Add metrics from information_schema.tables

Collect metrics from information_schema.tables
* Allow filtered list of databases to avoid performance problems on
  MySQL < 5.6
* Default to all databases
* Include extra labels in `table_version`, useful for detecting crashed
  tables `{engine="NONE"}`

Author: SuperQ

README.md

@@ -17,6 +17,8 @@ Name                                       | Description
 collect.auto_increment.columns             | Collect auto_increment columns and max values from information_schema.
 collect.binlog_size                        | Compute the size of all binlog files combined (as specified by "SHOW MASTER LOGS")
 collect.info_schema.userstats              | If running with userstat=1, set to true to collect user statistics.
+collect.info_schema.tables                 | Collect metrics from information_schema.tables.
+collect.info_schema.tables.databases       | The list of databases to collect table stats for, or '`*`' for all.
 collect.perf_schema.eventsstatements       | Collect metrics from performance_schema.events_statements_summary_by_digest.
 collect.perf_schema.eventsstatements.limit | Limit the number of events statements digests by response time. (default: 250)
 collect.perf_schema.eventsstatements.digest_text_limit | Maximum length of the normalized statement text. (default: 120)

mysqld_exporter.go

@@ -30,6 +30,14 @@ var (
 		"collect.info_schema.processlist", false,
 		"Collect current thread state counts from the information_schema.processlist",
 	)
+	collectTableSchema = flag.Bool(
+		"collect.info_schema.tables", true,
+		"Collect metrics from information_schema.tables",
+	)
+	tableSchemaDatabases = flag.String(
+		"collect.info_schema.tables.databases", "*",
+		"The list of databases to collect table stats for, or '*' for all",
+	)
 	collectGlobalStatus = flag.Bool(
 		"collect.global_status", true,
 		"Collect from SHOW GLOBAL STATUS",
@@ -204,7 +212,29 @@ const (
 		  COUNT_MISC, SUM_TIMER_MISC
 		  FROM performance_schema.file_summary_by_event_name
 		`
-	userStatQuery = `SELECT * FROM information_schema.USER_STATISTICS`
+	userStatQuery    = `SELECT * FROM information_schema.USER_STATISTICS`
+	tableSchemaQuery = `
+		SELECT
+		    TABLE_SCHEMA,
+		    TABLE_NAME,
+		    TABLE_TYPE,
+		    ifnull(ENGINE, 'NONE') as ENGINE,
+		    ifnull(VERSION, '0') as VERSION,
+		    ifnull(ROW_FORMAT, 'NONE') as ROW_FORMAT,
+		    ifnull(TABLE_ROWS, '0') as TABLE_ROWS,
+		    ifnull(DATA_LENGTH, '0') as DATA_LENGTH,
+		    ifnull(INDEX_LENGTH, '0') as INDEX_LENGTH,
+		    ifnull(DATA_FREE, '0') as DATA_FREE,
+		    ifnull(CREATE_OPTIONS, 'NONE') as CREATE_OPTIONS
+		  FROM information_schema.tables
+		  WHERE TABLE_SCHEMA = '%s'
+		`
+	dbListQuery = `
+		SELECT
+		    SCHEMA_NAME
+		  FROM information_schema.schemata
+		  WHERE SCHEMA_NAME NOT IN ('mysql', 'performance_schema', 'information_schema')
+		`
 )
 
