diff --git a/api_test.go b/api_test.go
new file mode 100644
index 000000000..6a9e9c8f2
--- /dev/null
+++ b/api_test.go
@@ -0,0 +1,1012 @@
+package main
+
+import (
+	"encoding/json"
+	"net/http"
+	"net/http/httptest"
+	"reflect"
+	"testing"
+
+	"github.com/blang/semver"
+	"github.com/cloudflare/unsee/mock"
+	"github.com/cloudflare/unsee/models"
+)
+
+type groupTest struct {
+	labels     map[string]string
+	receiver   string
+	alerts     []models.Alert
+	hash       string
+	id         string
+	stateCount map[string]int
+}
+
+var groupTests = []groupTest{
+	groupTest{
+		receiver: "by-name",
+		labels: map[string]string{
+			"alertname": "Memory_Usage_Too_High",
+		},
+		alerts: []models.Alert{
+			models.Alert{
+				Annotations: map[string]string{
+					"alert": "Memory usage exceeding threshold",
+				},
+				Labels: map[string]string{
+					"alertname": "Memory_Usage_Too_High",
+					"cluster":   "prod",
+					"instance":  "server2",
+					"job":       "node_exporter",
+				},
+				State: models.AlertStateActive,
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Receiver: "by-name",
+				Links: map[string]string{
+					"dashboard": "http://localhost/dashboard.html",
+				},
+			},
+		},
+		id:   "099c5ca6d1c92f615b13056b935d0c8dee70f18c",
+		hash: "e20fa82867c7f7929c7302893c303295ec2576ef",
+		stateCount: map[string]int{
+			models.AlertStateActive:      1,
+			models.AlertStateSuppressed:  0,
+			models.AlertStateUnprocessed: 0,
+		},
+	},
+	groupTest{
+		receiver: "by-cluster-service",
+		labels: map[string]string{
+			"alertname": "Memory_Usage_Too_High",
+			"cluster":   "prod",
+		},
+		alerts: []models.Alert{
+			models.Alert{
+				Annotations: map[string]string{
+					"alert": "Memory usage exceeding threshold",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "Memory_Usage_Too_High",
+					"cluster":   "prod",
+					"instance":  "server2",
+					"job":       "node_exporter",
+				},
+				Links: map[string]string{
+					"dashboard": "http://localhost/dashboard.html",
+				},
+				State:    models.AlertStateActive,
+				Receiver: "by-cluster-service",
+			},
+		},
+		hash: "828aa22d70e6d953eee0ea9f2c80f8b3df9ee775",
+		id:   "0b1963665aac588dc4b18e17c7a4f70466c622ea",
+		stateCount: map[string]int{
+			models.AlertStateActive:      1,
+			models.AlertStateSuppressed:  0,
+			models.AlertStateUnprocessed: 0,
+		},
+	},
+	groupTest{
+		receiver: "by-cluster-service",
+		labels: map[string]string{
+			"alertname": "Host_Down",
+			"cluster":   "staging",
+		},
+		alerts: []models.Alert{
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "staging",
+					"instance":  "server3",
+					"job":       "node_ping",
+				},
+				Links:    map[string]string{},
+				State:    models.AlertStateActive,
+				Receiver: "by-cluster-service",
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "staging",
+					"instance":  "server4",
+					"job":       "node_ping",
+				},
+				Links:    map[string]string{},
+				State:    models.AlertStateActive,
+				Receiver: "by-cluster-service",
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "staging",
+					"instance":  "server5",
+					"job":       "node_ping",
+				},
+				Links:    map[string]string{},
+				State:    models.AlertStateActive,
+				Receiver: "by-cluster-service",
+			},
+		},
+		hash: "db53e38245a7afe18f923518146326b6fe53109a",
+		id:   "2d3f39413b41c873cb72e0b8065aa7b8631e983e",
+		stateCount: map[string]int{
+			models.AlertStateActive:      3,
+			models.AlertStateSuppressed:  0,
+			models.AlertStateUnprocessed: 0,
+		},
+	},
+	groupTest{
+		receiver: "by-cluster-service",
+		labels: map[string]string{
