Repository navigation
feat: Alerts gather what needs someone in Management, with a count in the sidebar, the phone's bar and the account menu - #2423
Merged
Conversation
36 of 52 tasks
Contributor
Changes
0.90 test lines per prod line. Coverage
Patch coverage: 100.0%, 206 of 206 changed lines covered. |
… dead jobs and open exceptions A module that owns a cause implements `AlertSource`, so the alerts module reads no other module's data. Alerts are raised afresh on every read and clear once their cause is gone. A reader sees the alerts their roles can act on and may hide one for themself, which lasts until the alert clears. Hiding, showing and forgetting a hide are each one statement that may run twice, since the sidebar and the page read and write at the same moment.
… the sidebar, the phone's bar and the account menu Each alert links to where it is dealt with, and the reader hides one for themself or shows it again, on a phone as well. The count is read again on every page. A read that fails keeps the alerts it had and the page says the alerts could not be read.
ExtraToast
force-pushed
the
feat/2385-alerts
branch
from
October 13, 2026 10:02
a156b9c to
c194cfe
Compare
Contributor
The SQL
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Part of #2385 (step 6, its last pull request). Brings back #1941 from the archive tag.
What this is
One Alerts page, and a count in the sidebar, the phone's bar and the account menu, for what needs someone in Management.
AlertSource(alerts :: api), so the alerts module reads no other module's data. Alerts are raised afresh on every read and clear once their cause is gone.New module
alerts, new tablehidden_alert, three endpoints under/management/alerts(board only), one page and a new top entry in the portal's navigation.What differs from the archive, and why
AlertKindhasJOB_DEADandEXCEPTION_OPEN. Drift, a cohort without its list, failed email, a role waiting on two-factor and the Discord kinds come with the steps that own their cause, each adding its kind, its words indomains/alerts/reading.tsand its source.job-dead,exception-open) and the alert carries a mark that rises only when a new cause arrives. A hide covers the alert up to the mark it was taken at: retry one of three dead jobs and it stays hidden, a fourth dies and it shows again.Known, not changed
hidden_alerthas no foreign key to the user. A row left by a deleted user is a few bytes that nobody reads.How it was checked
AlertsIT,RecordedExceptionITand the jobs integration tests pass against a throwaway MariaDB: 62 tests, among them a hide that outlives a retry of the newest dead job and ends at a new one.services/api/src/mainandservices/frontend/srcis hit by a unit test, measured against jacoco and lcov.tests/e2e/alerts.spec.tspasses on a dev server of this branch, and the page was looked at on a wide window, with the account menu open and at phone width.