Incorporate supplied upstream frontier for diagnostic-test review

Retain the batch-start main frontier before integrating the independently reviewed documentation-test candidate. No publication or deployment.

Change-source: pulse-maintainer
This commit is contained in:
pulse-triage[bot]
2026-10-05 05:58:53 +01:00
12 changed files with 564 additions and 77 deletions
@@ -1761,6 +1761,17 @@ payload shape change when the portal presents compact client rows.
`frontend-modern/src/components/Settings/AvailabilitySettingsPanel.tsx`,
`frontend-modern/src/components/Settings/availabilitySettingsModel.ts`,
and `AvailabilityTargetSlot.tsx` own the browser transport shape.
The settings summary and status pills classify each target through
`getAvailabilityTargetHealth`. When the target's unified
`network-endpoint` resource is loaded, the Machines tab's
`getStandaloneResourceStatusIndicator` decides, so stale evidence and
unresolved identity count the same on both pages. Otherwise the target's
own status is read the way `availability_poller.go` derives the resource
status: aggregate and single-location failures are offline only once
consecutive failures reach the failure threshold (default 2), a missing
count is zero, and a never-checked target is pending. Settings and the
Machines availability tab therefore report the same counts in the same
words.
Availability target probe-result and error notices may compose
frontend-primitives' `CalloutCard` for shared settings callout chrome;
API contracts own the target CRUD/test payload semantics and endpoint
@@ -7338,7 +7338,9 @@ successful evidence must render an amber `Stale` state with no green
must not appear as a primary row in the Machines `Availability checks` tab;
the distinct source-owned `network-endpoint` for that configured check must
appear there regardless of whether its correlation state is `attached`,
`standalone`, `ambiguous`, or `unresolved`.
`standalone`, `ambiguous`, or `unresolved`. The tab's summary card owns the one
"Manage checks" link; the checks table toolbar keeps only "Add service/device
check".
Operational navigation for those agentless endpoints belongs to the
frontend-primitives-owned Machines surface as a focused Availability checks tab
rather than a new primary nav item. The page may show availability checks beside
@@ -3246,7 +3246,11 @@ as separate buckets in `standalonePageModel.ts`, include every source-owned
`network-endpoint` regardless of correlation state, exclude matched machine or
service projections from the check inventory by resource type, and let
`AvailabilityChecksTable.tsx` render saved probe method, target, latest result,
check age, failure count, and cadence from the canonical availability payload.
24h uptime, check age, last healthy time and failure count from the canonical
availability payload, with cadence restored only on the widest tables. Uptime
comes from the shared availability history summary the fleet view reads and
stays blank below 90 percent observed. The failure count reads "N in a row",
with the offline threshold on hover, not an "N/M" fraction.
Recent check timing and fuller failure context may stay in tooltip or drawer
detail, but the table row must not duplicate the same probe protocol and
result text across both identity and metric cells.