Skip to content

armgit / Pulse

Public Imported from GitHub
0

Real-time monitoring dashboard for Proxmox VE, PBS, Docker, Kubernetes, TrueNAS and vSphere. Self-hosted, with alerts for failed backups, full disks, unhealthy containers and offline nodes.

Avoid overlapping backup documentation insertions

pulse-triage[bot]c4205208ad61
diff --git a/docs/release-control/v6/internal/subsystems/storage-recovery.md b/docs/release-control/v6/internal/subsystems/storage-recovery.md
index b447ef0517..093af7ff0f 100644
--- a/docs/release-control/v6/internal/subsystems/storage-recovery.md
+++ b/docs/release-control/v6/internal/subsystems/storage-recovery.md
@@ -21,6 +21,24 @@
## Purpose
+### Protection assessment batches share the consuming read's lifetime
+
+Every protection-posture batch carries a cancellable transport signal owned by
+the one joined query read. Unmounting, replacing a read, changing the resource
+scope or switching organisations cancels all its pending batches. A failed batch
+cancels its siblings without replacing the original error; only a complete
+successful retry restores assessments. The abort listener is removed when the
+joined read settles. Existing stale-result guards and organisation/cache
+boundaries remain in force even for a transport that ignores cancellation.
+
+The registered `useProtectionPostures.test.ts` and
+`useProtectionPostures.cancellation.test.ts` cover successful shared signals,
+scope/disposal/organisation cancellation, late results, polling, failure and
+full-read recovery. Source-bound browser proof uses the real Backups consumer
+and HTTP transport at desktop and narrow widths. This stops unnecessary stale
+reads; it does not establish native guest recovery or change posture policy,
+batch size, cadence, API, identity or access grants.
+
### Same-named local backup disks retain independent read origins
An accepted storage listing on one PVE node cannot erase last-known files on
@@ -49,24 +67,6 @@ Retained rows are not a fresh read, archive-availability or installed recovery
claim, and this does not establish #3248's original stale-timestamp cause.
-### Protection assessment batches share the consuming read's lifetime
-
-Every protection-posture batch carries a cancellable transport signal owned by
-the one joined query read. Unmounting, replacing a read, changing the resource
-scope or switching organisations cancels all its pending batches. A failed batch
-cancels its siblings without replacing the original error; only a complete
-successful retry restores assessments. The abort listener is removed when the
-joined read settles. Existing stale-result guards and organisation/cache
-boundaries remain in force even for a transport that ignores cancellation.
-
-The registered `useProtectionPostures.test.ts` and
-`useProtectionPostures.cancellation.test.ts` cover successful shared signals,
-scope/disposal/organisation cancellation, late results, polling, failure and
-full-read recovery. Source-bound browser proof uses the real Backups consumer
-and HTTP transport at desktop and narrow widths. This stops unnecessary stale
-reads; it does not establish native guest recovery or change posture policy,
-batch size, cadence, API, identity or access grants.
-
### Backup-location filters keep exact repository identity
A backup-location filter uses a case-preserving, position-preserving tuple of