Spider Server System Variables
You are viewing an old version of this article. View
the current version here.
The following variables are available when the Spider storage engine has been installed.
See Server System Variables for a complete list of system variables and instructions on setting them.
Contents
- spider_auto_increment_mode
- spider_bgs_first_read
- spider_bgs_mode
- spider_bgs_second_read
- spider_bka_engine
- spider_bka_mode
- spider_block_size
- spider_bulk_access_free
- spider_bulk_size
- spider_bulk_update_mode
- spider_bulk_update_size
- spider_conn_recycle_mode
- spider_conn_recycle_strict
- spider_connect_mutex
- spider_connect_retry_count
- spider_connect_retry_interval
- spider_connect_timeout
- spider_crd_bg_mode
- spider_crd_interval
- spider_crd_mode
- spider_crd_sync
- spider_crd_type
- spider_crd_weight
- spider_direct_dup_insert
- spider_direct_order_limit
- spider_error_read_mode
- spider_error_write_mode
- spider_first_read
- spider_force_commit
- spider_general_log
- spider_init_sql_alloc_size
- spider_internal_limit
- spider_internal_offset
- spider_internal_optimize
- spider_internal_optimize_local
- spider_internal_sql_log_off
- spider_internal_unlock
- spider_internal_xa
- spider_internal_xa_snapshot
- spider_local_lock_table
- spider_lock_exchange
- spider_log_result_errors
- spider_low_mem_read
- spider_max_order
- spider_multi_split_read
- spider_net_read_timeout
- spider_net_write_timeout
- spider_ping_interval_at_trx_start
- spider_quick_mode
- spider_quick_page_size
- spider_read_only_mode
- spider_remote_access_charset
- spider_remote_autocommit
- spider_remote_default_database
- spider_remote_sql_log_off
- spider_remote_time_zone
- spider_remote_trx_isolation
- spider_reset_sql_alloc
- spider_same_server_link
- spider_second_read
- spider_select_column_mode
- spider_selupd_lock_mode
- spider_semi_split_read
- spider_semi_split_read_limit
- spider_semi_table_lock
- spider_semi_table_lock_connection
- spider_semi_trx
- spider_semi_trx_isolation
- spider_skip_default_condition
- spider_split_read
- spider_sts_bg_mode
- spider_sts_interval
- spider_sts_mode
- spider_sts_sync
- spider_support_xa
- spider_sync_autocommit
- spider_sync_time_zone
- spider_sync_trx_isolation
- spider_table_init_error_interval
- spider_udf_ct_bulk_insert_interval
- spider_udf_ct_bulk_insert_rows
- spider_udf_ds_bulk_insert_rows
- spider_udf_ds_table_loop_mode
- spider_udf_ds_use_real_table
- spider_udf_table_lock_mutex_count
- spider_udf_table_mon_mutex_count
- spider_use_all_conns_snapshot
- spider_use_consistent_snapshot
- spider_use_default_database
- spider_use_flash_logs
- spider_use_handler
- spider_use_pushdown_udf
- spider_use_snapshot_with_flush_tables
- spider_use_table_charset
spider_auto_increment_mode
- Description: The auto increment mode.
-1
The table parameter is adopted.0
Normal mode. Uses a count that Spider gets from the remote server with an exclusive lock for the auto increment value. Slow. Switch to quick mode if you use Spider tables with the table partitioning feature and the auto increment column is first column of index.1
Quick mode. Uses an internal Spider count for the auto increment value. Fast, but it is possible for duplicates to occur if you update the same table from multiple Spider and self tables.2
Set zero value. Auto increment value is given by the remote server. Sets zero value even if you set a value to the auto increment column. Sets zero value after choosing an inserted partition if you use a table with the table partitioning feature.3
The auto increment value is given by the remote server, if you set an auto increment column to a null value. The auto increment value is given by the local server, if you set a zero value to the auto increment column. Rather usespider_reset_auto_increment=2 or 3
if you want to use an auto increment column on the remote server.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
-1
- Default Table Value:
0
spider_bgs_first_read
- Description: Number of first read records when background search is used. To start a range scan on the remote server the engine first need to send the first record . Fetching a second record can save a network round trip stopping the plan if the shard get a single record on the remote data source. First read and second read are used for warming up for background search, third read without using spider_split_read and spider_semi_split_read will fetch the remaining data source in a single last fetch.
-1
The table parameter is adopted.0
Records are usually retrieved.1 or more
Number of records.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
-1
- Default Table Value:
2
spider_bgs_mode
- Description: Background search mode, default partitioning plugin scan partitions one after the other to optimize memory usage. In case of remote shard read can be done in parallel when the plan prune multiple partitions.
-1
The table parameter is adopted.0
Background search is disabled.1
Background search is used if search with no lock.2
Background search is used if search with no lock or shared lock.3
Background search is used regardless of the lock.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
-1
- Default Table Value:
0
spider_bgs_second_read
- Description: Number of second read records when background search is used. When the first records are found from
spider_bgs_first_read
, the engine will scan using limit spider_bgs_first_read, spider_bgs_second_read on the remote data source.-1
The table parameter is adopted.0
Records are usually retrieved.1 or more
Number of records.
