Sentry

KEEP IT

The value is the network, the data or the infrastructure. Keep paying.

Catches errors and performance problems in running software. An SDK in your app sends exceptions with the stack trace, the release, the browser and the user action that led there, then groups them so one broken deploy is one issue and not ten thousand.

Promptfree, for everyone, and the only version there is

Build me a minimal error tracker for one small application. It replaces the interface, not the infrastructure.

STACK
- Node with Fastify, SQLite via better-sqlite3, server-rendered HTML
- Hosted on separate infrastructure from the application it watches

INGEST
- JSON payload: message, exception type, stack trace, release, environment, request context, arbitrary tags
- Authenticated by a project key

GROUPING
- Fingerprint on normalised exception type plus the topmost application stack frames
- Ignore line numbers, which shift between builds and would split one issue into fifty

ISSUES
- List ordered by last seen, with occurrence counts and a sparkline of the last 24 hours
- An issue page showing the newest event, the parsed stack trace with source-map resolution when a map is present, and the tag distribution across occurrences

SURVIVING A CRASH LOOP
- Rate limiting and sampling at ingest, so one bad deploy cannot fill the disk
- A visible counter of dropped events, because silent dropping is how you stop trusting it

ALERTS
- A webhook when a new issue appears, or an existing one crosses a threshold

Include a README saying plainly that a collector sharing a failure domain with the thing it watches is worse than no collector.

What you lose

Ingestion that stays up while your application is falling over — the one moment your own error collector would also be down
Source map and debug symbol processing across JavaScript, iOS, Android, Java and native binaries
Grouping heuristics that turn ten thousand identical exceptions into one issue with a count
Maintained SDKs for well over a hundred platforms, including breadcrumbs, release tracking and performance traces
Storage and retention of high-volume event data, which is the cost you are actually paying for

If you would rather not build

Sentry itself, self-hosted — the official Docker Compose distribution
Highlight — open-source monitoring with session replay

The escape hatch

open source · no votes, no paid placement

Sentry (self-hosted)

Sentry's own stack run by you: the same features, your servers and your storage bill.

getsentry/self-hostedfree · open source

GlitchTip

Error tracking that speaks the Sentry SDK protocol and is far lighter to run.

GlitchTip/glitchtipfree · open source

Why this verdict

our own opinion · changed only by a person

28/100

Verdict no at 28: fingerprinting and a list view are approachable, but the value is availability during your outage, symbolication across runtimes, and event retention at scale. Sentry is also open source and self-hostable, which makes rebuilding it from scratch the worst of both options.

History

tracked since 6 Aug 2026 · nothing is ever overwritten

Interest · last 30 days

Nothing recorded yet. This chart fills in once the page has visitors, votes or prompt copies — it will not draw a flat line to look busy.

Questions about Sentry

answered from the record above

Is Sentry free?

No — the plan we track is $26 a month. Team plan, $26/mo with a default prepaid event allowance. The real bill is driven by event volume, and an incident is exactly when volume spikes.

Can you replace Sentry by building your own?

KEEP IT. The value is the network, the data or the infrastructure. Keep paying. Replacement score 28 out of 100, build time longer than it saves. Read what you lose before you decide.

How much does Sentry cost?

$26 a month on Team — $312 a year. Recorded 6 Aug 2026.

What do you lose by replacing Sentry?

Ingestion that stays up while your application is falling over — the one moment your own error collector would also be down; Source map and debug symbol processing across JavaScript, iOS, Android, Java and native binaries; Grouping heuristics that turn ten thousand identical exceptions into one issue with a count; Maintained SDKs for well over a hundred platforms, including breadcrumbs, release tracking and performance traces; Storage and retention of high-volume event data, which is the cost you are actually paying for. If any of those carry weight for you, keep paying.

Is there an open-source alternative to Sentry?

Yes: Sentry (self-hosted), GlitchTip. The prompt on this page is for when you want it your way instead.

Related entries

same category first, most replaced first

All 14 in Monitoring & uptime

Coming soon

Every week, something stops being worth paying for.

New verdicts, prices that moved, entries added. One email a week. Unsubscribe in one click.

free forever · no tracking pixel · every prompt stays free on the site