+			"alertname": "Host_Down",
+			"cluster":   "dev",
+		},
+		alerts: []models.Alert{
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:   "default",
+						State:  models.AlertStateSuppressed,
+						Source: "localhost/prometheus",
+						Silences: map[string]models.Silence{
+							"168f139d-77e4-41d6-afb5-8fe2cfd0cc9d": models.Silence{
+								ID:        "168f139d-77e4-41d6-afb5-8fe2cfd0cc9d",
+								CreatedBy: "john@example.com",
+								Comment:   "Silenced Host_Down alerts in the dev cluster",
+							},
+						},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "dev",
+					"instance":  "server6",
+					"job":       "node_ping",
+				},
+				Links:    map[string]string{},
+				State:    models.AlertStateSuppressed,
+				Receiver: "by-cluster-service",
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:   "default",
+						State:  models.AlertStateSuppressed,
+						Source: "localhost/prometheus",
+						Silences: map[string]models.Silence{
+							"168f139d-77e4-41d6-afb5-8fe2cfd0cc9d": models.Silence{
+								CreatedBy: "john@example.com",
+								Comment:   "Silenced Host_Down alerts in the dev cluster",
+							},
+							"378eaa69-097d-41c4-a8c2-fe6568c3abfc": models.Silence{
+								CreatedBy: "john@example.com",
+								Comment:   "Silenced server7",
+							},
+						},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "dev",
+					"instance":  "server7",
+					"job":       "node_ping",
+				},
+				Links:    map[string]string{},
+				State:    models.AlertStateSuppressed,
+				Receiver: "by-cluster-service",
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:   "default",
+						State:  models.AlertStateSuppressed,
+						Source: "localhost/prometheus",
+						Silences: map[string]models.Silence{
+							"168f139d-77e4-41d6-afb5-8fe2cfd0cc9d": models.Silence{
+								CreatedBy: "john@example.com",
+								Comment:   "Silenced Host_Down alerts in the dev cluster",
+							},
+						},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "dev",
+					"instance":  "server8",
+					"job":       "node_ping",
+				},
+				Links:    map[string]string{},
+				State:    models.AlertStateSuppressed,
+				Receiver: "by-cluster-service",
+			},
+		},
+		hash: "bcb440cdee1d6f818599cf405c40f3382a4b1229",
+		id:   "3c09c4156e6784dcf6d5b2e1629253798f82909b",
+		stateCount: map[string]int{
+			models.AlertStateActive:      0,
+			models.AlertStateSuppressed:  3,
+			models.AlertStateUnprocessed: 0,
+		},
+	},
+	groupTest{
+		receiver: "by-name",
+		labels: map[string]string{
+			"alertname": "Host_Down",
+		},
+		alerts: []models.Alert{
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "prod",
+					"instance":  "server1",
+					"job":       "node_ping",
+				},
+				State: models.AlertStateActive,
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Receiver: "by-name",
+				Links: map[string]string{
+					"url": "http://localhost/example.html",
+				},
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "prod",
+					"instance":  "server2",
+					"job":       "node_ping",
+				},
+				State: models.AlertStateActive,
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Receiver: "by-name",
+				Links:    map[string]string{},
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "staging",
+					"instance":  "server3",
+					"job":       "node_ping",
+				},
+				State: models.AlertStateActive,
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Receiver: "by-name",
+				Links:    map[string]string{},
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "staging",
+					"instance":  "server4",
+					"job":       "node_ping",
+				},
