Prompt
I'm building an internal platform and want self-hosted error tracking for compliance. What options make sense?
Latest observation
If you need self-hosted error tracking for compliance, the main question is usually:
- Do you need full control over data residency and retention?
- Do you want only error tracking, or also performance/APM, logs, and replay?
- How much operational overhead can your team handle?
Here are the most sensible options.
Strong self-hosted choices
1) Sentry Self-Hosted
Best overall for most teams.
- Very mature and widely used
- Covers error tracking, performance, releases, basic user context, source maps, alerting
- Open source and self-hostable
- Good SDK support across many languages/frameworks
Pros
- Feature-rich
- Good developer experience
- Large ecosystem and lots of docs
- Fits compliance needs if deployed in your environment
Cons
- Operationally heavy compared to SaaS
- Can be resource-intensive at scale
- Upgrades can be nontrivial
Use if: you want the closest thing to a commercial-grade error platform but hosted by you.
2) GlitchTip
Best lightweight alternative to Sentry.
- Open source
- Sentry-compatible SDKs in many cases
- Easier to run than Sentry in many environments
- Good for teams wanting simpler ops
Pros
- Simpler deployment
- Lower overhead
- Familiar workflow if you’ve used Sentry
Cons
- Fewer advanced features than Sentry
- Smaller ecosystem
- May lag on deep functionality or edge-case SDK support
Use if: you want “good enough Sentry-like” functionality with less operational burden.
3) Errbit
Best if you only want basic exception aggregation.
- Open source
- Error catcher for application exceptions
- Historically used as a self-hosted Airbrake-like solution
Pros
- Simple
- Lightweight
- Good for straightforward exception tracking
Cons
- Less modern
- Limited compared to Sentry/GlitchTip
- Less ideal for distributed tracing or richer workflow
Use if: you need basic exception capture and want minimal complexity.
4) Better Stack / Airbrake / Rollbar / Bugsnag
These are generally SaaS-first, not ideal if you need self-hosting for compliance.
Some may offer private deployment or enterprise options, but they’re usually not the first choice if “must self-host” is the priority.
If you want “platform” rather than just errors
Sometimes compliance requirements come with broader observability needs. In that case, you may want a stack like:
- OpenTelemetry for instrumentation
- Grafana stack for metrics/logs/traces
- Loki for logs
- Tempo for traces
- Grafana dashboards/alerting
This won’t give you Sentry-style exception triage out of the box, but it’s excellent for a self-hosted observability platform.
Practical recommendation
Pick Sentry Self-Hosted if:
- You want the richest error tracking experience
- You need performance monitoring too
- You can support the operational overhead
Pick GlitchTip if:
- You want something simpler and cheaper to run
- You mainly need exception tracking and alerting
- You want a Sentry-like experience with less complexity
Pick Errbit if:
- You only need basic exception collection
- You care more about simplicity than features
Compliance considerations
No matter which tool you choose, confirm:
- Data retention controls
- PII scrubbing/redaction
- Encryption at rest and in transit
- Role-based access control
- Audit logs
- Backups and restore testing
- Tenant/data isolation if needed
- Whether source maps or stack traces may contain sensitive data
My default recommendation
For most internal platforms needing compliance-friendly self-hosted error tracking:
- Start with GlitchTip if you want simplicity
- Choose Sentry Self-Hosted if you need maximum capability and can operate it
If you want, I can also give you a comparison table of Sentry vs GlitchTip vs Errbit, or suggest the best option based on your stack and traffic volume.