written criteria

    From GLPI, iTop and Znuny, pick the two that best match your scope from step one, and record why the third is out. Use the scope, the agent count and your existing sign-in and database standards as the filter. Confirm the database and PHP requirements against each vendor's current documentation, because they differ between tools and versions. Done means a written shortlist with the criteria, the named sources and the open questions for each vendor.

  4. Load a representative data sample

    Build a test instance on infrastructure that matches your planned hosting, then import a representative sample of users, groups, assets and existing configuration items, using the tool's import route such as CSV import where it is supported. Include messy records, duplicates and missing owners so you can see how the data model and consistency checks handle them. Done means a documented import log showing which fields mapped cleanly, which needed manual cleanup and how long your team spent on each.

  5. Run one change and one incident end to end

    Have agents run one standard change from request through approval, implementation and closure, and one incident from intake through assignment, escalation, SLA tracking and resolution, in the test instance. Link the affected assets and check whether impact analysis or related-item views show what you expect. Done means a short sign-off from the change manager and service desk lead that each workflow works with your approval steps, SLA rules and notifications, with any gaps logged as configuration or customization work.

  6. Test LDAP or SAML sign-in

    Connect the test instance to your directory and confirm how users and groups are created, mapped to profiles and removed when they leave. Test one account from each group that holds a different role, and check what happens when a directory group is missing. GLPI tries internal database, LDAP, IMAP and external methods in a defined order, so confirm that order matches your policy. Done means a documented test matrix that the identity team has reviewed and accepted.

  7. Check the upgrade procedure before committing

    Read the vendor's update documentation and walk through it on a copy of the test instance. For GLPI, confirm the backup set, including the key files in the config directory, the file replacement step and the database update command, and check how plugins behave after a new version. For iTop and Znuny, confirm the supported release lines and the matching database and PHP versions. Done means a written upgrade runbook, a rollback plan and a list of questions for the vendor or for DeployOpen, so you know the recurring effort before you sign.

Questions buyers ask before hiring help with open-source ITSM

What drives the cost and effort of self-hosting GLPI or iTop?

Cost is driven mainly by the scope of your CMDB, the number of agents or assets, the integrations you need, and how much of the setup you do internally. Self-hosting means you pay for infrastructure, the implementation work, and ongoing operations, and those recur every year. Licenses differ: GLPI is GPL-3.0-or-later, while iTop is AGPL-3.0 with paid Atelier and Forge tiers sized by agent count, so check the tier that matches your agent count. Ask any implementation partner for a fixed-scope quote that lists CMDB data sources, integrations, hosting, and support hours separately, and confirm whether paid vendor support or a subscription is included.

How long does a migration from ServiceNow or spreadsheets take?

No supplied fact gives a migration duration, so treat any number you hear as a question to confirm. The main factors are the volume and quality of existing asset and configuration data, how many ITIL processes you move at once, the number of integrations to rebuild, and whether you migrate history or only open items. Spreadsheet sources usually need cleaning and deduplication before import, while ServiceNow exports need field and workflow mapping. Ask for a written scope that lists data sources, mapping rules, test cycles, and cutover steps, and ask how the plan handles parallel running.

Can a team run GLPI or iTop without a consultant?

Yes, a capable internal team can run these tools, but the answer depends on skills and capacity rather than on the software alone. GLPI runs on PHP 8.3 or later with MariaDB 10.11 or later or MySQL 8.0 or later, and iTop 3.3.x runs on PHP 8.2 to 8.4 with MariaDB preferred and needs GraphViz. Your team will own web server configuration, database backups, patching, and the CMDB data model. A consultant is most useful for the initial data model, integrations, and hardening, after which your staff can take over operations. Confirm the skills needed in your own environment before deciding.

What does the work need from internal staff?

Internal staff need to own decisions that no tool can make for you: which configuration items belong in the CMDB, who approves changes, what the SLAs mean, and who is accountable for data quality. Technically, you need people who can run a LAMP-style stack, manage a database, read logs, and handle scheduled jobs. For GLPI, add the GLPI Agent rollout on endpoints and, for network discovery and SNMP, the separate GLPI Inventory plugin. For iTop, plan for GraphViz and for extension management through the iTop Hub. Ask a partner which named roles they expect you to fill during the engagement.

How do upgrades work for GLPI, and what must be backed up?

GLPI upgrades follow a documented sequence: back up the database, the files, and the config directory, replace the application files, and run php bin/console db:update. The config directory includes glpi.key and glpicrypt.key, and losing them can make encrypted data unreadable, so back them up with every upgrade. Since GLPI 11.0, plugins are suspended after a new version is detected and must be reactivated after major updates, so plan for plugin compatibility testing before cutover. Test the procedure in a staging copy first, and confirm the release notes for any version you are moving to, such as GLPI 12.0.0.

