What does root-cause analysis on a Rius alert do?
Rius can hand a firing alert to an analysis agent that writes a report on the firing’s page: what failed, since when, and its reading of the cause. The analysis is optional, switched on per alert and per workspace. It runs against a monthly analysis budget, with at most one analysis per service every 5 minutes. Admins can also run one by hand from the firing page.
How it works
A firing is one occurrence of an alert’s condition. When root-cause analysis is on, Rius hands the firing to the analysis agent, and the report is written back to the firing’s page as soon as the run completes. See Alerts for how firings open, renotify and resolve.
Two switches control it:
- Workspace level: root-cause analysis must be enabled for the workspace.
- Alert level: each alert has its own Root-cause analysis switch. When both are on, every firing of that alert goes to the analysis agent.
Two limits apply:
- Monthly budget: each analysis counts against the workspace’s monthly analysis budget. When the budget is used up, firings are not analyzed and show Over budget. User summaries count against the same budget, one analysis each.
- Rate limit: Rius runs at most one analysis per service every 5 minutes. A firing held back by this cooldown shows Skipped.
Admins can click Run analysis on a firing page whatever the alert’s switch says, and without waiting for the cooldown. It counts as one analysis against the budget. It is not offered while an analysis is queued or running, or on a firing that already has a report or is over budget.
Set it up
-
Open the alert’s details page and turn on its Root-cause analysis switch. Configuring alerting is an admin action.
-
Make sure root-cause analysis is enabled for the workspace. Check the budget in the Intelligence section of the workspace settings. When analyses are limited, it shows how many are used this period and when the count resets.
Settings → Workspace → Intelligence -
Wait for a firing, or open an existing firing and click Run analysis if you are an admin. The report appears on that firing’s page when the run completes.
-
Track progress in the analysis status on the Firings list and on each firing’s page:
Status Meaning Off Analysis is off for this alert or workspace Queued Waiting to run Running The analysis agent is working on it Report Done; the report is on the firing’s page Skipped Skipped by the cooldown, at most one analysis per service every 5 minutes Over budget The workspace’s monthly analysis budget is used up Failed The run failed No report Accepted, but no report came back A firing that was never analyzed shows a dash.
-
On the Firings page, turn on the Has analysis filter to list only firings with a report. Your filters are kept in the page URL, so you can share the filtered view as a link.
What you get
- A written report on the firing’s page: what failed, since when, and the analysis agent’s reading of the cause.
- The report sits beside the rest of the firing: when it opened, when the condition was last seen, the affected traces linked to the trace view, and the delivery list for each channel.
- View all in Traces opens the trace list pre-filtered to the firing’s signature and time window, so you can check the analysis against the traces.
- A clear status on every firing, so you can tell a queued run from a skipped, over-budget, failed or missing one.
- Control over spend: the budget caps analyses each month, and the Intelligence section shows how many you have used.
When something else fits
If you want no analysis at all, leave the alert’s switch off. The alert still fires and notifies your channels, and the firing page still lists the affected traces. If a firing is skipped by the cooldown or shows Over budget, work from the affected traces and the trace list filtered to the firing, or ask an admin to click Run analysis where it is offered.