SHOW STATUS EXTENDED

Aborted_clients

0

The number of aborted connections caused by the client dying without closing the connection properly.

Aborted_connects

0

The number of failed attempts to connect to the server.

Alloc_large_variable

Tracks memory allocated for variable length columns inside rowstore tables, or for other variable length memory allocations inside query execution larger than 128KB. Can take a lot of RAM.

Bytes_received

202

The total number of bytes received from all clients.

Bytes_sent

2556

The number of bytes sent to clients.

Connections

11

The number of successful or non-successful connection attempts to the server.

Max_used_connections

2

The maximum number of simultaneous connections the server has seen since it started.

Queries

78

The number of statements the server has executed, including statements from clients and statements within stored programs.

Questions

6

The number of statements the server has executed, excluding statements within stored programs.

Threads_cached

2

Number of kernel threads currently idle waiting for queries.

Threads_connected

2

The number of connections currently open.

Threads_created

10

The number of threads created to handle connections.

Threads_running

1

The number of threads that are currently active.

Uptime

11381

The amount of time, in seconds, that the server has been up.

aggregator_id

A unique identifier for each aggregator. This is allocated by the master aggregator the first time a child aggregator connects.

Seconds_until_expiration

2539106

Seconds until expiration

License_key

The license key. Not present in secure mode.

License_type

Developer or Enterprise depending on the license. Not present in secure mode.

Maximum_cluster_capacity

18446744073709551615 MB

The allowed memory capacity in the cluster based on the license. Only on the master aggregator. Only present if the global variable license_visibility is set to true.

used_cluster_capacity

The amount of memory capacity existent in the cluster. This is the sum of maximum_memory from all nodes in the cluster. This metric is meaningful from the master aggregator as it is the state of the cluster as a whole and not any particular node. Only present if the global variable license_visibility is set to true.

Query_compilations

0

Number of query compilations.

Query_compilation_failures

0

Number of query compilation failures.

GCed_versions_last_sweep

0

Number of versioned nodes reclaimed during the last iteration of the garbage collector.

Average_garbage_collection_duration

0 ms

Average duration of garbage collection, in milliseconds.

Total_server_memory

81.6 MB

Tracks the node’s overall memory use. Will not surpass the maximum_memory engine variable.

Alloc_thread_stacks

17.0 MB

The memory allocated by threads for thread stacks on this node.

Malloc_active_memory

364.1 MB

Tracks memory that is allocated directly from the Linux operating system and managed by the C runtime allocators. This is in contrast to memory managed by SingleStore's built-in memory allocators that use the Buffer Manager. includes, but is not limited to, columnstore tables, open connections, and metadata about tables, columns, etc. For most workloads, the memory use here should be approximately 1-2 GBs

malloc_cumulative_memory

68815.7 (+0.3) MB

The total memory ever allocated by glibc malloc.

Alloc_variable_cached_buffers

0.4 MB

The number of unused 128K memory buffers cached by the variable allocator.

Ssl_accept_renegotiates  

0

Returns the number of start renegotiations in server mode.

Ssl_accepts    

0

Returns the number of started SSL/TLS handshakes in server mode.

Ssl_callback_cache_hits  

0

Returns the number of successfully retrieved sessions from the external session cache in server mode.

Ssl_client_connects  

0

Returns the number of started SSL/TLS handshakes in client mode.

Ssl_connect_renegotiates 

0

Returns the number of start renegotiations in client mode.

Ssl_ctx_verify_depth     

18446744073709551615

Returns the verification depth limit currently set in CTX. If no limit has been set, -1 is returned and the default value is used.

Ssl_ctx_verify_mode      

0

Returns the verification mode currently set in CTX.

Ssl_default_timeout      

0

Returns the default timeout value assigned to SSL_SESSION objects negotiated for the protocol valid for SSL.

Ssl_finished_accepts     

0

Returns the number of successfully established SSL/TLS sessions in server mode.