+				State: models.AlertStateActive,
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Receiver: "by-name",
+				Links:    map[string]string{},
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "staging",
+					"instance":  "server5",
+					"job":       "node_ping",
+				},
+				State: models.AlertStateActive,
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Receiver: "by-name",
+				Links:    map[string]string{},
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "dev",
+					"instance":  "server6",
+					"job":       "node_ping",
+				},
+				State: models.AlertStateSuppressed,
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:   "default",
+						State:  models.AlertStateSuppressed,
+						Source: "localhost/prometheus",
+						Silences: map[string]models.Silence{
+							"168f139d-77e4-41d6-afb5-8fe2cfd0cc9d": models.Silence{
+								CreatedBy: "john@example.com",
+								Comment:   "Silenced Host_Down alerts in the dev cluster",
+							},
+						},
+					},
+				},
+				Receiver: "by-name",
+				Links:    map[string]string{},
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "dev",
+					"instance":  "server7",
+					"job":       "node_ping",
+				},
+				State: models.AlertStateSuppressed,
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:   "default",
+						State:  models.AlertStateSuppressed,
+						Source: "localhost/prometheus",
+						Silences: map[string]models.Silence{
+							"168f139d-77e4-41d6-afb5-8fe2cfd0cc9d": models.Silence{
+								CreatedBy: "john@example.com",
+								Comment:   "Silenced Host_Down alerts in the dev cluster",
+							},
+							"378eaa69-097d-41c4-a8c2-fe6568c3abfc": models.Silence{
+								CreatedBy: "john@example.com",
+								Comment:   "Silenced server7",
+							},
+						},
+					},
+				},
+				Receiver: "by-name",
+				Links:    map[string]string{},
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "dev",
+					"instance":  "server8",
+					"job":       "node_ping",
+				},
+				State: models.AlertStateSuppressed,
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:   "default",
+						State:  models.AlertStateSuppressed,
+						Source: "localhost/prometheus",
+						Silences: map[string]models.Silence{
+							"168f139d-77e4-41d6-afb5-8fe2cfd0cc9d": models.Silence{
+								CreatedBy: "john@example.com",
+								Comment:   "Silenced Host_Down alerts in the dev cluster",
+							},
+						},
+					},
+				},
+				Receiver: "by-name",
+				Links:    map[string]string{},
+			},
+		},
+		id:   "58c6a3467cebc53abe68ecbe8643ce478c5a1573",
+		hash: "68d0ac6e27b890e0f854611963b03b51b37242cf",
+		stateCount: map[string]int{
+			models.AlertStateActive:      5,
+			models.AlertStateSuppressed:  3,
+			models.AlertStateUnprocessed: 0,
+		},
+	},
+	groupTest{
+		receiver: "by-cluster-service",
+		labels: map[string]string{
+			"alertname": "Free_Disk_Space_Too_Low",
+			"cluster":   "staging",
+		},
+		alerts: []models.Alert{
+			models.Alert{
+				Annotations: map[string]string{
+					"alert": "Less than 10% disk space is free",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "Free_Disk_Space_Too_Low",
+					"cluster":   "staging",
+					"instance":  "server5",
+					"job":       "node_exporter",
+				},
+				Links: map[string]string{
+					"dashboard": "http://localhost/dashboard.html",
+				},
+				State:    models.AlertStateActive,
+				Receiver: "by-cluster-service",
+			},
+		},
+		hash: "4917eff113e7d22d7f1e5dba1e6dbb6d7f0969ad",
+		id:   "8ca8151d9e30baba2334507dca53e16b7be93c5e",
+		stateCount: map[string]int{
+			models.AlertStateActive:      1,
+			models.AlertStateSuppressed:  0,
