MySql Exporter
In a MySQL Exporter dashboard in Grafana, you typically have multiple panels, each displaying different metrics or information related to the MySQL database being monitored. Here’s an explanation of some common panels you might find in a MySQL Exporter dashboard:
mysql_global_status_uptime: This metric shows the total time, in seconds, that the MySQL server has been running since it was last started.
mysql_global_status_queries: This metric counts the total number of SQL statements executed by the MySQL server.
mysql_global_status_bytes_received: This metric indicates the total number of bytes received by the MySQL server from all clients.
mysql_global_status_bytes_sent: This metric indicates the total number of bytes sent by the MySQL server to all clients.
mysql_global_status_max_used_connections: This metric records the highest number of connections used simultaneously since the MySQL server.
mysql_global_variables_max_connections: This variable defines the maximum number of concurrent connections allowed to the MySQL server. It sets the upper limit for the max_used_connections metric and helps in managing server resources and preventing overload.
mysql_global_status_threads_connected: This metric shows the current number of client connections that are currently open and connected to the MySQL server. It reflects the active connections handling queries or waiting for results.
mysql_global_status_threads_running: This metric indicates the current number of threads that are actively executing queries in the MySQL server. It provides insights into the level of concurrent query execution and server workload.
mysql_global_status_threads_cached: This metric indicates the current number of threads stored in the thread cache.
mysql_global_status_threads_created: This metric counts the total number of threads created by the MySQL server.
mysql_global_status_created_tmp_tables: This metric counts the number of temporary tables created in memory by MySQL since it was last started or reset. Temporary tables are used for intermediate results during query execution.
mysql_global_status_created_tmp_disk_tables: This metric counts the number of temporary tables that could not be created in memory and were instead created on disk by MySQL.
mysql_global_status_created_tmp_files: This metric counts the total number of temporary files created by MySQL.
mysql_global_status_questions: This metric counts the number of SQL statements (queries) executed by the MySQL server.
mysql_global_status_select_full_join: This metric counts the number of times MySQL has executed a full join during SELECT queries.
mysql_global_status_select_full_range_join: This metric counts the number of times MySQL has executed a full range join during SELECT queries.
mysql_global_status_select_range: This metric counts the number of times MySQL has used range access during SELECT queries.
mysql_global_status_sort_rows: This metric counts the number of rows sorted by MySQL using sorting operations such as ORDER BY or GROUP BY.
mysql_global_status_sort_range: This metric counts the number of times MySQL has used range sorting during query execution.
mysql_global_status_sort_merge_passes: This metric counts the number of merge passes that MySQL has performed.
mysql_global_status_slow_queries: This metric counts the number of queries that have taken longer than the threshold defined by long_query_time to execute.
mysql_global_status_aborted_connects: This metric counts the number of failed attempts to connect to the MySQL server
mysql_global_status_aborted_clients: This metric counts the number of client connections that were aborted or closed unexpectedly by clients
mysql_global_status_table_locks_immediate: This metric counts the number of times MySQL was able to acquire a table lock immediately without having to wait.
mysql_global_status_table_locks_waited: This metric counts the number of times MySQL had to wait to acquire a table lock.
mysql_global_variables_innodb_buffer_pool_size: This metric indicates the size of the InnoDB buffer pool in bytes. The buffer pool is a key component of MySQL’s InnoDB storage engine, used to cache data and indexes in memory to improve database performance.
mysql_global_status_innodb_page_size: This metric indicates the size of InnoDB pages in bytes. The default page size is typically 16KB.
mysql_global_status_buffer_pool_pages: This metric shows the total number of pages managed by the buffer pool in InnoDB.
mysql_global_variables_innodb_log_buffer_size: This variable defines the size of the buffer that InnoDB uses to store log data before writing it to the log files.
mysql_global_variables_key_buffer_size: This variable specifies the size of the key buffer, also known as the key cache.
mysql_global_variables_query_cache_size: This variable defines the total amount of memory allocated for the query cache.
mysql_global_variables_thread_cache_size: This variable specifies the size of the thread cache in MySQL, which stores threads that can be reused for client connections.