Ssl_finished_connects    

0

Returns the number of successfully established SSL/TLS sessions in client mode.

Ssl_session_cache_hits   

0

Returns the number of successfully reused sessions.

Ssl_session_cache_misses 

0

Returns the number of sessions proposed by clients that were not found in the internal session cache in server mode.

Ssl_session_cache_overflows

0

Returns the number of sessions that were removed because the maximum session cache size was exceeded.

Ssl_session_cache_size   

20480

Returns the currently valid session cache size.

Ssl_session_cache_timeout

0

Returns the number of sessions proposed by clients and either found in the internal or external session cache in server mode, but that were invalid due to timeout.

Ssl_sessions_reused      

0

Returns 0 or 1 depending on whether a reused session was negotiated during the handshake.

Ssl_used_session_cache_entries

0

Returns the current number of sessions in the internal session cache.

Ssl_verify_depth         

0

Returns the verification depth limit currently set in SSL for this connection. If no limit has been set, -1 is returned and the default value is used.

Ssl_verify_mode          

0

Returns the verification mode currently set in SSL for this connection.

Ssl_cipher               

Returns the name of the cipher being used for the current connection.

Ssl_cipher_list          

Returns a pointer to the name of the SSL_CIPHER listed for SSL with priority.

Ssl_version              

Returns the name of the protocol used for the connection SSL.

Ssl_session_cache_mode   

SERVER

Returns the cache mode currently being used.

Threads_background

5

Background threads managed by the query scheduler, i.e. replication master queries and the scheduler thread.

Threads_idle

0

The number of idle connections not associated with threads.

Ready_queue

0

The number of connections with pending queries not associated with threads.

Idle_queue

0

The number of connections in the idle queue. This is a subset of threads_idle.

Context_switches

57

The number of times any thread has switched connections because it deemed a connection idle.

Context_switch_misses

0

The number of times the scheduler tried to switch connections but failed.

Workload_management_queued_queries

0

The number of queries being delayed due to workload management on this aggregator.

Workload_management_active_queries

0

The number of queries currently running on this aggregator.

Workload_management_active_threads

0

The number of threads currently running per leaf due to queries originating from this aggregator.

Workload_management_active_connections

0

The number of connections per leaf allocated to queries originating from this aggregator.

Columnstore_ingest_management_queued_queries

0

The number of queued concurrent columnstore writes.

Columnstore_ingest_management_active_queries

0

The number of running columnstore writes inserting into the in-memory segment.

Columnstore_ingest_management_max_concurrency

0

The maximum number of queries that can run concurrently. The value 0 indicates that this limit is not set.

Columnstore_ingest_management_estimated_segments_to_flush

0

The number of segments estimated to be created based on the amount of data in the in-memory segment.

Columnstore_ingest_management_estimated_memory

0.000 MB

The estimated amount of memory used by the in-memory segments.

Auto_attach_remaining_seconds

0

The time remaining until the automatic rebalance triggered by an auto-attach operation will occur. Only applies to aggregators.

Data_directory

/var/lib/memsql/data

The directory containing snapshots, transaction logs, and columnstore blob files.

Plancache_directory

/var/lib/memsql/plancache

Directory path for the plancache directory, which contains compiled plans used for codegen.

Transaction_logs_directory

var/lib/memsql/master-3306/data/logs

Always the logs subfolder of datadir. This is where transaction logs are stored.

Segments_directory

var/lib/memsql/master-3306/data/columns

Always the columns subfolder of datadir. This is where columnstore blobs are stored.

Snapshots_directory

var/lib/memsql/master-3306/data/snapshots

Always the snapshots subfolder of datadir. This is where snapshot files live.

threads_waiting_for_disk_space

0

The number of threads currently blocked because the disk is full.

Malloc_transaction_cached_memory

323.3 MB

The memory allocated to caching transaction objects.

