The workflows teams expect from Sentry, without the stack tax.

urgentry keeps issue triage, tracing, replay, profiling, alerts, and query-heavy investigation in one product surface. The goal is simple: let teams validate the workflows they already rely on while carrying a much smaller self-hosted footprint.

Maintained by Wraxle LLC
Last updated
218/218 source-scanned Sentry operations covered on the public compatibility line
2.2x more same-host self-hosted throughput than Sentry self-hosted
52.3 MB Tiny-mode peak memory on the current reference benchmark lane
1 DSN keep the SDKs your team already ships and validate one workflow at a time

The headline numbers come from the current same-host benchmark line: urgentry self-hosted sustained 2,200 eps at 391.8 MB peak memory, versus 1,000 eps and 8191.7 MB for the current Sentry reference row. Read the benchmark method before repeating the shorthand.

Group the issue, follow the trace, ask the query.

The strongest story on this page is the shortest path from a bad event to a useful answer.

Issue tracking

Group errors deterministically, merge or unmerge bad groupings, and keep the triage path close to assignments, comments, releases, and regression checks.

  • Automatic grouping by exception type and stack trace
  • Assignments, bookmarks, subscriptions, and comments
  • Release tracking, regressions, and state workflow
issue list payload
GET /api/0/projects/acme/backend/issues/

[
  {
    "id": "grp-001",
    "shortId": "URGENTRY-42",
    "title": "ConnectionError: timeout",
    "culprit": "db.query in handler.go",
    "count": "847",
    "userCount": 23
  }
]

Discover and logs

Query issues, transactions, and logs in one place so an alert or issue can widen into a broader investigation without a tool switch.

  • Aggregations, group-by, and time-series views
  • Saved queries with sharing
  • Structured OTLP log ingestion
discover query
{
  "dataset": "transactions",
  "select": [
    {"alias": "transaction", "expr": {"field": "transaction"}},
    {"alias": "p95", "expr": {"call": "p95", "args": [{"field": "duration.ms"}]}},
    {"alias": "count", "expr": {"call": "count"}}
  ],
  "groupBy": [{"field": "transaction"}]
}

Performance

Ingest traces through the Sentry SDKs or OTLP, then follow transaction timing, span waterfalls, and latency percentiles without bolting on another product.

  • OTLP v1 trace and log ingestion
  • Span waterfall and latency breakdowns
  • Tag-based filtering and grouping
trace ingest
curl -X POST http://localhost:8080/api/proj-1/otlp/v1/traces/ \
  -H "Content-Type: application/json" \
  -d @traces.json

sentry_sdk.init(
    dsn="http://key@localhost:8080/1",
    traces_sample_rate=1.0,
)

See the session. Find the hot path.

Replay and profiling stay tied to the same release, trace, and issue context instead of becoming a disconnected add-on.

Session replay

DOM snapshots, timeline playback, click tracking, and linked issue context so support and engineering can look at the same session.

Profiling

CPU profiles with flamegraphs, call trees, and hot-path analysis tied back to releases and traces.

Product screens

The clearest proof is the product itself. These screens show the current issue, replay, and profile surfaces.

Issues

Grouped issue triage with quick actions and the main team queue.

Replay

Session replay with timeline, linked issue context, and asset access.

Profiling

Profile detail with call trees, hot paths, and trace-aware analysis.

Know when the issue lands. Know when the job stops.

Issue alerts, metric alerts, and cron monitors cover the paths most teams reach for before they need another incident tool.

Issue alerts

Trigger on new issues, regressions, or threshold breaches with project and environment filters.

Metric alerts

Track error count, transaction count, p95 latency, failure rate, and apdex windows.

Cron monitors

Detect missed, late, and failed scheduled jobs with check-in heartbeats.

Email and Slack

Use the built-in delivery paths for issue context, team routing, and the first escalation layer.

Webhooks

Send structured JSON payloads to the systems you already use for routing and incident workflows.

PagerDuty and Jira

Create incidents and linked tickets when the problem belongs in the on-call or project-management lane.

Keep the SDKs your team already ships.

Use the existing SDK packages, validate the common ingest path, then widen the rollout when the boring checks pass.

Migration proof
218/218
source-scanned Sentry operations covered on the public compatibility line
  • Keep the existing SDKs and change the DSN
  • Validate the route with the benchmark and operator docs
  • Use Tiny first, self-hosted when the team already needs shared services

Pick the next proof and keep moving.

Once you know which workflow matters most, validate it with the migration guide, the SDK snippets, and the benchmark or operator pages that fit your rollout.