 // landingPage contains the HTML served at '/'.
@@ -255,6 +285,21 @@ var (
 		"The max value of an auto_increment column from information_schema.",
 		[]string{"schema", "table", "column"}, nil,
 	)
+	infoSchemaTablesVersionDesc = prometheus.NewDesc(
+		prometheus.BuildFQName(namespace, informationSchema, "table_version"),
+		"The version number of the table's .frm file",
+		[]string{"schema", "table", "type", "engine", "row_format", "create_options"}, nil,
+	)
+	infoSchemaTablesRowsDesc = prometheus.NewDesc(
+		prometheus.BuildFQName(namespace, informationSchema, "table_rows"),
+		"The estimated number of rows in the table from information_schema.tables",
+		[]string{"schema", "table"}, nil,
+	)
+	infoSchemaTablesSizeDesc = prometheus.NewDesc(
+		prometheus.BuildFQName(namespace, informationSchema, "table_size"),
+		"The size of the table components from information_schema.tables",
+		[]string{"schema", "table", "component"}, nil,
+	)
 	globalPerformanceSchemaLostDesc = prometheus.NewDesc(
 		prometheus.BuildFQName(namespace, globalStatus, "performance_schema_lost_total"),
 		"Total number of MySQL instrumentations that could not be loaded or created due to memory constraints.",
@@ -676,6 +721,12 @@ func (e *Exporter) scrape(ch chan<- prometheus.Metric) {
 			return
 		}
 	}
+	if *collectTableSchema {
+		if err = scrapeTableSchema(db, ch); err != nil {
+			log.Println("Error scraping table schema:", err)
+			return
+		}
+	}
 	if *collectAutoIncrementColumns {
 		if err = scrapeInformationSchema(db, ch); err != nil {
 			log.Println("Error scraping information schema:", err)
@@ -1457,6 +1508,93 @@ func scrapeProcesslist(db *sql.DB, ch chan<- prometheus.Metric) error {
 	return nil
 }
 
+func scrapeTableSchema(db *sql.DB, ch chan<- prometheus.Metric) error {
+	var dbList []string
+	if *tableSchemaDatabases == "*" {
+		dbListRows, err := db.Query(dbListQuery)
+		if err != nil {
+			return err
+		}
+		defer dbListRows.Close()
+
+		var database string
+
+		for dbListRows.Next() {
+			if err := dbListRows.Scan(
+				&database,
+			); err != nil {
+				return err
+			}
+			dbList = append(dbList, database)
+		}
+	} else {
+		dbList = strings.Split(*tableSchemaDatabases, ",")
+	}
+
+	for _, database := range dbList {
+		tableSchemaRows, err := db.Query(fmt.Sprintf(tableSchemaQuery, database))
+		if err != nil {
+			return err
+		}
+		defer tableSchemaRows.Close()
+
+		var (
+			tableSchema   string
+			tableName     string
+			tableType     string
+			engine        string
+			version       uint64
+			rowFormat     string
+			tableRows     uint64
+			dataLength    uint64
+			indexLength   uint64
+			dataFree      uint64
+			createOptions string
+		)
+
+		for tableSchemaRows.Next() {
+			err = tableSchemaRows.Scan(
+				&tableSchema,
+				&tableName,
+				&tableType,
+				&engine,
+				&version,
+				&rowFormat,
+				&tableRows,
+				&dataLength,
+				&indexLength,
+				&dataFree,
+				&createOptions,
+			)
+			if err != nil {
+				return err
+			}
+			ch <- prometheus.MustNewConstMetric(
+				infoSchemaTablesVersionDesc, prometheus.GaugeValue, float64(version),
+				tableSchema, tableName, tableType, engine, rowFormat, createOptions,
+			)
+			ch <- prometheus.MustNewConstMetric(
+				infoSchemaTablesRowsDesc, prometheus.GaugeValue, float64(tableRows),
+				tableSchema, tableName,
+			)
+			ch <- prometheus.MustNewConstMetric(
+				infoSchemaTablesSizeDesc, prometheus.GaugeValue, float64(dataLength),
+				tableSchema, tableName, "data_length",
+			)
+			ch <- prometheus.MustNewConstMetric(
+				infoSchemaTablesSizeDesc, prometheus.GaugeValue, float64(indexLength),
+				tableSchema, tableName, "index_length",
+			)
+			ch <- prometheus.MustNewConstMetric(
+				infoSchemaTablesSizeDesc, prometheus.GaugeValue, float64(dataFree),
+				tableSchema, tableName, "data_free",
+			)
+		}
+	}
+
+	return nil
+}
+
 func newDesc(subsystem, name, help string) *prometheus.Desc {
 	return prometheus.NewDesc(
 		prometheus.BuildFQName(namespace, subsystem, name),