mysql_global_status_commands_total: This metric counts the total number of SQL commands executed by the MySQL server since it was last started or reset. It includes all types of commands such as queries (SELECT, INSERT, UPDATE, DELETE), administrative commands (GRANT, FLUSH), and others.
mysql_global_status_handlers_total: This metric counts the total number of handler operations executed by the MySQL server since it was last started or reset. Handler operations include operations such as read, write, update, and delete operations on tables or indexes.
mysql_info_schema_threads: The metric mysql_info_schema_threads typically refers to a table in the MySQL performance_schema database named threads. This table provides information about threads that are currently running within the MySQL server instance. It includes details such as thread ID, thread status, thread type, and other relevant attributes.
mysql_global_status_opened_files: This metric tracks the total number of files opened by the MySQL server. It includes files opened for various operations such as table data, indexes, logs, temporary files, and others.
mysql_global_status_open_files: This metric counts the number of files currently open in MySQL, showing how many resources are in use. It helps monitor MySQL’s file usage, ensuring it stays within system limits.
mysql_global_status_opened_tables: This metric counts the total number of tables that have been opened by MySQL since it was last started or reset.
mysql_global_status_table_open_cache_hits: This metric counts the number of times MySQL successfully retrieves a table from the table open cache without needing to reopen it from disk.
mysql_global_status_table_open_cache_misses: This metric counts the number of times MySQL fails to find a table in the table open cache and has to reopen it from disk.
mysql_global_status_table_open_cache_overflows: This metric counts the number of times MySQL attempts to add a table to the table open cache but the cache is full, leading to eviction of older entries.
mysql_global_status_open_tables: This metric indicates the current number of tables that are currently open and in use by MySQL.
mysql_global_variables_table_open_cache: This variable specifies the size of the table open cache in MySQL. The table open cache stores information about tables that have been accessed, reducing the need to reopen tables from disk.
mysql_global_status_open_table_definitions: This metric indicates the current number of table definitions that are currently open and in use by MySQL.
mysql_global_variables_table_definition_cache: This variable specifies the size of the table definition cache in MySQL. The table definition cache stores metadata about table structures, reducing the need to reload table definitions from disk.
node_vmstat_pgpgin: This metric indicates the total number of memory pages that have been brought into the system from disk per second. It reflects the rate at which data is read from disk into memory.
node_vmstat_pgpgout: This metric indicates the total number of memory pages that have been written out to disk per second. It reflects the rate at which data is written from memory to disk.
node_memory_MemTotal_bytes: Total amount of physical memory (RAM) available on the node in bytes. This represents the entire memory capacity of the system.
node_memory_MemFree_bytes: Amount of free (unused) physical memory available on the node in bytes. It shows how much memory is currently not in use by applications or the operating system.
node_memory_Buffers: Amount of memory used for buffers by the operating system in kilobytes. Buffers are used to temporarily hold data being transferred between devices or between processes.
node_memory_Cached: Amount of memory used for caching of data by the operating system in kilobytes. Cached memory includes data that has been read from disk and stored in memory for faster access.
node_cpu_seconds_total: This metric tracks the total amount of CPU time consumed by the node since it started. It includes user-mode, system-mode, and idle time across all CPUs or cores.
node_load1: This metric represents the average system load over the last 1 minute. System load indicates the number of processes waiting to run or running on CPUs. A load of 1 means the system’s CPUs are fully utilized, while values above 1 indicate potential CPU congestion.
node_network_receive_bytes_total: This metric represents the total number of bytes received on all network interfaces since the node started. It measures incoming network traffic.
node_network_transmit_bytes_total: This metric represents the total number of bytes transmitted on all network interfaces since the node started. It measures outgoing network traffic.
node_vmstat_pswpin: This metric indicates the total number of swap pages brought into memory per second. It represents the rate at which data is read from swap space (disk) into memory to fulfill memory demands.
node_vmstat_pswpout: This metric indicates the total number of swap pages swapped out per second. It reflects the rate at which data is written from memory to swap space (disk) due to memory pressure.
MySQL Uptime Panel
The amount of time since the last restart of the MySQL server process.
Current QPS Panel
The Current QPS (Queries Per Second) panel shows how many queries the MySQL server is handling each second.
