Fix native health polling and isolate file operation traffic

docs/native-hosts.md

@@ -5,6 +5,13 @@ instance never uses Komodo or a filesystem mounted into the controller. The
 controller owns desired state, campaigns, collections, safety queues and history;
 the agent owns local execution and its durable operation journal.
 
+Fleet health checks query RCON even when container inspection is unavailable. A
+fresh successful RCON reply can establish game health while container state stays
+unknown; confirmed stopped, unhealthy, and restarting states still take precedence.
+Agent 1.4.4 retries transient failures of the read-only `status` query up to three
+times, with 250 ms and 500 ms delays. Commands that change the game are not retried.
+Status replies still expire normally, and native timeout diagnostics are retained.
+
 ## Install and launch
 
 1. Run the controller with persistent `/data` storage. Set `EC_PUBLIC_URL` to its
@@ -479,9 +486,14 @@ of which task delayed a previous deployment.
 
 Campaign, beta and maintenance observers run independently of provisioning.
 New requests and completed agent commands wake provisioning; an arrival during a
-pass schedules another pass. File drift scans use bounded batches of four reads
-per scan to reduce agent polling round trips. Startup and disruption safety gates
-remain enforced at command admission.
+pass schedules another pass. Routine file drift scans run every five minutes;
+manual checks bypass that interval. Overlapping identical native observations
+share one operation until it completes, without caching results for later checks.
+Agent 1.4.5 reserves two observation workers for instance health and read-only RCON,
+and one for files, logs, and host inventory. Background reads arrive in batches of
+up to four. File receipt uploads release the instance lock before contacting the
+controller. Older agents retain the shared read lane until upgraded. Startup and
+disruption safety gates remain enforced at command admission.
 
 Older completed operations whose progress snapshots were already discarded
 cannot recover missing check results from this change. Their surviving events