From ce1607694e08292d89e48ce2b6e31f725fd464c3 Mon Sep 17 00:00:00 2001
From: rcourtman <courtmanr@gmail.com>
Date: Thu, 28 May 2026 14:11:22 +0100
Subject: [PATCH] Run ceph pool alerts for agent-sourced clusters

Fixes #1341

When a Pulse host-agent reports Ceph data, ApplyHostReport upserts the
cluster into state but only the Proxmox-API polling path ran
cephPoolAlertStorageTargets. Users with agent-reported Ceph (instance
prefix "agent:hostname") saved per-pool overrides under
agent-prefixed IDs that the alert manager never evaluated. The
threshold appeared to save (and showed Custom in the UI), but the
polling cycle was checking a different storage ID, so the alert
silently stayed dormant. Run CheckStorage for each pool right after
the agent upsert so the override key actually drives evaluation.
---
 internal/monitoring/monitor.go                | 11 +++
 .../monitoring/monitor_host_agents_test.go    | 76 +++++++++++++++++++
 2 files changed, 87 insertions(+)

diff --git a/internal/monitoring/monitor.go b/internal/monitoring/monitor.go
index 2cd8be455..d8bdadb72 100644
--- a/internal/monitoring/monitor.go
+++ b/internal/monitoring/monitor.go
@@ -3659,6 +3659,17 @@ func (m *Monitor) ApplyHostReport(report agentshost.Report, tokenRecord *config.
 			Str("health", cephCluster.Health).
 			Int("osds", cephCluster.NumOSDs).
 			Msg("Updated Ceph cluster from host agent")
+
+		// #1341: the agent-reported Ceph cluster used to land in state but
+		// never get evaluated for pool alerts. Only the Proxmox-API polling
+		// path ran cephPoolAlertStorageTargets, so overrides set against
+		// agent-prefixed pool IDs (e.g. agent:hostname-ceph-pool-foo) were
+		// dormant. Run the alert check here so agent-sourced pools fire.
+		if m.alertManager != nil {
+			for _, storage := range cephPoolAlertStorageTargets(cephCluster) {
+				m.alertManager.CheckStorage(storage)
+			}
+		}
 	}
 
 	if m.alertManager != nil {
diff --git a/internal/monitoring/monitor_host_agents_test.go b/internal/monitoring/monitor_host_agents_test.go
index b647749f6..09b665fca 100644
--- a/internal/monitoring/monitor_host_agents_test.go
+++ b/internal/monitoring/monitor_host_agents_test.go
@@ -1160,3 +1160,79 @@ func TestApplyHostReport_FallbackIdentifier(t *testing.T) {
 		t.Errorf("expected hostname 'fallback-host', got %q", host.Hostname)
 	}
 }
+
+// Regression for #1341: when Ceph is reported by a Pulse host-agent (not the
+// Proxmox API), the agent-sourced cluster used to land in state but the
+// pool alert evaluation only ran in the Proxmox-API polling path. Per-pool
+// overrides keyed against the agent-prefixed pool ID (e.g.
+// `agent:hostname-ceph-pool-name`) were silently dormant. Verify the
+// agent path now fires CheckStorage for each pool so overrides match.
+func TestApplyHostReportFiresCephPoolAlertsForAgentSourcedCluster(t *testing.T) {
+	monitor := &Monitor{
+		state:             models.NewState(),
+		alertManager:      alerts.NewManager(),
+		hostTokenBindings: make(map[string]string),
+		config:            &config.Config{},
+		rateTracker:       NewRateTracker(),
+	}
+	t.Cleanup(func() { monitor.alertManager.Stop() })
+
+	overrideTrigger := alerts.HysteresisThreshold{Trigger: 50, Clear: 45}
+	cfg := monitor.alertManager.GetConfig()
+	cfg.MinimumDelta = 0
+	if cfg.TimeThresholds == nil {
+		cfg.TimeThresholds = make(map[string]int)
+	}
+	cfg.TimeThresholds["storage"] = 0
+	cfg.StorageDefault = alerts.HysteresisThreshold{Trigger: 95, Clear: 90}
+	cfg.Overrides = map[string]alerts.ThresholdConfig{
+		"agent:pve5-ceph-pool-data_replication": {Usage: &overrideTrigger},
+	}
+	monitor.alertManager.UpdateConfig(cfg)
+
+	report := agentshost.Report{
+		Agent: agentshost.AgentInfo{
+			ID:              "agent-ceph",
+			Version:         "1.0.0",
+			IntervalSeconds: 30,
+		},
+		Host: agentshost.HostInfo{
+			ID:       "pve5-host",
+			Hostname: "pve5",
+			Platform: "linux",
+		},
+		Timestamp: time.Now().UTC(),
+		Ceph: &agentshost.CephCluster{
+			FSID:   "ceph-fsid-1341",
+			Health: agentshost.CephHealth{Status: "HEALTH_OK"},
+			Pools: []agentshost.CephPool{
+				{
+					ID:             2,
+					Name:           "data_replication",
+					BytesUsed:      611,
+					BytesAvailable: 389,
+					PercentUsed:    61.1,
+				},
+			},
+		},
+	}
+
+	if _, err := monitor.ApplyHostReport(report, nil); err != nil {
+		t.Fatalf("ApplyHostReport: %v", err)
+	}
+
+	active := monitor.alertManager.GetActiveAlerts()
+	found := false
+	for _, alert := range active {
+		if alert.ID == "agent:pve5-ceph-pool-data_replication-usage" {
+			found = true
+			if alert.Threshold != 50 {
+				t.Fatalf("Ceph pool alert fired but threshold = %.1f, want 50 (override)", alert.Threshold)
+			}
+			break
+		}
+	}
+	if !found {
+		t.Fatalf("expected agent-sourced Ceph pool alert to fire under override at 61.1%% usage; got %d active alerts", len(active))
+	}
+}