+			models.AlertStateUnprocessed: 0,
+		},
+	},
+	groupTest{
+		receiver: "by-cluster-service",
+		labels: map[string]string{
+			"alertname": "Host_Down",
+			"cluster":   "prod",
+		},
+		alerts: []models.Alert{
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "prod",
+					"instance":  "server1",
+					"job":       "node_ping",
+				},
+				Links: map[string]string{
+					"url": "http://localhost/example.html",
+				},
+				State:    models.AlertStateActive,
+				Receiver: "by-cluster-service",
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "Host_Down",
+					"cluster":   "prod",
+					"instance":  "server2",
+					"job":       "node_ping",
+				},
+				Links:    map[string]string{},
+				State:    models.AlertStateActive,
+				Receiver: "by-cluster-service",
+			},
+		},
+		hash: "eee0a9960be86ab7308f50a8ff438caed5cf8540",
+		id:   "98c1a53d0f71af9c734c9180697383f3b8aff80f",
+		stateCount: map[string]int{
+			models.AlertStateActive:      2,
+			models.AlertStateSuppressed:  0,
+			models.AlertStateUnprocessed: 0,
+		},
+	},
+	groupTest{
+		receiver: "by-name",
+		labels: map[string]string{
+			"alertname": "HTTP_Probe_Failed",
+		},
+		alerts: []models.Alert{
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+					"help":    "Example help annotation",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:   "default",
+						State:  models.AlertStateSuppressed,
+						Source: "localhost/prometheus",
+						Silences: map[string]models.Silence{
+							"0804764c-6163-4c64-b0a9-08feebe2db4b": models.Silence{
+								CreatedBy: "john@example.com",
+								Comment:   "Silenced instance",
+							},
+						},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "HTTP_Probe_Failed",
+					"cluster":   "dev",
+					"instance":  "web1",
+					"job":       "node_exporter",
+				},
+				Links: map[string]string{
+					"url": "http://localhost/example.html",
+				},
+				State:    models.AlertStateSuppressed,
+				Receiver: "by-name",
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "HTTP_Probe_Failed",
+					"cluster":   "dev",
+					"instance":  "web2",
+					"job":       "node_exporter",
+				},
+				Links:    map[string]string{},
+				State:    models.AlertStateActive,
+				Receiver: "by-name",
+			},
+		},
+		hash: "cc1b20a6b0ded9265ab96699638d844a4c992614",
+		id:   "bc4845fec77585cdfebe946234279d785ca93891",
+		stateCount: map[string]int{
+			models.AlertStateActive:      1,
+			models.AlertStateSuppressed:  1,
+			models.AlertStateUnprocessed: 0,
+		},
+	},
+	groupTest{
+		receiver: "by-name",
+		labels: map[string]string{
+			"alertname": "Free_Disk_Space_Too_Low",
+		},
+		alerts: []models.Alert{
+			models.Alert{
+				Annotations: map[string]string{
+					"alert": "Less than 10% disk space is free",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "Free_Disk_Space_Too_Low",
+					"cluster":   "staging",
+					"instance":  "server5",
+					"job":       "node_exporter",
+				},
+				Links: map[string]string{
+					"dashboard": "http://localhost/dashboard.html",
+				},
+				State:    models.AlertStateActive,
+				Receiver: "by-name",
+			},
+		},
+		hash: "a596259a6ff3d8a5fdabf1a91c6d2b7e680d05d7",
+		id:   "bf78806d2a80b1c8150c1391669813722428e858",
+		stateCount: map[string]int{
+			models.AlertStateActive:      1,
+			models.AlertStateSuppressed:  0,
+			models.AlertStateUnprocessed: 0,
+		},
+	},
+	groupTest{
+		receiver: "by-cluster-service",
+		labels: map[string]string{
+			"alertname": "HTTP_Probe_Failed",
+			"cluster":   "dev",
+		},
+		alerts: []models.Alert{
+			models.Alert{
+				Annotations: map[string]string{