InnoDB Buffer Pool Size Panel
InnoDB maintains a storage area called the buffer pool for caching data and indexes in memory.The InnoDB Buffer Pool Size panel in the MySQL dashboard indicates the amount of memory allocated to the InnoDB buffer pool. This buffer pool is used to cache data and indexes, which helps improve database performance by reducing disk I/O operations.
Buffer Pool Size of Total RAM Panel
The “Buffer Pool Size of Total RAM” panel in the MySQL dashboard shows how much of your computer’s memory is used for caching data and indexes in InnoDB. This helps speed up the database by reducing the need to read from the disk frequently.
MySQL Connections Panel
The “MySQL Connections” panel in the dashboard gives details on the current and peak usage of database connections. It shows the average number of active connections (Connections), the highest number of connections used (Max Used Connections), and the maximum allowed connections set in the configuration (Max Connections).
MySQL Client Thread Activity Panel
The “MySQL Client Thread Activity” panel in the dashboard provides insights into how MySQL handles client connections and query execution. It includes “Peak Threads Connected,” which tracks the highest number of simultaneous client connections. “Peak Threads Running” shows the maximum threads actively processing queries. The “Avg Threads Running” metric gives an average count of threads executing queries over a period.
MySQL Questions Panel
The “MySQL Questions” panel in the dashboard provides insights into the number of statements executed by the MySQL server. These statements include queries sent directly from clients to the server and exclude statements executed within stored programs such as procedures or triggers.
MySQL Thread Cache Panel
The “MySQL Thread Cache” panel in the dashboard provides insights into how MySQL manages and caches threads for client connections. The “Thread Cache Size” indicates the current configured size of the thread cache, which determines how many threads MySQL can keep cached for reuse instead of creating new ones. “Threads Cached” shows the average number of threads that are currently cached and ready for reuse. “Threads Created” measures the rate at which new threads are being created by MySQL to handle incoming client connections.
MySQL Temporary Objects Panel
The “MySQL Temporary Objects” panel in the dashboard shows how MySQL handles temporary data during queries. “Created Tmp Tables” measures how often MySQL makes temporary tables in memory for tasks like sorting or combining large datasets. If memory limits are exceeded, “Created Tmp Disk Tables” shows how often these temporary tables are stored on disk instead, which can slow down performance due to disk operations. “Created Tmp Files” tracks how often MySQL creates temporary files on disk for tasks needing more storage than available memory.
MySQL Select Types Panel
The “MySQL Select Types” panel in the dashboard shows how MySQL handles different types of SELECT queries. “Select Full Join” tells us how often MySQL combines data from two tables with matching columns using full joins. “Select Range” shows how often queries fetch data within specific ranges using conditions like BETWEEN in WHERE clauses. “Select Scan” indicates how often MySQL scans entire tables or large parts of them to find data, usually without using indexes.
MySQL Sorts Panel
The “MySQL Sorts” panel in the dashboard shows how often MySQL arranges query results. “Sort Rows” counts how often MySQL organizes rows in queries by criteria like alphabetical order. Sorting is essential for presenting data neatly. “Sort Scan” counts how often MySQL looks through sorted data to find what a query needs. It’s about finding and using sorted data effectively.
MySQL Slow Queries Panel
The “MySQL Slow Queries” panel in the dashboard tracks the frequency of queries that take longer than a predefined threshold to execute. It measures “Slow Queries,” showing how often MySQL processes queries that exceed the expected performance time.
MySQL Aborted Connections Panel
The “MySQL Aborted Connections” panel in the dashboard tracks cases where connections to the MySQL server were cut off prematurely. It shows “Aborted Connects (attempts),” which counts failed connection tries that didn’t complete.Also, “Aborted Clients (timeout)” measures how often client connections are ended due to timeouts, when clients take too long to finish their tasks.
MySQL Table Locks Panel
The “MySQL Table Locks” panel in the dashboard provides information on how table locks are managed during database operations. It includes “Table Locks Immediate,” which tracks the average number of times table locks are acquired immediately without waiting.“Table Locks Waited” measures the instances where table locks were not acquired immediately and had to wait
MySQL Network Traffic Panel
The “MySQL Network Traffic” panel in the dashboard shows the amount of data being transferred to and from the MySQL server. “Inbound” measures the average rate of incoming data, indicating how much data the server receives from clients. “Outbound” tracks the average rate of outgoing data, showing how much data the server sends to clients.
