ccl, metrics: Fill in chart catalog with missed metrics

This change adds missing metrics to chart catalog. Release note: None
cockroachdb · May 24, 2022 · 2004f02 · 2004f02
1 parent 3cb6d73
commit 2004f02
Show file tree

Hide file tree

Showing 5 changed files with 136 additions and 27 deletions.
diff --git a/pkg/ccl/serverccl/chart_catalog_test.go b/pkg/ccl/serverccl/chart_catalog_test.go
@@ -91,18 +91,8 @@ func TestChartCatalogMetrics(t *testing.T) {
 
 // deleteSeenMetrics removes all metrics in a section from the metricMetadata map.
 func deleteSeenMetrics(c *catalog.ChartSection, metadata map[string]metric.Metadata, t *testing.T) {
-	// if c.Title == "SQL" {
-	// 	t.Log(c)
-	// }
 	for _, x := range c.Charts {
-		if x.Title == "Connections" || x.Title == "Byte I/O" {
-			t.Log(x)
-		}
-
 		for _, metric := range x.Metrics {
-			if metric.Name == "sql.new_conns" || metric.Name == "sql.bytesin" {
-				t.Logf("found %v\n", metric.Name)
-			}
 			_, ok := metadata[metric.Name]
 			if ok {
 				delete(metadata, metric.Name)

diff --git a/pkg/jobs/schedule_metrics.go b/pkg/jobs/schedule_metrics.go
@@ -42,6 +42,12 @@ type SchedulerMetrics struct {
 	// Number of schedules that are malformed: that is, the schedules
 	// we cannot parse, or even attempt to execute.
 	NumMalformedSchedules *metric.Gauge
+	// Number of started schedules
+	NumTotalStartedSchedules *metric.Gauge
+	// Number of succeeded schedules
+	NumTotalSucceededSchedules *metric.Gauge
+	// Number of failed schedules
+	NumTotalFailedSchedules *metric.Gauge
 }
 
 // MakeSchedulerMetrics returns metrics for scheduled job daemon.
@@ -81,6 +87,27 @@ func MakeSchedulerMetrics() SchedulerMetrics {
 			Measurement: "Schedules",
 			Unit:        metric.Unit_COUNT,
 		}),
+
+		NumTotalStartedSchedules: metric.NewGauge(metric.Metadata{
+			Name:        "schedules.total.started",
+			Help:        "Number of started schedules",
+			Measurement: "Schedules",
+			Unit:        metric.Unit_COUNT,
+		}),
+
+		NumTotalSucceededSchedules: metric.NewGauge(metric.Metadata{
+			Name:        "schedules.total.succeeded",
+			Help:        "Number of succeeded schedules",
+			Measurement: "Schedules",
+			Unit:        metric.Unit_COUNT,
+		}),
+
+		NumTotalFailedSchedules: metric.NewGauge(metric.Metadata{
+			Name:        "schedules.total.failed",
+			Help:        "Number of failed schedules",
+			Measurement: "Schedules",
+			Unit:        metric.Unit_COUNT,
+		}),
 	}
 }
 

diff --git a/pkg/ts/catalog/catalog_generator.go b/pkg/ts/catalog/catalog_generator.go
@@ -200,6 +200,13 @@ var chartCatalog = []ChartSection{
 		Description:     `Your cluster executes various background jobs, as well as scheduled jobs`,
 		Level:           0,
 	},
+	{
+		Title:           Tenants,
+		LongTitle:       Tenants,
+		CollectionTitle: "tenants-all",
+		Description:     `TBD`, // TODO (koorosh): provide a meaningful description.
+		Level:           0,
+	},
 }
 
 var catalogGenerated = false
@@ -215,6 +222,7 @@ var catalogKey = map[string]int{
 	StorageLayer:       5,
 	Timeseries:         6,
 	Jobs:               7,
+	Tenants:            8,
 }
 
