Logs and Errors

Uptrace collects logs and exceptions alongside traces, correlating them with spans automatically. The Logs & Errors section provides grouped and raw views of log records with filtering, search, and per-entry detail.

Groups view

By default, similar logs are grouped automatically by their _group_id. The Groups view shows each unique log pattern as a row with its frequency and last occurrence.

Default aggregations are perMin(count()) and max(_time). Sparklines on the right show how group frequency changes over time.

The Grouping rules button in the app bar opens the project's grouping rules. Its Debug page runs the rules against log lines you type or pick from the project's logs, and shows which group each line lands in — see Debugging rules.

Logs tab

The Logs tab shows individual log records in chronological order with a volume histogram at the top.

Filters

Logs are filterable by any indexed attribute — service_name, log_severity, host_name, deployment_environment, and any custom attributes recorded on the log entry. Filters support equality, exclusion, regex, and range expressions. The Show filters panel lists all available attributes with their values for quick selection.

Search

A search term is part of the query. Type it into the + menu of the Filters row and pick Search "…" on this page, or search from the command palette (⌘K or Ctrl+K). The term then stands first in the row as a chip with a magnifier icon, which you edit or remove like any other filter. Clear and Reset remove it together with the other filters.

Search supports full-text matching across group-by columns with AND/OR logic, exclusion prefixes, and attribute-scoped queries (_kind:client, -key:value). Matching terms are highlighted inline. Links with the older ?search= parameter still work and open with the term as a chip.

See Searching for the full syntax reference.

Log entry detail panel

Selecting a log record opens a detail panel. Some tabs appear only when the record has the data for them.

Attrs — all key-value attributes recorded on the log entry, with type annotations (::str, ::int, ::bool). Attributes are grouped by category: code, service, telemetry, other. A filter field narrows the list by key name.

Stacktrace — the exception stack trace, for an error record that carries one.

Breadcrumbs — the Sentry breadcrumbs that led up to an error, as a timeline.

Compare — the log group's records per minute in the current time range against the same range an hour, a day, or a week earlier, with a chart that marks this record.

Traces — for a record written inside a trace, the span groups that co-occurred in the same traces as this log entry, with configurable columns and an Open in explorer button.

Context — the log records written just before and after this one on the same host or in the same log file. The Filters box picks which of these attributes must match.

JSON — full internal record including span metadata (id, parentId, traceId, groupId, type, system) and the complete attrs object.

Tags — selects dashboard tags. For each selected tag, the panel adds a tab with the metrics of the dashboards that carry the tag and match the record's attributes.

Timeseries view

The Timeseries tab shows log group frequency over time as a line chart. Individual groups can be toggled to isolate patterns.

Error monitors automatically fire alerts when logs reach ERROR or FATAL severity, or match a specific message pattern or exception type.