Buffer_manager_memory

7.4 (+7.4) MB

Tracks memory that is allocated by the Buffer Manager for SingleStore Helios’s built-in memory allocators.

Buffer_manager_cached_memory

1.0 (+1.0) MB

Tracks memory that was allocated by the Buffer Manager but is now cached and not in use. Limited to 25% of maximum_memory.

Buffer_manager_unrecycled_memory

0.0 MB

Number of buffers ready to be recycled which have not yet been reclaimed.

Alloc_vector_index

Tracks memory used by vector indexes in megabytes.

Tracks memory used by vector indexes in megabytes.

Alloc_skiplist_tower

Tracks the memory used by the towers for skiplist indexes. Each skiplist index utilizes on average 40 bytes of memory per row using this allocator. It's important to note that the exact amount of memory used per row is probabilistic because it depends on the randomized tower height of the particular row being indexed. Skiplist indexes are a part of SingleStore’s indexing mechanism, especially for rowstore tables. These indexes are designed to maintain data in a sorted order to optimize search operations. It also gets used for in-memory rowstore segments for columnstores.

Alloc_variable

Tracks the memory allocated for variable-length columns inside rowstore tables, or for other variable-length memory allocations inside query execution. This can include memory used for operations like sorting, joins, or other in-memory operations that require variable-length data structures. This is relevant especially for workloads that heavily utilize variable-length columns or perform complex query operations necessitating significant variable-length memory allocations. It also gets used for in-memory rowstore segments for columnstores.

Alloc_table_autostats

Memory usage for autostats on columnstore tables.

Alloc_table_primary

Tracks the memory used for on-row data for rowstore tables and also the in-memory segment of columnstore tables. It represents the amount of memory allocated for storing the actual data rows that reside in the primary storage of the system.

Alloc_deleted_version

Memory used to mark rows as deleted in rowstore tables.

Alloc_internal_key_node

Bytes used for HASH index buckets.

Alloc_hash_buckets

Memory used for HASH index buckets.

Alloc_table_metadata_cache

Bytes used for table entries used by DDL.

Alloc_unit_images

This variable measures the bytes allocated to unit images, which contain metadata and bytecode necessary to execute an individual plan. In essence, unit images are a representation of compiled query plans that SingleStore uses to execute SQL queries efficiently.

It also measures memory used for compiled table code modules that are in active use. These code modules contain code used to access rows in the tables.

Alloc_unit_ifn_thunks

Bytes allocated to thunks. These are stub functions assembled by the interpreter to obey the x64 ABI.

Alloc_object_code_images

Bytes attributed to mapping .mo files into memory.

Alloc_compiled_unit_sections

This variable is used to track the memory allocated by the dynamic object loader for code, table code objects and globals specific to each compiled unit or plan. In simpler terms, it represents the memory footprint for sections of compiled query plans or units, which are essential for query execution in SingleStore.

Alloc_databases_list_entry

Bytes used for database entries used by DDL.

Alloc_plan_cache

Bytes allocated for plan objects within the plancache.

Alloc_query_execution

Memory used for running queries on the node. Only present if queries are currently executing.

Alloc_durability_large

Tracks memory used to allocate transaction buffers to group commit rows to disk, as well as temporary buffers to write backups or take snapshots.

alloc_client_connection

Bytes allocated for outgoing connections to other nodes in the cluster.

Alloc_protocol_packet

Bytes allocated for result set protocol objects.

Alloc_table_memory

Tracks the memory stored inside all rowstore tables.

Alloc_variable_bucket_<number>

number = (16, 24, 32, 40, 48,...5680)

The variable allocator is split into buckets with different allocation sizes. Allocations <= 16 bytes use the 16 bytes bucket; allocations between 17 and 24 bytes use the 24 bytes bucket, and so on. Alloc_variable_bucket_16 shows information about the 16 bytes bucket. Each of these is a subset of Alloc_variable.

Alloc_variable_allocated

Memory currently active from the variable length allocator.

