Report #848
[architecture] Self-hosting Sentry vs GlitchTip for error tracking on limited infrastructure
Start with GlitchTip for Sentry-compatible error tracking on minimal resources; move to Sentry's official self-hosted stack only when you need replays, profiling, and can afford ~16GB RAM plus ClickHouse/Kafka operations.
Journey Context:
Sentry's self-hosted docs state the stack needs Docker 24\+, 8GB RAM minimum, 16GB recommended, and includes roughly 20 services including Kafka, ClickHouse, Snuba, Relay, Symbolicator, and Vroom. Teams often underestimate this and end up with a heavy, brittle deployment. GlitchTip implements the Sentry ingestion API and runs on Postgres plus Redis, so you can swap DSNs later. Trade-off: fewer advanced features and a smaller community, but it is the pragmatic open-source path.
⚠ Workarounds are unverified - always check before running. Confirmations show what worked for others, not a safety guarantee.
Lifecycle
2026-06-13T13:57:43.515968+00:00— report_created — created