Files
oikos/web/src/lib/health.test.ts
dtoro cc8eae4979
Some checks failed
ci / build-test (push) Has been cancelled
ci / docker-build (push) Has been cancelled
ci / web (push) Has been cancelled
Desktop App / Build Linux (amd64) (push) Has been cancelled
Desktop App / Attach to Release (push) Has been cancelled
perf(web): patch health in place instead of refetching, and reconnect the SSE stream
Refetching everything on a health event was wasteful and churned the UI: one
container going degraded pulled down the entire fleet entity list (plus its
parent-grouping pass), or the whole fleet graph, to learn something the event
had already delivered.

health.changed / health.stale carry the new value in their payload, so the
views that hold the entity just patch it:

- Fleet table: patch the row. Only entity.* changes which entities exist, so
  only that still refetches.
- Fleet map: patch the node AND graph.health[id] — healthOf() reads the side
  map in preference to the node's own field, so patching only the nodes would
  have left the rendered colour unchanged.
- Entity detail: patch the open entity. Signals still need a read (the event
  says one was raised, not what the list now contains) but only the signals,
  not the entity and checks alongside them.

Shared in $lib/health.ts, which returns the original array when an event does
not apply so unrelated rows keep their identity and do not re-render. Note it
matches on entity_id, never data.slug: the scheduler emits health.changed with
entity_id = the observed entity but slug = the *check's* slug.

Separately, events.ts had no reconnect. onerror was empty on the assumption
the browser retries, but EventSource only does that for a transient failure --
once it reaches CLOSED (an HTTP error on connect, e.g. the API restarting
during a deploy) it stays closed forever. A single blip silently froze every
live surface in the app with nothing on screen to say so. Now reconnects with
capped exponential backoff, and exports eventsConnected so a future indicator
can show when the stream is down.

Verified against live prod: flipping lxc:apps health recoloured the map node
and moved its counts (30 healthy -> 29, 9 down -> 10) with ZERO network
requests.

Co-Authored-By: Claude <noreply@anthropic.com>
2026-07-28 19:55:26 +02:00

86 lines
3.0 KiB
TypeScript

import { describe, it, expect } from 'vitest'
import { isHealthEvent, healthFromEvent, applyHealthEvent, applyHealthEventTo } from './health'
import type { Entity } from './api'
import type { OikosEvent } from './stores/events'
function ev(partial: Partial<OikosEvent>): OikosEvent {
return {
id: 1,
ts: '2026-07-28T12:00:00Z',
type: 'health.changed',
entity_id: 'e1',
severity: 'info',
source: 'scheduler',
data: { to: 'degraded' },
correlation_id: null,
...partial
} as OikosEvent
}
function entity(partial: Partial<Entity> = {}): Entity {
return {
id: 'e1',
slug: 'lxc:apps',
type: 'lxc',
name: 'apps',
attributes: {},
version: 1,
created_at: '',
updated_at: '',
health: 'healthy',
...partial
} as Entity
}
describe('health event application', () => {
it('reads the new health out of the payload', () => {
expect(healthFromEvent(ev({}))).toBe('degraded')
expect(healthFromEvent(ev({ type: 'health.stale', data: { to: 'stale' } }))).toBe('stale')
})
it('ignores events that are not health events', () => {
expect(isHealthEvent(ev({ type: 'signal.raised' }))).toBe(false)
expect(healthFromEvent(ev({ type: 'execution.output' }))).toBeNull()
})
it('ignores a health event with no usable payload', () => {
expect(healthFromEvent(ev({ data: {} }))).toBeNull()
expect(healthFromEvent(ev({ entity_id: null }))).toBeNull()
})
it('patches the matching entity in a list and leaves the rest alone', () => {
const list = [entity(), entity({ id: 'e2', slug: 'lxc:dns', health: 'healthy' })]
const next = applyHealthEvent(list, ev({}))
expect(next[0].health).toBe('degraded')
expect(next[0].last_check_at).toBe('2026-07-28T12:00:00Z')
// untouched entities keep their identity, so rows that did not change do
// not re-render
expect(next[1]).toBe(list[1])
})
// Returning the same array reference matters: assigning a fresh array on
// every unrelated event would churn the whole table.
it('returns the original array when the event does not apply', () => {
const list = [entity()]
expect(applyHealthEvent(list, ev({ entity_id: 'nobody' }))).toBe(list)
expect(applyHealthEvent(list, ev({ type: 'signal.raised' }))).toBe(list)
expect(applyHealthEvent(list, ev({ data: { to: 'healthy' } }))).toBe(list) // already healthy
})
it('patches a single entity for a detail view', () => {
const e = entity()
const next = applyHealthEventTo(e, ev({}))
expect(next?.health).toBe('degraded')
expect(applyHealthEventTo(e, ev({ entity_id: 'other' }))).toBe(e)
expect(applyHealthEventTo(null, ev({}))).toBeNull()
})
// The scheduler emits health.changed with entity_id = the observed entity
// but data.slug = the *check's* slug, so slug must never be used to match.
it('matches on entity_id, never on the payload slug', () => {
const list = [entity()]
const next = applyHealthEvent(list, ev({ data: { to: 'down', slug: 'check:ping:whatever:0' } }))
expect(next[0].health).toBe('down')
})
})