Back to blog
TL;DR for busy leaders
Bare minimum cloud footprint: ≈ $500/mo CapEx style spend
Ongoing Ops & support: ≈ $600–$800/mo OpEx
Unplanned outages = brand & SLA risk owned by you
In 6–9 months the TCO of “free” often exceeds a turnkey SaaS subscription.
Many teams looking to adopt tools like ReportPortal are drawn to the idea of self-hosting – especially because it seems like the "free" option. ReportPortal is open-source and doesn’t charge you to download or run the software yourself. But before you decide to go the On-Premises route, it’s important to do the math properly and understand what "free" really means in this context.
Running ReportPortal on your own infrastructure comes with more than just hardware expenses. Here’s what you need to consider:
1. CapEx and Baseline Infrastructure
Even if you already have servers, you'll need to allocate resources to run ReportPortal reliably, including backups, scaling and availability configurations. You can check our Hardware Requirements to estimate what’s needed for a typical deployment. This includes:
Compute power to run the services and application containers - App + API - OpenSearch - PostgreSQL database
Storage for logs, history, and analytics
Backup systems and disaster recovery
Networking and bandwidth considerations
You’ll also need to plan for future growth. As your team expands, you may need to scale the system: allocate more disk space, CPU, and memory to accommodate larger data volumes and increased usage.
2. OpEx - Maintenance & DevOps
Self-hosting means you own the responsibility for keeping everything running – both technically and operationally. That includes:
Regular updates and security patches
Compatibility checks after each upgrade
Downtime recovery and root cause analysis
Docker orchestration or Kubernetes configuration
CI/CD pipelines to deploy updates
Secrets and environment management
While ReportPortal is build to be as plug-and-play as possible, Installing and running production grade ReportPortal is not a trivial task to be handled by your QA team. You’ll need a DevOps specialist to deploy, update, and maintain the environment. For example, PostgreSQL requires regular maintenance – like scheduled VACUUM FULL operations – to keep performance steady. Between ReportPortal versions, your DevOps team will also need to make configuration changes, adjust file formats, and validate system compatibility.
After every ReportPortal update, someone will need to validate that everything is working correctly. It’s not guaranteed that your DevOps team will catch all the issues – so your QA or engineering team may need to step in and help. That means additional effort.
3. Monitoring & Observability – risk management
SaaS platforms come with monitoring built in – but if you go On-Premises, you need to build and maintain that yourself:
Observability stack in place - health checks, uptime alerts, and usage metrics
Runbooks & RTO/RPO targets.
Log aggregation and retention
Actionable alerting when something breaks – ideally before your users notice
We strongly recommend setting up monitoring, for example with Grafana, so you can track critical parameters: CPU usage, memory, disk space (for attachments, database, and OpenSearch), RabbitMQ queues, and more. You’ll need to keep an eye on these to ensure you’re delivering a reliable service to your team or enterprise – because if something runs out, parts of your system can go down.
Based on this monitoring, you’ll also be able to adjust the resources allocated to each service to keep things running smoothly and efficiently.
4. User Support
When users run into issues – errors, slowness – they’ll come to your team.
They’ll reach out with a wide range of questions when something stops working or doesn’t behave as expected. Your QA lead becomes Tier‑1 support.
You’ll need to respond with things like: "Try regenerating your API key", "Double-check that your properties are configured correctly", "Update this parameter in the widget settings".
All of this consumes time and usually pulls engineers or QA leads away from their core work.
5. Compliance & Security
When self-hosting, you are responsible for:
Controlling access to user data
Implementing encryption and secure access
If you plan to integrate with an external identity provider (like SAML), you will likely need to engage your internal team that manages the identity platform. That means additional communication, coordination, and time – especially during the initial setup phase.
Yes, ReportPortal is free to use. But hosting it yourself comes with a hidden cost. With an On-Premises setup, you need to consider the real costs – engineering time for setup, updates, and maintenance, cloud or hardware expenses, internal user support, and the risk of something breaking without anyone noticing.
If your team doesn’t want to spend valuable time managing servers, chasing down issues, or maintaining infrastructure just to keep things running, then the SaaS option is a smarter choice. It’s more scalable, more secure, and lets your team focus on what really matters – testing, delivering quality, and moving faster.
Bottom Line for Execs
Selfhosting ReportPortal is a strategic choice, not a bargain. Unless data sovereignty mandates dictate onprem, the math favors SaaS:
Breakeven ≈ 1 month at the starter scale outlined above. ≈ 3-4 months at the enterprise scale setup. After that, selfhosting is more expensive and more distracting.
Focus your people on shipping value, not babysitting infrastructure. Upgrade to SaaS and grow without the growing pains.