Tour of the app

This page answers “where do I find what?” It walks through the parts every page shares, the navigation rail and the header, and says what each page is for, with a link to its full guide.

The Stories page, the home of the app, with the rail on the left and the header at the top.

A typical investigation

Open Stories and pick the top story group: the problem that produced the most failing or slow requests in the range.
Read its latest story in the inspector, then select Open story to see the root cause, the waterfall, the comparison with normal and the logs.
From the story, follow the root-cause service to the Service map to see its health and its callers, or a log line to its log template to see how often that line appears and whether it alerted.
Use Traces to find other requests like it, and Alerts to see what the logs flagged around the same time.

The rail on the left links to the five sections. Hover over an icon, or focus it, to see its name. The section you are in is highlighted. On a phone the rail becomes a bar at the bottom of the screen.

The links keep the current time range, and leave page filters behind. Logs and templates opens the log alerts; the templates are one link away from there.

The header

Page title and breadcrumb (1): where you are, for example Logs / Templates / Template 5461…7332. Long ids are shortened to their first and last four characters.
Services degraded: a red or amber pill, such as 2 services degraded, appears beside the title when services had errors or ran slow in the last 15 minutes, whatever the page’s range. Hover over it to see their names; select it to open the service map over the last 15 minutes. It is hidden while every service is healthy. (Not numbered in the screenshot.)
Command palette (2): “Jump to service, trace id, template…”. Select it or press ⌘+K or Ctrl+K to jump anywhere. See Command palette and keyboard shortcuts.
Time range presets (3) and Custom… (4): the window every page shows. See Time range and live mode.
Live (5): refreshes the page every 10 seconds.
Theme switch (6): cycles light, dark and system. See Theme.
User menu: only when sign-in is turned on. See Sign-in.

Press Tab once on any page to reveal a Skip to content link that jumps past the rail and the header.

The pages

StoriesWhat is failing or slow now: story groups ranked by count, with the latest story beside them.

Story detailOne request explained: root cause, request path, waterfall, comparison with normal and logs.

TracesSearch traces by service, endpoint, duration and errors; open any trace as a waterfall.

Service mapServices and their calls with health, rate, errors and p99; a drawer per service.

Logs and templatesLog lines grouped into templates, with hits over time and the silence alert setting.

AlertsNew, spike and silence alerts on log templates, with example traces.

PipelineWhether Tayga itself is healthy: job status, throughput charts and consumer lag.

Command palette and shortcutsJump to a page, service, trace, story group or template from the keyboard.

States you will see on every page

Loading: grey placeholders in the shape of the content.
A panel that failed: a red banner, “Could not load …”, with the HTTP status, the message and Try again. Other panels keep working.
A refresh that failed: the panel keeps its last data and says “Refresh failed · showing data from 14:03:07”, with Try again.
Storage unavailable: a red banner under the header while the database cannot be reached, with “Retrying on the next refresh.” It clears on its own when a request succeeds.
Page not found: an address that matches no page shows “Page not found” with Go to stories. The rail and the header stay usable.

Tips

Every view’s state (filters, selection, open drawer, time range) is in its URL. Copy the address to share exactly what you see.
Press ? on any page for the list of keyboard shortcuts.