MySQL Network Usage Hourly Panel
The “MySQL Network Usage Hourly” panel in the dashboard shows data transfer over the last 24 hours. “Received” tracks the average amount of data the server received each hour, while “Sent” shows the average amount of data sent out each hour.
MySQL Internal Memory Overview Panel
The “MySQL Internal Memory Overview” panel in the dashboard provides insights into the allocation and usage of memory within the MySQL server. “System Memory” indicates the total memory available on the server. “InnoDB Buffer Pool Data” shows the amount of memory used by the InnoDB storage engine for caching data. “InnoDB Log Buffer Size” displays the memory allocated for storing log data before it’s written to disk. “Key Buffer Size” indicates the memory allocated for caching index blocks.
Top Command Counters Panel
The “Top Command Counters” panel in the MySQL dashboard tracks the frequency of different types of SQL commands executed on the server. This includes commands like “Com_admin_commands” for administrative tasks, “Com_commit” for transaction commits, and “Com_select” for data retrieval queries. It also tracks “Com_set_option” for setting session variables, “Com_show_binlogs” for viewing binary logs, “Com_show_replica_status” and “Com_show_slave_status” for replication status, “Com_show_status” for viewing server status, “Com_show_variables” for displaying server variables, and “Com_update” for updating data.
Top Command Counters Hourly Panel
The “Top Command Counters Hourly” panel in the MySQL dashboard provides an overview of the frequency of various SQL commands executed on the server over the last 24 hours.This includes commands as same as Top Command Counters.
MySQL Handlers Panel
The “MySQL Handlers” panel in the dashboard provides insights into how MySQL interacts with different types of data access operations, categorized by handler functions.“external_lock” measures the rate of external locking operations, crucial for managing concurrent access to database resources. “read_first” tracks initial record reads from an index. “read_key” counts operations where MySQL retrieves rows based on index lookups. “read_next” indicates sequential reads of rows following an initial index read. “read_rnd_next” measures random access reads after an initial index read. “update” monitors updates to rows in tables. “write” reflects the frequency of write operations to the database.
MySQL Transaction Handlers Panel
The “MySQL Transaction Handlers” panel in the dashboard gives insights into how MySQL manages transactions, specifically focusing on commit and prepare operations. These show how often these actions occur on average during a given time frame. “Commit” tracks how frequently transactions are completed and changes are permanently stored in the database, ensuring data integrity and durability. “Prepare” measures how often transactions get ready for execution, which is crucial for ensuring consistency in distributed databases.
Process States Panel
The “Process States” panel in the MySQL dashboard provides an overview of the current states of MySQL server processes. It includes categories such as “idle” and “other,” which summarize how many processes are in each state at a given time. “Idle” indicates processes that are not actively executing tasks and are waiting for new instructions. “Other” encompasses various states that processes might enter during different phases of their lifecycle, such as connecting to clients, performing administrative tasks, or waiting for resources.
Top Process States Hourly Panel
The “Top Process States Hourly” panel in the MySQL dashboard shows how MySQL processes are active. “Executing” means processes are actively running queries. “Idle” means processes are waiting for new tasks or client requests. “Other” covers various states like administrative tasks or managing connections.
MySQL File Openings Panel
The “MySQL File Openings” panel in the dashboard tracks the frequency of file openings performed by MySQL over a given period. This metric indicates how often MySQL accesses files, including data files, log files, and configuration files.
MySQL Open Files Panel
The “MySQL Open Files” panel shows how many files MySQL has open and its limits. “Open Files” is how many files MySQL has open on average, including data and log files. “Open Files Limit” is the max number of files MySQL can have open at once. “InnoDB Open Files” is how many files the InnoDB storage engine has open on average.
MySQL Table Open Cache Status Panel
The “MySQL Table Open Cache Status” panel shows how MySQL uses a cache to store information about open tables for faster query performance. Openings tracks how often tables are opened in the cache. Hits measures how often MySQL finds table information in the cache. Misses counts when MySQL needs to load table information from disk. Misses due to Overflows shows when the cache can’t hold all requested tables. Table Open Cache Hit Ratio indicates how well the cache reduces disk accesses.
