Error Tracking Overview

Tindra captures unhandled exceptions, logged errors, and manually captured events from your applications.

What's in each event#

• Exception type and message: what went wrong
• Stacktrace: file, line, and surrounding code context
• Breadcrumbs: a timeline of activity leading up to the error
• User context: ID, email, username (if set by your SDK)
• Request context: URL, method, headers, body
• Environment: release version, environment tag, server name

Issues vs. events#

Tindra groups similar errors into issues. Two exceptions with the same type and stacktrace fingerprint land in the same issue.

• Issue: the unique problem (e.g., TypeError: Cannot read property 'id' of null)
• Event: a single occurrence of that problem

Events become searchable after ingestion workers persist and process them. An accepted SDK request does not by itself confirm that an event is already stored.

Issue lifecycle#

An issue opens when a new fingerprint is seen for the first time. Resolve it manually when the fix ships. If the same error reappears after the next release, the issue re-opens automatically.

Filtering#

Use the shared filters to select projects, environment, time, and an affected application user. The issues list also has title search, status, level, assignee, and tag filters. From an event's user card, follow the same user into logs and traces.

Next steps#

• Issues: searching, triaging, and resolving
• Environments: separate production from staging
• Releases: track which deploy introduced an error
• Alerts: get notified when error rates spike