 // unitsKey converts between metric.Unit and catalog.AxisUnits which is

diff --git a/pkg/ts/catalog/chart_catalog.go b/pkg/ts/catalog/chart_catalog.go
@@ -1337,23 +1337,82 @@ var charts = []sectionDescription{
 				},
 			},
 			{
-				Title: "Assorted (Counters)",
+				Title: "Frontier Updates",
+				Metrics: []string{
+					"changefeed.frontier_updates",
+				},
+			},
+			{
+				Title: "Ingested Events",
+				Metrics: []string{
+					"streaming.events_ingested",
+					"streaming.resolved_events_ingested",
+				},
+			},
+			{
+				Title: "Flushes",
+				Metrics: []string{
+					"streaming.flushes",
+				},
+			},
+			{
+				Title: "Ingested Bytes",
+				Metrics: []string{
+					"streaming.ingested_bytes",
+				},
+			},
+			{
+				Title: "Flushed Bytes",
+				Metrics: []string{
+					"changefeed.flushed_bytes",
+				},
+			},
+			{
+				Title: "Forwarded Resolved Messages",
+				Metrics: []string{
+					"changefeed.forwarded_resolved_messages",
+				},
+			},
+			{
+				Title: "Messages Size",
+				Metrics: []string{
+					"changefeed.message_size_hist",
+				},
+			},
+			{
+				Title: "Commits Latency",
+				Metrics: []string{
+					"changefeed.commit_latency",
+				},
+			},
+			{
+				Title: "Time spent",
+				Metrics: []string{
+					"changefeed.checkpoint_hist_nanos",
+					"changefeed.flush_hist_nanos",
+					"changefeed.sink_batch_hist_nanos",
+				},
+			},
+			{
+				Title: "Backfill",
 				Metrics: []string{
 					"changefeed.backfill_count",
 					"changefeed.backfill_pending_ranges",
+				},
+			},
+			{
+				Title: "Time Spent Waiting",
+				Metrics: []string{
 					"changefeed.buffer_pushback_nanos",
+					"changefeed.queue_time_nanos",
+				},
+			},
+			{
+				Title: "Throttled Time",
+				Metrics: []string{
 					"changefeed.bytes.messages_pushback_nanos",
-					"changefeed.checkpoint_hist_nanos",
-					"changefeed.commit_latency",
-					"changefeed.flush.messages_pushback_nanos",
-					"changefeed.flush_hist_nanos",
-					"changefeed.flushed_bytes",
-					"changefeed.forwarded_resolved_messages",
-					"changefeed.frontier_updates",
-					"changefeed.message_size_hist",
 					"changefeed.messages.messages_pushback_nanos",
-					"changefeed.queue_time_nanos",
-					"changefeed.sink_batch_hist_nanos",
+					"changefeed.flush.messages_pushback_nanos",
 				},
 			},
 		},
@@ -2787,6 +2846,12 @@ var charts = []sectionDescription{
 					"schedules.BACKUP.failed",
 				},
 			},
+			{
+				Title: "Last Completed Backups",
+				Metrics: []string{
+					"schedules.BACKUP.last-completed-time",
+				},
+			},
 		},
 	},
 	{
@@ -3118,19 +3183,34 @@ var charts = []sectionDescription{
 		},
 	},
 	{
-		Organization: [][]string{{Tenants}},
+		Organization: [][]string{{Tenants, "Consumption"}},
 		Charts: []chartDescription{
 			{
-				Title: "Assorted",
+				Title: "Total number of KV read/write requests (Gauge_Count_Requests)",
 				Metrics: []string{
-					"tenant.consumption.kv_request_units",
-					"tenant.consumption.pgwire_egress_bytes",
-					"tenant.consumption.read_bytes",
+					"tenant.consumption.write_requests",
 					"tenant.consumption.read_requests",
+				},
+			},
+			{
+				Title: "Total RU consumption (Gauge_Count_Request_units)",
+				Metrics: []string{
 					"tenant.consumption.request_units",
+					"tenant.consumption.kv_request_units",
+				},
+			},
+			{
+				Title: "Total number of bytes written to KV (Gauge_Count_CPU_seconds)",
+				Metrics: []string{
 					"tenant.consumption.sql_pods_cpu_seconds",
+				},
+			},
+			{
+				Title: "Total number of bytes (Gauge_count_bytes)",
+				Metrics: []string{
 					"tenant.consumption.write_bytes",
-					"tenant.consumption.write_requests",
+					"tenant.consumption.pgwire_egress_bytes",
+					"tenant.consumption.read_bytes",
 				},
 			},
 		},

diff --git a/pkg/ts/catalog/metrics.go b/pkg/ts/catalog/metrics.go
@@ -93,6 +93,10 @@ var histogramMetricsNames = map[string]struct{}{
 	"raft.process.applycommitted.latency":       {},
 	"sql.stats.txn_stats_collection.duration":   {},
 	"rebalancing.l0_sublevels_histogram":        {},
+	"changefeed.admit_latency":                  {},
+	"changefeed.message_size_hist":              {},
+	"changefeed.commit_latency":                 {},
+	"changefeed.sink_batch_hist_nanos":          {},
 }
 
 func allInternalTSMetricsNames() []string {