+					"help":    "Example help annotation",
+					"summary": "Example summary",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:   "default",
+						State:  models.AlertStateSuppressed,
+						Source: "localhost/prometheus",
+						Silences: map[string]models.Silence{
+							"0804764c-6163-4c64-b0a9-08feebe2db4b": models.Silence{
+								CreatedBy: "john@example.com",
+								Comment:   "Silenced instance",
+							},
+						},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "HTTP_Probe_Failed",
+					"cluster":   "dev",
+					"instance":  "web1",
+					"job":       "node_exporter",
+				},
+				Links: map[string]string{
+					"url": "http://localhost/example.html",
+				},
+				State:    models.AlertStateSuppressed,
+				Receiver: "by-cluster-service",
+			},
+			models.Alert{
+				Annotations: map[string]string{
+					"summary": "Example summary",
+				},
+				Alertmanager: []models.AlertmanagerInstance{
+					models.AlertmanagerInstance{
+						Name:     "default",
+						State:    models.AlertStateActive,
+						Source:   "localhost/prometheus",
+						Silences: map[string]models.Silence{},
+					},
+				},
+				Labels: map[string]string{
+					"alertname": "HTTP_Probe_Failed",
+					"cluster":   "dev",
+					"instance":  "web2",
+					"job":       "node_exporter",
+				},
+				Links:    map[string]string{},
+				State:    models.AlertStateActive,
+				Receiver: "by-cluster-service",
+			},
+		},
+		hash: "1dd655dc8ac8ed51aca51a702e70b1a2f442f434",
+		id:   "ecefc3705b1ab4e4c3283c879540be348d2d9dce",
+		stateCount: map[string]int{
+			models.AlertStateActive:      1,
+			models.AlertStateSuppressed:  1,
+			models.AlertStateUnprocessed: 0,
+		},
+	},
+}
+
+var countsMap = models.LabelsCountMap{
+	"@receiver": map[string]int{
+		"by-cluster-service": 12,
+		"by-name":            12,
+	},
+	"@state": map[string]int{
+		"active":     16,
+		"suppressed": 8,
+	},
+	"alertname": map[string]int{
+		"Free_Disk_Space_Too_Low": 2,
+		"HTTP_Probe_Failed":       4,
+		"Host_Down":               16,
+		"Memory_Usage_Too_High":   2,
+	},
+	"cluster": map[string]int{
+		"dev":     10,
+		"prod":    6,
+		"staging": 8,
+	},
+	"instance": map[string]int{
+		"server1": 2,
+		"server2": 4,
+		"server3": 2,
+		"server4": 2,
+		"server5": 4,
+		"server6": 2,
+		"server7": 2,
+		"server8": 2,
+		"web1":    2,
+		"web2":    2,
+	},
+	"job": map[string]int{
+		"node_exporter": 8,
+		"node_ping":     16,
+	},
+}
+
+var filtersExpected = []models.Filter{
+	models.Filter{
+		Hits:    24,
+		IsValid: true,
+	},
+}
+
+func compareAlertGroups(testCase groupTest, group models.AlertGroup) bool {
+	if testCase.receiver != group.Receiver {
+		return false
+	}
+	if len(testCase.labels) != len(group.Labels) {
+		return false
+	}
+	for key, val := range testCase.labels {
+		v, found := group.Labels[key]
+		if !found {
+			return false
+		}
+		if val != v {
+			return false
+		}
+	}
+	return true
+}
+
+func compareAlerts(expectedAlert, gotAlert models.Alert) bool {
+	if gotAlert.Receiver != expectedAlert.Receiver {
+		return false
+	}
+	if len(gotAlert.Labels) != len(expectedAlert.Labels) {
+		return false
+	}
+	for key, val := range expectedAlert.Labels {
+		v, found := gotAlert.Labels[key]
+		if !found {
+			return false
+		}
+		if val != v {
+			return false
+		}
+	}
+	return true
+}
+
+func testAlert(version string, t *testing.T, expectedAlert, gotAlert models.Alert) {
+	if gotAlert.Receiver != expectedAlert.Receiver {
+		t.Errorf("[%s] Expected '%s' receiver but got '%s' on alert labels=%v",
+			version, expectedAlert.Receiver, gotAlert.Receiver, expectedAlert.Labels)
+	}