- Scope: Global, Session
- Dynamic: Yes
- Default Session Value:
-1
- Default Table Value:
100
spider_bka_engine
- Description: The temporary table engine when the
bka_mode
table parameter is1
. The table parameter default is memory. - Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
-1
- Default Table Value:
Memory
spider_bka_mode
- Description: The internal action for using
multi_split_read=0
.-1
The table parameter is adopted.0
Use "union all".1
Use temporary table if it is judged acceptable.2
Same as1
. Avoiding replication delay.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
-1
- Default Table Value:
1
spider_block_size
- Description: Pseudosize of block used in mysql. Can usually be left unchanged.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
16384
spider_bulk_access_free
- Description: Free mode of bulk access resources.
-1
The table parameter is adopted.0
in reset1
in close
- Data Type:
numeric
- Default Session Value:
-1
- Introduced: Spider 3.0
spider_bulk_size
- Description: Size in bytes of insert query generated at bulk insert.
-1
The table parameter is adopted.0 or more
: Size of buffer.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
-1
- Default Table Value:
16000
spider_bulk_update_mode
- Description: Bulk update and delete mode. If
spider_bgs_mode
orspider_split_read
are changed from their default settings, Spider treats this parameter setting as2
.-1
The table parameter is adopted.0
Send "update" and "delete" statements one by one.1
Send collected multiple "update" and "delete" statements.Collected statements are sent one by one.2
Send collected multiple "update" and "delete" statements. Collected statements are sent together.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
-1
- Default Table Value:
0
spider_bulk_update_size
- Description: Size in bytes of the "update" and "delete" query generated at bulk update.
-1
The table parameter is adopted.0 or more
Size of buffer.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
-1
- Default Table Value:
16000
spider_conn_recycle_mode
- Description: Connection recycle mode.
0
Disconnect.1
Recycles by all sessions.2
Recycles in the same session.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
0
spider_conn_recycle_strict
spider_connect_mutex
- Description: Serialized remote servers connections.
You can use this parameter if you gets an error or slowdown by too many processes of connecting.
0
Don't serialized.1
: Serialized.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
0
spider_connect_retry_count
- Description: Connect-retry-count for connecting failure caused by too many processes of connecting.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
1000
spider_connect_retry_interval
- Description: Connect-retry-interval for connecting failure caused by too many processes of connecting.(micro second)
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
1000
spider_connect_timeout
spider_crd_bg_mode
spider_crd_interval
spider_crd_mode
spider_crd_sync
spider_crd_type
spider_crd_weight
spider_direct_dup_insert
spider_direct_order_limit
spider_error_read_mode
spider_error_write_mode
spider_first_read
spider_force_commit
spider_general_log
spider_init_sql_alloc_size
spider_internal_limit
spider_internal_offset
spider_internal_optimize
spider_internal_optimize_local
spider_internal_sql_log_off
spider_internal_unlock
spider_internal_xa
spider_internal_xa_snapshot
spider_local_lock_table
spider_lock_exchange
spider_log_result_errors
spider_low_mem_read
spider_max_order
spider_multi_split_read
spider_net_read_timeout
spider_net_write_timeout
spider_ping_interval_at_trx_start
spider_quick_mode
spider_quick_page_size
spider_read_only_mode
spider_remote_access_charset
spider_remote_autocommit
spider_remote_default_database
spider_remote_sql_log_off
spider_remote_time_zone
spider_remote_trx_isolation
spider_reset_sql_alloc
spider_same_server_link
spider_second_read
spider_select_column_mode
spider_selupd_lock_mode
spider_semi_split_read
spider_semi_split_read_limit
spider_semi_table_lock
spider_semi_table_lock_connection
spider_semi_trx
spider_semi_trx_isolation
spider_skip_default_condition
spider_split_read
spider_sts_bg_mode
- Description: Mode of table statistic at background.
Background confirmation is enabled by default at the table level, one thread per partition is created to maintain table status. Disable when sts_interval=0, effective as default: sts_interval is 10.
-1
The table parameter is adopted.0
Background confirmation is disabled.1
Background confirmation is enabled.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
-1
- Default Table Value:
1
spider_sts_interval
- Description: Time interval of table statistic.
Set to 0 to always get the newest information from remote servers.
-1
The table parameter is adopted.1 or more
Interval in second table state confirmation.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
-1
- Default Table Value:
10
spider_sts_mode
- Description: Mode for table statistic
The show command is used at the table level default.
-1,0
The table parameter is adopted.1
The show command is used.2
The information schema is used.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
-1
- Default Table Value:
1
spider_sts_sync
- Description: Table statistic synchronization in partitioned table.
-1
The table parameter is adopted.0
No synchronization.1
Table state is synchronized when opening a table, no more synchronization later.2
Synchronization.
- Scope: Global, Session
- Dynamic: Yes
- Data Type:
numeric
- Default Session Value:
-1
- Default Table Value:
0
spider_support_xa
spider_sync_autocommit
spider_sync_time_zone
spider_sync_trx_isolation
spider_table_init_error_interval
spider_udf_ct_bulk_insert_interval
spider_udf_ct_bulk_insert_rows
spider_udf_ds_bulk_insert_rows
spider_udf_ds_table_loop_mode
spider_udf_ds_use_real_table
spider_udf_table_lock_mutex_count
spider_udf_table_mon_mutex_count
spider_use_all_conns_snapshot
spider_use_consistent_snapshot
spider_use_default_database
spider_use_flash_logs
spider_use_handler
spider_use_pushdown_udf
spider_use_snapshot_with_flush_tables
spider_use_table_charset
Comments
Comments loading...
Content reproduced on this site is the property of its respective owners,
and this content is not reviewed in advance by MariaDB. The views, information and opinions
expressed by this content do not necessarily represent those of MariaDB or any other party.