MySQL Open Tables Panel
The “MySQL Open Tables” panel in the dashboard provides insights into the current status of open tables managed by MySQL. “Open Tables” indicates the average number of tables currently open and in use by MySQL processes. This includes tables accessed for queries and other operations. “Table Open Cache” specifies the configured size of the table open cache, which is used to store table descriptors to avoid reopening tables repeatedly.
MySQL Table Definition Cache Panel
The “MySQL Table Definition Cache” panel in the dashboard provides insights into how MySQL manages table definitions for optimizing query performance. “Open Table Definitions” indicates the average number of table definitions currently open and in use by MySQL. These definitions include metadata and schema information necessary for query execution. “Table Definitions Cache Size” specifies the configured size of the cache used to store these definitions, ensuring quick access without needing to fetch them repeatedly from disk.
I/O Activity Panel
The “I/O Activity” panel in the MySQL dashboard provides crucial insights into the input/output operations handled by the database server. It includes “Page In,” which measures the average rate of data read from storage into memory. “Page Out” indicates the average rate of data written from memory back to storage, often to manage memory usage efficiently by moving less frequently accessed data out of memory.
Memory Distribution Panel
The “Memory Distribution” panel in the MySQL dashboard provides an overview of how system memory is allocated and utilized. It includes “Free,” which indicates the average amount of unused memory available on the system. This metric is crucial for monitoring system health, as sufficient free memory ensures that MySQL and other applications can operate without resource constraints.
CPU Usage / Load Panel
The “CPU Usage / Load” panel in MySQL’s dashboard shows how the CPU is used by the system and MySQL itself. It includes I/Owait that tracks time spent waiting for disk operations and revealing potential disk performance issues. Nice tracks CPU time for lower-priority tasks, affecting how less critical jobs are managed, softirq time handling software interrupts, crucial for network and I/O tasks. Steal tracks CPU time taken by the hypervisor in virtualized environments. System tracks CPU time for kernel tasks like scheduling and managing I/O. User tracks CPU time for applications and processes, including MySQL. Load 1m Average system load over the last minute, indicating how many processes are waiting to use the CPU.
Network Traffic Panel
The “Network Traffic” panel in the MySQL dashboard tracks data transfer rates for inbound and outbound traffic. Inbound measures how much data MySQL receives from external sources or clients on average, while outbound tracks the average amount of data sent from MySQL to external destinations or clients.
Swap Activity Panel
The “Swap Activity” panel in MySQL dashboard monitors how often data is read from (Swap In) and written to (Swap Out) the swap space. Swap In shows how frequently MySQL retrieves data from swap to active memory, indicating possible memory constraints or management issues. Swap Out tracks how often data is moved from memory to swap, freeing up RAM for critical tasks. High Swap Out rates suggest heavy memory usage, suggesting a need for better memory management or additional RAM for improved performance.
MysqlDown
mysql_up < 3MysqlTooManyConnections(>80%)
max_over_time(mysql_global_status_threads_connected[1m]) / mysql_global_variables_max_connections * 100 > 80MysqlHighThreadsRunning
max_over_time(mysql_global_status_threads_running[1m]) / mysql_global_variables_max_connections * 100 > 60MysqlSlaveIoThreadNotRunning
( mysql_slave_status_slave_io_running and ON (instance) mysql_slave_status_master_server_id > 0 ) == 0MysqlSlaveSqlThreadNotRunning
( mysql_slave_status_slave_sql_running and ON (instance) mysql_slave_status_master_server_id > 0) == 0MysqlSlaveReplicationLag
( (mysql_slave_status_seconds_behind_master - mysql_slave_status_sql_delay) and ON (instance) mysql_slave_status_master_server_id > 0 ) > 30MysqlSlowQueries
increase(mysql_global_status_slow_queries[1m]) > 0MysqlInnodbLogWaits
rate(mysql_global_status_innodb_log_waits[15m]) > 10MysqlRestarted
mysql_global_status_uptime < 60