+	if gotAlert.State != expectedAlert.State {
+		t.Errorf("[%s] Expected state '%s' but got '%s' on alert receiver='%s' labels=%v",
+			version, expectedAlert.State, gotAlert.State, gotAlert.Receiver, expectedAlert.Labels)
+	}
+	if !reflect.DeepEqual(gotAlert.Annotations, expectedAlert.Annotations) {
+		t.Errorf("[%s] Annotation mismatch on alert receiver='%s' labels=%v, expected %v but got %v",
+			version, expectedAlert.Receiver, expectedAlert.Labels, expectedAlert.Annotations, gotAlert.Annotations)
+	}
+	if !reflect.DeepEqual(gotAlert.Labels, expectedAlert.Labels) {
+		t.Errorf("[%s] Labels mismatch on alert receiver='%s', expected labels=%v but got %v",
+			version, expectedAlert.Receiver, expectedAlert.Labels, gotAlert.Labels)
+	}
+	if !reflect.DeepEqual(gotAlert.Links, expectedAlert.Links) {
+		t.Errorf("[%s] Links mismatch on alert receiver='%s' labels=%v expected %v but got %v",
+			version, expectedAlert.Receiver, expectedAlert.Labels, expectedAlert.Links, gotAlert.Links)
+	}
+	if len(gotAlert.Alertmanager) != len(expectedAlert.Alertmanager) {
+		t.Errorf("[%s] Expected %d alertmanager instances but got %d on alert receiver='%s' labels=%v",
+			version, len(expectedAlert.Alertmanager), len(gotAlert.Alertmanager), gotAlert.Receiver, expectedAlert.Labels)
+	}
+	for _, expectedAM := range expectedAlert.Alertmanager {
+		found := false
+		for _, gotAM := range gotAlert.Alertmanager {
+			if gotAM.Name == expectedAM.Name {
+				found = true
+				if gotAM.State != expectedAM.State {
+					t.Errorf("[%s] Expected alertmanager '%s' to have state '%s' but got '%s' on alert receiver='%s' labels=%v",
+						version, expectedAM.Name, expectedAM.State, gotAM.State, gotAlert.Receiver, expectedAlert.Labels)
+				}
+				if gotAM.Source != expectedAM.Source {
+					t.Errorf("[%s] Expected alertmanager '%s' to have source '%s' but got '%s' on alert receiver='%s' labels=%v",
+						version, expectedAM.Name, expectedAM.Source, gotAM.Source, gotAlert.Receiver, expectedAlert.Labels)
+				}
+				// multiple silences only work for >=0.6.1
+				versionRange := semver.MustParseRange(">=0.6.1")
+				if versionRange(semver.MustParse(version)) {
+					if len(gotAM.Silences) != len(expectedAM.Silences) {
+						t.Errorf("[%s] Expected alertmanager '%s' to have %d silences but got %d on alert receiver='%s' labels=%v",
+							version, expectedAM.Name, len(expectedAM.Silences), len(gotAM.Silences), gotAlert.Receiver, expectedAlert.Labels)
+					}
+					for _, es := range expectedAM.Silences {
+						foundSilence := false
+						for _, gs := range gotAM.Silences {
+							if es.Comment == gs.Comment && es.CreatedBy == gs.CreatedBy && es.JiraID == gs.JiraID && es.JiraURL == es.JiraURL {
+								foundSilence = true
+							}
+						}
+						if !foundSilence {
+							t.Errorf("[%s] Silence %v not found on alertmanager '%s' on alert receiver='%s' labels=%v",
+								version, es, expectedAM.Name, expectedAlert.Receiver, expectedAlert.Labels)
+						}
+					}
+				}
+				break
+			}
+		}
+		if !found {
+			t.Errorf("[%s] Alertmanager instances '%s' not found on alert receiver='%s' labels=%v",
+				version, expectedAM.Name, gotAlert.Receiver, expectedAlert.Labels)
+		}
+	}
+}
+
+func testAlertGroup(version string, t *testing.T, testCase groupTest, group models.AlertGroup) {
+	if testCase.hash != group.Hash {
+		// FIXME this is different per mock version due to startsAt / endsAt
+		// t.Errorf("[%s] Alert group.Hash mismatch, expected '%s' but got '%s' for group %v",
+		// version, testCase.hash, group.Hash, group.Labels)
+	}
+	if testCase.id != group.ID {
+		t.Errorf("[%s] Alert group.ID mismatch, expected '%s' but got '%s' for group %v",