Successful_read_queries

0

Number of read queries which have successfully run since server startup.

Successful_write_queries

0

Number of write queries which have successfully run since server startup.

Failed_read_queries

0

Number of read-query executions that have failed since server startup on this node, including queries that were rolled back, aborted, or encountered a runtime error. This value is maintained in memory and reset to zero when the node restarts.

For distributed queries, this count represents failed partition-level tasks rather than failed SQL statements. A single failed statement can therefore increase the count by more than one.

The count includes statements run by stored procedures (internal and user-defined) as well as user queries. Therefore, a non-zero value does not necessarily indicate that an application query failed.

Query the MV_PLANCACHE information schema view to identify the failed queries, typically queries with ROLLBACKS > 0. Refer to Investigate Failed Queries for related information.

Failed_write_queries

0

Number of write-query executions that have failed since server startup on this node, including queries that were rolled back, aborted, or encountered a runtime error. This value is maintained in memory and reset to zero when the node restarts.

For distributed queries, this count represents failed partition-level tasks rather than failed SQL statements. A single failed statement can therefore increase the count by more than one.

The count includes statements run by stored procedures (internal and user-defined) as well as user queries. Therefore, a non-zero value does not necessarily indicate that an application query failed.

Query the MV_PLANCACHE information schema view to identify the failed queries, typically queries with ROLLBACKS > 0. Refer to Investigate Failed Queries for related information.

Rows_returned_by_reads

0

Number of rows returned by read queries since server startup.

Rows_affected_by_writes

0

Number of rows affected by write queries since server startup.

Execution_time_of_reads

0 ms

Total wall-time of all read queries since server startup.

Execution_time_of_write

0 ms

Total wall-time of all writer queries since server startup.

Transaction_buffer_wait_time

0 ms

Provides a view into transaction buffer log performance. For more information, see the Using Durability and Recovery topic.

Transaction_log_flush_wait_time

0 ms

Cumulative amount of time commits have been waiting to flush to disk or across the network, if the database is running with sync durability or sync replication. For async, the value of Transaction_log_flush_wait_time should always be 0.

Row_lock_wait_time

0 ms

Cumulative amount of time spent waiting on row locks by any write query.

Ingest_errors_disk_space_use

59580 Bytes

Tracks the total disk space (in bytes) currently used by stored ingest errors from LOAD DATA and PIPELINES.

Free_io_pool_memory

Max 16 MB

Size of cached free buffers, which tracks how much memory in the IO pool is currently not in use. The maximum limit applies to each pool per partition on a node and can shrink if memory is reused. This variable can only be used to view the unused IO pool memory.

Total_io_pool_memory

Max 16 MB

Total size of all buffers allocated by the IO pool for that partition, including buffers currently in use and free buffers in Free_io_pool_memory. The maximum limit applies to each pool per partition on a node and does not shrink from its high value. This variable can only be used to view the total IO pool size.

Gv_clock

Represents the current timestamp of the server's logical clock.

Effective_module_cache_max_size_mb

Memory allocated (in MBs) used to cache query object code images, as derived from compiled_images_eviction_memory_limit_percent and compiled_images_eviction_memory_limit_mb. Refer to Managing Plancache Memory and Disk Usage for more information.

Alloc_table_memory

Alloc_skiplist_tower

Alloc_variable

Alloc_large_variable

Alloc_table_primary

Alloc_deleted_version

Alloc_internal_key_node

Alloc_hash_buckets

Alloc_table_autostats

Total_server_memory

Alloc_thread_stacks

Malloc_active_memory

Buffer_manager_memory

Total_io_pool_memory

Alloc_compiled_unit_sections

Alloc_object_code_images

Alloc_unit_ifn_thunks

Alloc_unit_images

Alloc_mmap_memory

Alloc_table_memory

Alloc_vector_index

Buffer_manager_memory

All allocators except Malloc_active_memory