Does Znuny's LTS line change the upgrade plan?

Yes, it changes the decision about which release line you run. Znuny publishes a feature line and a separate LTS line, and znuny.org lists Znuny 7.3.8 and Znuny LTS 6.5.26 as its latest releases in October 2026. An LTS deployment usually means fewer major changes over time, while a feature line brings newer functionality and more frequent change. Znuny's documentation portal keeps separate documentation for the feature version and the LTS version, so use the one that matches your line. Ask Znuny GmbH or your partner to confirm the support window for the line you select.

What running these tools asks of your team

Running GLPI, iTop and Znuny is mostly a question of web server and runtime discipline, and the first check is the web root. GLPI's installation docs require the web server's document root to point to the public directory rather than the installation root, with URL rewriting routing requests through index.php on Apache, Nginx, lighttpd or IIS. Get that wrong and configuration files become reachable over HTTP. Pin your PHP and database versions before you pick a release: GLPI 12.0.0's README asks for PHP 8.3 or later with MariaDB 10.11 or later or MySQL 8.0 or later, while the 11.0 installation docs list PHP 8.2 with MariaDB 10.6, and only MySQL and MariaDB are supported. Confirm which branch your team will run, because the matrix differs between them.

iTop 3.3.x requires PHP 8.2 to 8.4, MySQL 5.7 or later or MariaDB 10.6 to 11.8, and GraphViz for its impact analysis views. Combodo recommends MariaDB and discourages MySQL 8 for performance reasons. Combodo's sizing guidance suggests an all-in-one server with 2 vCPU and 4 GB of memory for fewer than 200 tickets a month and under 20 console users, with separate web and database servers of up to 8 vCPU and 16 GB for larger installations. Treat those figures as a starting point and ask Combodo how they apply to your ticket volume and any planned federation. Znuny runs on Perl 5.16.0 or higher with Apache 2 and mod_perl2 recommended, so your team needs Perl module maintenance skills, not only PHP or Linux administration.

Routines matter more than the initial install. For GLPI, the update procedure requires backing up the database, the files and the config directory, including glpi.key and glpicrypt.key, because losing those keys can make encrypted data unreadable. Since GLPI 11.0, plugins are suspended when a new version is detected and must be reactivated after major updates, so plan a plugin review in every upgrade window. Ask each vendor for its supported upgrade paths, its recommended backup method for your database engine, and how long older branches receive fixes. Znuny's LTS line and feature line have separate documentation, so decide which one you will run before writing the runbook.

How DeployOpen scopes and delivers ITSM selection and implementation

DeployOpen plans, implements, customizes, integrates, hardens, backs up, documents and supports self-hosted open-source platforms inside your own infrastructure. For ITSM, that means turning a shortlist of GLPI, iTop, Znuny and Snipe-IT into a written scope that matches your processes, data and hosting constraints.

How we scope

  • Engagements are fixed-scope, with scope, deliverables and timelines agreed in writing before work starts.
  • The written scope names which ITIL processes, CMDB classes and asset types the chosen platform must cover.
  • Scoping confirms edition limits, such as agent counts and feature availability, against current vendor documentation before design begins.
  • Integration points with identity, monitoring and inventory sources are listed with owners and acceptance criteria.

How we build

  • Deployment runs in your cloud account, data center or air-gapped network, using the database and web server stack the chosen platform supports.
  • Configuration covers ticket workflows, CMDB classes, SLA rules, roles and LDAP or SAML sign-in as agreed in scope.
  • Data migration from spreadsheets or an existing ITSM tool is scripted, reconciled and checked against source records.
  • Security testing of the deployed application and its APIs is scoped separately so findings can be fixed before go-live.

How we hand over

  • Deliverables include runbooks for backup, restore, update and plugin reactivation tasks specific to the platform you chose.
  • Documentation covers configuration decisions, CMDB data model, integration maps and administrator responsibilities.
  • Knowledge transfer sessions are held with your platform, service desk and security staff named in the written scope.
  • Acceptance follows the criteria agreed at scoping, so go-live is measured against written requirements rather than impressions.

How we support

  • Ongoing support, upgrades and continuous testing are available as retainers after handover.
  • Retainer scope should state which vendor releases are applied, how test environments are used and who approves changes.
  • Upgrade work covers database and application updates, along with checks on custom code, extensions and integrations.
  • Ask any quote to separate one-time implementation effort from recurring support, licensing and vendor subscription costs.

Sources and further reading

Talk to our team.

Tell us what you're working on, whether it's a deployment, an audit, a security test or a cyber range. You'll speak with an engineer who can help you scope it.

  • 30-minute call: free, with no obligation.
  • NDA on request: we can sign before you share details.
  • Clear next steps: a scope and plan after the call.