+			version, testCase.id, group.ID, group.Labels)
+	}
+	for key, val := range testCase.stateCount {
+		v, found := group.StateCount[key]
+		if !found {
+			t.Errorf("[%s] Expected group.StateCount[%s]=%d not found", version, key, val)
+		} else if v != val {
+			t.Errorf("[%s] group.StateCount[%s] mismatch, expected %d but got %d", version, key, val, v)
+		}
+	}
+	if len(testCase.alerts) != len(group.Alerts) {
+		t.Errorf("[%s] Expected %d alert(s) but got %d on alert group receiver='%s' labels=%v",
+			version, len(testCase.alerts), len(group.Alerts), testCase.receiver, testCase.labels)
+	}
+	for _, expectedAlert := range testCase.alerts {
+		alertFound := false
+		for _, alert := range group.Alerts {
+			match := compareAlerts(expectedAlert, alert)
+			if match {
+				alertFound = true
+				testAlert(version, t, expectedAlert, alert)
+			}
+		}
+		if !alertFound {
+			t.Errorf("[%s] Expected alert receiver='%s' labels=%v not found in group: %v",
+				version, expectedAlert.Receiver, expectedAlert.Labels, group.Alerts)
+		}
+	}
+}
+
+func TestVerifyAllGroups(t *testing.T) {
+	mockConfig()
+	for _, version := range mock.ListAllMocks() {
+		mockAlerts(version)
+		r := ginTestEngine()
+		req, _ := http.NewRequest("GET", "/alerts.json", nil)
+		resp := httptest.NewRecorder()
+		r.ServeHTTP(resp, req)
+		if resp.Code != http.StatusOK {
+			t.Errorf("GET /alerts.json returned status %d", resp.Code)
+		}
+
+		ur := models.AlertsResponse{}
+		json.Unmarshal(resp.Body.Bytes(), &ur)
+
+		if len(ur.AlertGroups) != len(groupTests) {
+			t.Errorf("[%s] Got %d alert(s) in response, expected %d",
+				version, len(ur.AlertGroups), len(groupTests))
+		}
+		for _, testCase := range groupTests {
+			groupFound := false
+			for _, group := range ur.AlertGroups {
+				if compareAlertGroups(testCase, group) {
+					groupFound = true
+					testAlertGroup(version, t, testCase, group)
+				}
+			}
+			if !groupFound {
+				t.Errorf("[%s] Expected alert group not found receiver='%s' labels=%v",
+					version, testCase.receiver, testCase.labels)
+			}
+		}
+		for key, expectedCounts := range countsMap {
+			gotCounts, foundCounts := ur.Counters[key]
+			if !foundCounts {
+				t.Errorf("[%s] Counters missing for key '%s'", version, key)
+			} else if !reflect.DeepEqual(expectedCounts, gotCounts) {
+				t.Errorf("[%s] Counters mismatch for key '%s', expected %v but got %v",
+					version, key, expectedCounts, gotCounts)
+			}
+		}
+		if !reflect.DeepEqual(ur.Filters, filtersExpected) {
+			t.Errorf("[%s] Filters mismatch, expected %v but got %v", version, filtersExpected, ur.Filters)
+		}
+	}
+}
diff --git a/views_test.go b/views_test.go
index 7a83066e4..b0a9956e4 100644
--- a/views_test.go
+++ b/views_test.go
@@ -141,12 +141,15 @@ func TestAlerts(t *testing.T) {
 		if ur.Upstreams.Counters.Total == 0 {
 			t.Errorf("[%s] No instances in upstream counter: %v", version, ur.Upstreams.Counters)
 		}
-		if len(ur.Upstreams.Instances) == 0 {
-			t.Errorf("[%s] No instances in upstream status: %v", version, ur.Upstreams.Instances)
+		if ur.Upstreams.Counters.Healthy == 0 {
+			t.Errorf("[%s] No healthy instances in upstream counter: %v", version, ur.Upstreams.Counters)
 		}
 		if ur.Upstreams.Counters.Failed > 0 {
 			t.Errorf("[%s] %d error(s) in upstream status: %v", version, ur.Upstreams.Counters.Failed, ur.Upstreams)
 		}
+		if len(ur.Upstreams.Instances) == 0 {
+			t.Errorf("[%s] No instances in upstream status: %v", version, ur.Upstreams.Instances)
+		}
 		if ur.Status != "success" {
 			t.Errorf("[%s] Invalid status in response: %s", version, ur.Status)
 		}